Regulatory Open Forum

Dear RAPS Community,

What is CE design certificate and it is mandatory to obtain from NB for class IIa device (SaMD)?

Thank you for any info!

Dear RAPS Community,

What is CE design certificate and it is mandatory to obtain from NB for class IIa device (SaMD)?

Thank you for any info!

Hi Anon,

as Kevin says, the MDR does not use the expression "design certificate".

Annex XII that deals with the certificates issued by a notified body lists: "EU technical documentation assessment certificates, EU type-examination certificates and EU product verification certificates" in section 4(a) and "EU quality management system certificates and EU quality assurance certificates" in section 4(b). You could argue that the former are to some degree "device certificates", because they refer to an individual technical documentation, device type or even single devices, while the latter are "quality management certificates", but there is no such formulation in the MDR.

For a class IIa SaMD you usually would need either a "EU quality management system certificate" (Annex IX, chapters I and III) or a "EU quality assurence certificate" (Annex XI, Part A, Section 10). Both procedures would include the assessment of a technical documentation for at least one representative device sample for each device category, but as far as I know this assessment does not result in a separate certificate (as opposed to e.g. the full Annex IX assessment for class III devices or a type examination according to Annex X). If you only have one device, then of course the sampling in the end means that every device is assessed this way, but again this to my knowledge is not reflected in its own certificate. In theory Annex XI, Part B, Section 18, would be a possible procedure, but testing each device does not make sense for software and I do not think any notified body has this in their scope.

So in a loose sense, you do not obtain a design certificate for a class IIa software, but a quality management certificate (either full quality management system or product assurance). However, using the language of the MDR in describing the certificate is the better idea.

Best regards, Christoph

Dear RAPS Community,

What is CE design certificate and it is mandatory to obtain from NB for class IIa device (SaMD)?

Thank you for any info!

Under both the outgoing Directives (e.g., MDD 93/42/EEC) and the new regulations (e.g., MDR 2017/745), whenever a chosen conformity assessment route requires Notified Body (NB) assessment of the quality management system (QMS), it is always required to be paired with NB assessment of the Technical Documentation (TD).  This always results in two certificates issued in order to gain Notified Body (NB) approval for placing the corresponding devices on the market: One certificate for the quality management system conformity, and one certificate for the device conformity. This two-certificate ensemble is a most fundamental certification requirement that is core to any successful and real CE marking experience where the manufacturer chooses a route involving QMS conformity.

For example, if the Article 52(6) first paragraph Class IIa conformity assessment route is chosen, then, by definition, that route involves two types of conformity assessments and certificates by the NB: One for the QMS plus one for the device TD. This is the way it's always been under the outgoing Directives (e.g., MDD 93/42/EEC), and this has been maintained under the new regulations (e.g., MDR 2017/745).  Specifically, for example under the MDR per Article 52(6) first paragraph, the Annex IX Chapters I & III QMS conformity assessment is performed by the NB resulting in an EU QMS assessment certificate per Annex IX section 2.3 fourth paragraph PLUS an Annex IX Chapter II section 4 TD assessment is performed by the NB resulting in an EU TD assessment certificate per Annex IX section 4.9.

As further examples, the same type of two-certificate requirement applies to the Article 52(6) second paragraph Class IIa conformity assessment route (and others for other device classes).

Hi Anon,

as Kevin says, the MDR does not use the expression "design certificate".

Annex XII that deals with the certificates issued by a notified body lists: "EU technical documentation assessment certificates, EU type-examination certificates and EU product verification certificates" in section 4(a) and "EU quality management system certificates and EU quality assurance certificates" in section 4(b). You could argue that the former are to some degree "device certificates", because they refer to an individual technical documentation, device type or even single devices, while the latter are "quality management certificates", but there is no such formulation in the MDR.

For a class IIa SaMD you usually would need either a "EU quality management system certificate" (Annex IX, chapters I and III) or a "EU quality assurence certificate" (Annex XI, Part A, Section 10). Both procedures would include the assessment of a technical documentation for at least one representative device sample for each device category, but as far as I know this assessment does not result in a separate certificate (as opposed to e.g. the full Annex IX assessment for class III devices or a type examination according to Annex X). If you only have one device, then of course the sampling in the end means that every device is assessed this way, but again this to my knowledge is not reflected in its own certificate. In theory Annex XI, Part B, Section 18, would be a possible procedure, but testing each device does not make sense for software and I do not think any notified body has this in their scope.

So in a loose sense, you do not obtain a design certificate for a class IIa software, but a quality management certificate (either full quality management system or product assurance). However, using the language of the MDR in describing the certificate is the better idea.

Best regards, Christoph

Dear RAPS Community,

What is CE design certificate and it is mandatory to obtain from NB for class IIa device (SaMD)?

Thank you for any info!

Hi Kevin,

I respectfully disagree.

Under the MDD e.g. Annex II section 7 made it quite clear that for class IIa and IIb devices section 4 of Annex II (that contains provisions for design examination and a corresponding certificate) does not apply and instead TDs are sampled as part of the audit procedure (Section 3.3 of Annex II). The certificates for the quality management system (Annex II without section 4) that I know then stated that the validity for class III devices would depend on an additional design examination certificate (Annex II section 4) but I know of no additional certificate for class IIa/IIb devices where the TDs were sampled.

The MDR (of course) makes things less clear, but I think a not unusual interpretation of the additional TD assessment of samples in the case of IIa and IIb devices for e.g. Annex IX is that the assessment again is done as part of the audit activities (Chapter I of Annex IX, where also the additional TD sampling during surveillance in described) and while it is generally performed as described in Section 4 of Annex IX, the results are a part of the Annex IX chapter I (and III) certificate. Only class III and class IIb implantable devices where every TD is assessed receive an additional certificate in line with Section 4.9 and are subject to the approval of changes detailed in Section 4.10. I also think this makes sense: Only if every TD is assessed the reporting and approval of changes makes sense. If only samples are assessed, then what would the "approved device" in section 4.10 refer to? The whole generic device group / category of devices that the sample stands for or only the TDs that happen to have been sampled? If the NB does not know the TD the manufacturer is changing (because it has not yet been sampled), how are they assumed to assess changes? Would they have to do an additional full TD assessment?

And so far the MDR certificates I have seen are comparable to the ones under MDD, a certificate according to Annex IX Chapter I (and III) with the additional remark that in case of class III and IIb implantable devices an additional Annex IX Section 4 (or Chapter II) certificate is required (but not for class IIa/IIb).

However, I admit this is less clearly expressed than it was under the MDD and there might be NBs that interpret this differently and hand out Annex IX Chapter II certificates for class IIa/IIb (non implantable) devices. If you have examples for this, I would be interested to know them.

Best regards, Christoph

Under both the outgoing Directives (e.g., MDD 93/42/EEC) and the new regulations (e.g., MDR 2017/745), whenever a chosen conformity assessment route requires Notified Body (NB) assessment of the quality management system (QMS), it is always required to be paired with NB assessment of the Technical Documentation (TD).  This always results in two certificates issued in order to gain Notified Body (NB) approval for placing the corresponding devices on the market: One certificate for the quality management system conformity, and one certificate for the device conformity. This two-certificate ensemble is a most fundamental certification requirement that is core to any successful and real CE marking experience where the manufacturer chooses a route involving QMS conformity.

For example, if the Article 52(6) first paragraph Class IIa conformity assessment route is chosen, then, by definition, that route involves two types of conformity assessments and certificates by the NB: One for the QMS plus one for the device TD. This is the way it's always been under the outgoing Directives (e.g., MDD 93/42/EEC), and this has been maintained under the new regulations (e.g., MDR 2017/745).  Specifically, for example under the MDR per Article 52(6) first paragraph, the Annex IX Chapters I & III QMS conformity assessment is performed by the NB resulting in an EU QMS assessment certificate per Annex IX section 2.3 fourth paragraph PLUS an Annex IX Chapter II section 4 TD assessment is performed by the NB resulting in an EU TD assessment certificate per Annex IX section 4.9.

As further examples, the same type of two-certificate requirement applies to the Article 52(6) second paragraph Class IIa conformity assessment route (and others for other device classes).

Hi Anon,

as Kevin says, the MDR does not use the expression "design certificate".

Annex XII that deals with the certificates issued by a notified body lists: "EU technical documentation assessment certificates, EU type-examination certificates and EU product verification certificates" in section 4(a) and "EU quality management system certificates and EU quality assurance certificates" in section 4(b). You could argue that the former are to some degree "device certificates", because they refer to an individual technical documentation, device type or even single devices, while the latter are "quality management certificates", but there is no such formulation in the MDR.

For a class IIa SaMD you usually would need either a "EU quality management system certificate" (Annex IX, chapters I and III) or a "EU quality assurence certificate" (Annex XI, Part A, Section 10). Both procedures would include the assessment of a technical documentation for at least one representative device sample for each device category, but as far as I know this assessment does not result in a separate certificate (as opposed to e.g. the full Annex IX assessment for class III devices or a type examination according to Annex X). If you only have one device, then of course the sampling in the end means that every device is assessed this way, but again this to my knowledge is not reflected in its own certificate. In theory Annex XI, Part B, Section 18, would be a possible procedure, but testing each device does not make sense for software and I do not think any notified body has this in their scope.

So in a loose sense, you do not obtain a design certificate for a class IIa software, but a quality management certificate (either full quality management system or product assurance). However, using the language of the MDR in describing the certificate is the better idea.

Best regards, Christoph

Dear RAPS Community,

What is CE design certificate and it is mandatory to obtain from NB for class IIa device (SaMD)?

Thank you for any info!

Christoph,

I can't show actual clients' examples; but I'll be creative to show a couple Class IIa examples that are publicly/readily available among others on the internet:

• Example 1
• Example 2

Therein on these Class IIa device Annex IX Chapter II EU Technical Documentation Assessment Certificates is stated, "For placing devices listed in the Annex on the market, an additional certificate according to Annex IX, Chapter I and III is required."

Along these lines during my 28-year tenure with medical device CE marking, I've seen many instances of the MDD Article 11(2) second paragraph Class IIa Annex II full quality assurance route generating a QMS certificate plus an EC certificate for the chosen assessment route through the most well-known NBs.

All this said, your insights have educated me to the fact that there does seem to be more NBs employing the single-certificate route for Class IIa devices. For example, it hadn't yet dawned on me that BSI and TUV seem to be sticking to that single-certificate approach for Class IIa Annex IX devices.

My understanding is that the NBs have freedom regarding the particular documented format/way in which they answer the aforesaid two separate, distinct, concomitant Class IIa Annex IX certificate requirements (QMS certificate + TD certificate).  Thus, this liberty may be why we are indeed seeing some use a single-certificate Class IIa Annex IX approach where that one certificate is in fact entitled "EU Quality Management System Assessment Certificate".  But I don't believe that is in clear alignment with the aforesaid two-certificate requirement.  Given how much trouble NBs have had in recent years issuing timely certificates, perhaps they are stretching this issue to reduce the administrative burden rather than following the approach in the examples above which seems to more clearly match the intent of the MDR.

Hi Kevin,

I respectfully disagree.

Under the MDD e.g. Annex II section 7 made it quite clear that for class IIa and IIb devices section 4 of Annex II (that contains provisions for design examination and a corresponding certificate) does not apply and instead TDs are sampled as part of the audit procedure (Section 3.3 of Annex II). The certificates for the quality management system (Annex II without section 4) that I know then stated that the validity for class III devices would depend on an additional design examination certificate (Annex II section 4) but I know of no additional certificate for class IIa/IIb devices where the TDs were sampled.

The MDR (of course) makes things less clear, but I think a not unusual interpretation of the additional TD assessment of samples in the case of IIa and IIb devices for e.g. Annex IX is that the assessment again is done as part of the audit activities (Chapter I of Annex IX, where also the additional TD sampling during surveillance in described) and while it is generally performed as described in Section 4 of Annex IX, the results are a part of the Annex IX chapter I (and III) certificate. Only class III and class IIb implantable devices where every TD is assessed receive an additional certificate in line with Section 4.9 and are subject to the approval of changes detailed in Section 4.10. I also think this makes sense: Only if every TD is assessed the reporting and approval of changes makes sense. If only samples are assessed, then what would the "approved device" in section 4.10 refer to? The whole generic device group / category of devices that the sample stands for or only the TDs that happen to have been sampled? If the NB does not know the TD the manufacturer is changing (because it has not yet been sampled), how are they assumed to assess changes? Would they have to do an additional full TD assessment?

And so far the MDR certificates I have seen are comparable to the ones under MDD, a certificate according to Annex IX Chapter I (and III) with the additional remark that in case of class III and IIb implantable devices an additional Annex IX Section 4 (or Chapter II) certificate is required (but not for class IIa/IIb).

However, I admit this is less clearly expressed than it was under the MDD and there might be NBs that interpret this differently and hand out Annex IX Chapter II certificates for class IIa/IIb (non implantable) devices. If you have examples for this, I would be interested to know them.

Best regards, Christoph

Under both the outgoing Directives (e.g., MDD 93/42/EEC) and the new regulations (e.g., MDR 2017/745), whenever a chosen conformity assessment route requires Notified Body (NB) assessment of the quality management system (QMS), it is always required to be paired with NB assessment of the Technical Documentation (TD).  This always results in two certificates issued in order to gain Notified Body (NB) approval for placing the corresponding devices on the market: One certificate for the quality management system conformity, and one certificate for the device conformity. This two-certificate ensemble is a most fundamental certification requirement that is core to any successful and real CE marking experience where the manufacturer chooses a route involving QMS conformity.

For example, if the Article 52(6) first paragraph Class IIa conformity assessment route is chosen, then, by definition, that route involves two types of conformity assessments and certificates by the NB: One for the QMS plus one for the device TD. This is the way it's always been under the outgoing Directives (e.g., MDD 93/42/EEC), and this has been maintained under the new regulations (e.g., MDR 2017/745).  Specifically, for example under the MDR per Article 52(6) first paragraph, the Annex IX Chapters I & III QMS conformity assessment is performed by the NB resulting in an EU QMS assessment certificate per Annex IX section 2.3 fourth paragraph PLUS an Annex IX Chapter II section 4 TD assessment is performed by the NB resulting in an EU TD assessment certificate per Annex IX section 4.9.

As further examples, the same type of two-certificate requirement applies to the Article 52(6) second paragraph Class IIa conformity assessment route (and others for other device classes).

Hi Anon,

as Kevin says, the MDR does not use the expression "design certificate".

Annex XII that deals with the certificates issued by a notified body lists: "EU technical documentation assessment certificates, EU type-examination certificates and EU product verification certificates" in section 4(a) and "EU quality management system certificates and EU quality assurance certificates" in section 4(b). You could argue that the former are to some degree "device certificates", because they refer to an individual technical documentation, device type or even single devices, while the latter are "quality management certificates", but there is no such formulation in the MDR.

For a class IIa SaMD you usually would need either a "EU quality management system certificate" (Annex IX, chapters I and III) or a "EU quality assurence certificate" (Annex XI, Part A, Section 10). Both procedures would include the assessment of a technical documentation for at least one representative device sample for each device category, but as far as I know this assessment does not result in a separate certificate (as opposed to e.g. the full Annex IX assessment for class III devices or a type examination according to Annex X). If you only have one device, then of course the sampling in the end means that every device is assessed this way, but again this to my knowledge is not reflected in its own certificate. In theory Annex XI, Part B, Section 18, would be a possible procedure, but testing each device does not make sense for software and I do not think any notified body has this in their scope.

So in a loose sense, you do not obtain a design certificate for a class IIa software, but a quality management certificate (either full quality management system or product assurance). However, using the language of the MDR in describing the certificate is the better idea.

Best regards, Christoph

Dear RAPS Community,

What is CE design certificate and it is mandatory to obtain from NB for class IIa device (SaMD)?

Thank you for any info!

Hi Kevin,

thank you for providing the examples, that was definitely something I had not realised before. Just for others reading this here are two examples for publicly available current MDR certificates by the mentioned TÜV Süd and BSI that either declare that TD assessment is part of this certificate or that an Annex IX Chapter II certificate is only required for class III and IIb implantable devices (where every TD must be assessed).

With respect to the MDD: At least the old NBOG Best Practice Guide (2010-3) does not foresee a combination of Annex II excluding (4) with Annex II (4) ( for class IIa and IIb, but exclusively for class III. And in the corresponding MDD certificates to the examples you provided DQS used the singe-certificate approach under MDD (see here and here, where an additional Annex II section 4 certificate is limited to class III devices).

Obviously there is some freedom in the interpretaton of the MDR and the corresponding use of certificates. However I would really prefer if there was a common approach by the NBs. I will try to reach out to Team NB and see if they could imagine a common approach for their members.

But aside from the number of certificates, the main impact if Annex IX Chapter II is fully applicable to class IIa and IIb non-implantable devices (aside from the sampling aspect) in my opinion would be that this would also mean that significant design changes must be approved by the Notified Body and lead to a supplement to the (then existing) TD-certificate (of course also under the single-certificate approach each NB might have additional requirements on what needs to be reported by the manufacturer that might be more extensive, but I still think this would be different from a legal requirement with mandatory consequences for the certificates). At least in my experience that was handled differently under MDD for Annex II excluding (4) (at least if you go with the single-certificate approach and the conformity assessment routes given in NBOG 2010-3, see NBOG 2014-3).

There currently is no MDCG guidance on the interpretation of the need for approval of significant changes (with the exception of the specific case of Article 120), so their position is currently not clear to me. But how would this be handled by the NB in the case of TD sampling? If the manufacturer makes a significant change to a device within a group with a TD that has not yet been assessed as a sample, how does the NB assess the change? Based on the change notification only without any knowledge about the previous state of the TD? Do they perform an additional full TD assessment for this device? Or is the assessment only necessary if the TD already had been sampled? Maybe I am wrong but I think this would require a fair amount of additional resources and time, both currently in rather short supply.

Best regards, Christoph

Christoph,

I can't show actual clients' examples; but I'll be creative to show a couple Class IIa examples that are publicly/readily available among others on the internet:

• Example 1
• Example 2

Therein on these Class IIa device Annex IX Chapter II EU Technical Documentation Assessment Certificates is stated, "For placing devices listed in the Annex on the market, an additional certificate according to Annex IX, Chapter I and III is required."

Along these lines during my 28-year tenure with medical device CE marking, I've seen many instances of the MDD Article 11(2) second paragraph Class IIa Annex II full quality assurance route generating a QMS certificate plus an EC certificate for the chosen assessment route through the most well-known NBs.

All this said, your insights have educated me to the fact that there does seem to be more NBs employing the single-certificate route for Class IIa devices. For example, it hadn't yet dawned on me that BSI and TUV seem to be sticking to that single-certificate approach for Class IIa Annex IX devices.

My understanding is that the NBs have freedom regarding the particular documented format/way in which they answer the aforesaid two separate, distinct, concomitant Class IIa Annex IX certificate requirements (QMS certificate + TD certificate).  Thus, this liberty may be why we are indeed seeing some use a single-certificate Class IIa Annex IX approach where that one certificate is in fact entitled "EU Quality Management System Assessment Certificate".  But I don't believe that is in clear alignment with the aforesaid two-certificate requirement.  Given how much trouble NBs have had in recent years issuing timely certificates, perhaps they are stretching this issue to reduce the administrative burden rather than following the approach in the examples above which seems to more clearly match the intent of the MDR.

Hi Kevin,

I respectfully disagree.

Under the MDD e.g. Annex II section 7 made it quite clear that for class IIa and IIb devices section 4 of Annex II (that contains provisions for design examination and a corresponding certificate) does not apply and instead TDs are sampled as part of the audit procedure (Section 3.3 of Annex II). The certificates for the quality management system (Annex II without section 4) that I know then stated that the validity for class III devices would depend on an additional design examination certificate (Annex II section 4) but I know of no additional certificate for class IIa/IIb devices where the TDs were sampled.

The MDR (of course) makes things less clear, but I think a not unusual interpretation of the additional TD assessment of samples in the case of IIa and IIb devices for e.g. Annex IX is that the assessment again is done as part of the audit activities (Chapter I of Annex IX, where also the additional TD sampling during surveillance in described) and while it is generally performed as described in Section 4 of Annex IX, the results are a part of the Annex IX chapter I (and III) certificate. Only class III and class IIb implantable devices where every TD is assessed receive an additional certificate in line with Section 4.9 and are subject to the approval of changes detailed in Section 4.10. I also think this makes sense: Only if every TD is assessed the reporting and approval of changes makes sense. If only samples are assessed, then what would the "approved device" in section 4.10 refer to? The whole generic device group / category of devices that the sample stands for or only the TDs that happen to have been sampled? If the NB does not know the TD the manufacturer is changing (because it has not yet been sampled), how are they assumed to assess changes? Would they have to do an additional full TD assessment?

And so far the MDR certificates I have seen are comparable to the ones under MDD, a certificate according to Annex IX Chapter I (and III) with the additional remark that in case of class III and IIb implantable devices an additional Annex IX Section 4 (or Chapter II) certificate is required (but not for class IIa/IIb).

However, I admit this is less clearly expressed than it was under the MDD and there might be NBs that interpret this differently and hand out Annex IX Chapter II certificates for class IIa/IIb (non implantable) devices. If you have examples for this, I would be interested to know them.

Best regards, Christoph

Under both the outgoing Directives (e.g., MDD 93/42/EEC) and the new regulations (e.g., MDR 2017/745), whenever a chosen conformity assessment route requires Notified Body (NB) assessment of the quality management system (QMS), it is always required to be paired with NB assessment of the Technical Documentation (TD).  This always results in two certificates issued in order to gain Notified Body (NB) approval for placing the corresponding devices on the market: One certificate for the quality management system conformity, and one certificate for the device conformity. This two-certificate ensemble is a most fundamental certification requirement that is core to any successful and real CE marking experience where the manufacturer chooses a route involving QMS conformity.

For example, if the Article 52(6) first paragraph Class IIa conformity assessment route is chosen, then, by definition, that route involves two types of conformity assessments and certificates by the NB: One for the QMS plus one for the device TD. This is the way it's always been under the outgoing Directives (e.g., MDD 93/42/EEC), and this has been maintained under the new regulations (e.g., MDR 2017/745).  Specifically, for example under the MDR per Article 52(6) first paragraph, the Annex IX Chapters I & III QMS conformity assessment is performed by the NB resulting in an EU QMS assessment certificate per Annex IX section 2.3 fourth paragraph PLUS an Annex IX Chapter II section 4 TD assessment is performed by the NB resulting in an EU TD assessment certificate per Annex IX section 4.9.

As further examples, the same type of two-certificate requirement applies to the Article 52(6) second paragraph Class IIa conformity assessment route (and others for other device classes).

Hi Anon,

as Kevin says, the MDR does not use the expression "design certificate".

Annex XII that deals with the certificates issued by a notified body lists: "EU technical documentation assessment certificates, EU type-examination certificates and EU product verification certificates" in section 4(a) and "EU quality management system certificates and EU quality assurance certificates" in section 4(b). You could argue that the former are to some degree "device certificates", because they refer to an individual technical documentation, device type or even single devices, while the latter are "quality management certificates", but there is no such formulation in the MDR.

For a class IIa SaMD you usually would need either a "EU quality management system certificate" (Annex IX, chapters I and III) or a "EU quality assurence certificate" (Annex XI, Part A, Section 10). Both procedures would include the assessment of a technical documentation for at least one representative device sample for each device category, but as far as I know this assessment does not result in a separate certificate (as opposed to e.g. the full Annex IX assessment for class III devices or a type examination according to Annex X). If you only have one device, then of course the sampling in the end means that every device is assessed this way, but again this to my knowledge is not reflected in its own certificate. In theory Annex XI, Part B, Section 18, would be a possible procedure, but testing each device does not make sense for software and I do not think any notified body has this in their scope.

So in a loose sense, you do not obtain a design certificate for a class IIa software, but a quality management certificate (either full quality management system or product assurance). However, using the language of the MDR in describing the certificate is the better idea.

Best regards, Christoph

Dear RAPS Community,

What is CE design certificate and it is mandatory to obtain from NB for class IIa device (SaMD)?

Thank you for any info!

Thank you, Christoph. NBOG-3 (addressing certificate practices for the outgoing MDD) doesn't directly foresee a two-certificate approach yet does so indirectly in my experience.  Specifically, if the manufacturer, like most do, has chosen to employ an ISO standard as its means for meeting the Annex II Full Quality Assurance QMS requirements, then there is required to be a separate QMS certificate that is the basis for the Annex II Full Quality Assurance EC Certificate.

Also of note is NB-MED/2.5.1/Rec4 stating that the matter of interlinking of certificates has been considered and is recognized to be an administrative responsibility of the NB and should be traceable through the file of the NB. Inclusion of a cross-reference on the certificates themselves, however, would lead to practical difficulties.  Thus, the lack of such cross-references (like in the MDD certificates your cited) certainly doesn't necessarily mean that the certificate stands on its own.

Regarding the various additional questions you've raised (i.e., handling of significant changes for class IIa devices; NB practices for TD sampling after significant changes), those seem to be well beyond the scope of Anon's question.  I certainly have answers for you on those peripheral questions; but I think it's most appropriate to raise them in a separate threat so that this current thread can stay on track with Anon's intended question.

Hi Kevin,

thank you for providing the examples, that was definitely something I had not realised before. Just for others reading this here are two examples for publicly available current MDR certificates by the mentioned TÜV Süd and BSI that either declare that TD assessment is part of this certificate or that an Annex IX Chapter II certificate is only required for class III and IIb implantable devices (where every TD must be assessed).

With respect to the MDD: At least the old NBOG Best Practice Guide (2010-3) does not foresee a combination of Annex II excluding (4) with Annex II (4) ( for class IIa and IIb, but exclusively for class III. And in the corresponding MDD certificates to the examples you provided DQS used the singe-certificate approach under MDD (see here and here, where an additional Annex II section 4 certificate is limited to class III devices).

Obviously there is some freedom in the interpretaton of the MDR and the corresponding use of certificates. However I would really prefer if there was a common approach by the NBs. I will try to reach out to Team NB and see if they could imagine a common approach for their members.

But aside from the number of certificates, the main impact if Annex IX Chapter II is fully applicable to class IIa and IIb non-implantable devices (aside from the sampling aspect) in my opinion would be that this would also mean that significant design changes must be approved by the Notified Body and lead to a supplement to the (then existing) TD-certificate (of course also under the single-certificate approach each NB might have additional requirements on what needs to be reported by the manufacturer that might be more extensive, but I still think this would be different from a legal requirement with mandatory consequences for the certificates). At least in my experience that was handled differently under MDD for Annex II excluding (4) (at least if you go with the single-certificate approach and the conformity assessment routes given in NBOG 2010-3, see NBOG 2014-3).

There currently is no MDCG guidance on the interpretation of the need for approval of significant changes (with the exception of the specific case of Article 120), so their position is currently not clear to me. But how would this be handled by the NB in the case of TD sampling? If the manufacturer makes a significant change to a device within a group with a TD that has not yet been assessed as a sample, how does the NB assess the change? Based on the change notification only without any knowledge about the previous state of the TD? Do they perform an additional full TD assessment for this device? Or is the assessment only necessary if the TD already had been sampled? Maybe I am wrong but I think this would require a fair amount of additional resources and time, both currently in rather short supply.

Best regards, Christoph

Christoph,

I can't show actual clients' examples; but I'll be creative to show a couple Class IIa examples that are publicly/readily available among others on the internet:

• Example 1
• Example 2

Therein on these Class IIa device Annex IX Chapter II EU Technical Documentation Assessment Certificates is stated, "For placing devices listed in the Annex on the market, an additional certificate according to Annex IX, Chapter I and III is required."

Along these lines during my 28-year tenure with medical device CE marking, I've seen many instances of the MDD Article 11(2) second paragraph Class IIa Annex II full quality assurance route generating a QMS certificate plus an EC certificate for the chosen assessment route through the most well-known NBs.

All this said, your insights have educated me to the fact that there does seem to be more NBs employing the single-certificate route for Class IIa devices. For example, it hadn't yet dawned on me that BSI and TUV seem to be sticking to that single-certificate approach for Class IIa Annex IX devices.

My understanding is that the NBs have freedom regarding the particular documented format/way in which they answer the aforesaid two separate, distinct, concomitant Class IIa Annex IX certificate requirements (QMS certificate + TD certificate).  Thus, this liberty may be why we are indeed seeing some use a single-certificate Class IIa Annex IX approach where that one certificate is in fact entitled "EU Quality Management System Assessment Certificate".  But I don't believe that is in clear alignment with the aforesaid two-certificate requirement.  Given how much trouble NBs have had in recent years issuing timely certificates, perhaps they are stretching this issue to reduce the administrative burden rather than following the approach in the examples above which seems to more clearly match the intent of the MDR.

Hi Kevin,

I respectfully disagree.

Under the MDD e.g. Annex II section 7 made it quite clear that for class IIa and IIb devices section 4 of Annex II (that contains provisions for design examination and a corresponding certificate) does not apply and instead TDs are sampled as part of the audit procedure (Section 3.3 of Annex II). The certificates for the quality management system (Annex II without section 4) that I know then stated that the validity for class III devices would depend on an additional design examination certificate (Annex II section 4) but I know of no additional certificate for class IIa/IIb devices where the TDs were sampled.

The MDR (of course) makes things less clear, but I think a not unusual interpretation of the additional TD assessment of samples in the case of IIa and IIb devices for e.g. Annex IX is that the assessment again is done as part of the audit activities (Chapter I of Annex IX, where also the additional TD sampling during surveillance in described) and while it is generally performed as described in Section 4 of Annex IX, the results are a part of the Annex IX chapter I (and III) certificate. Only class III and class IIb implantable devices where every TD is assessed receive an additional certificate in line with Section 4.9 and are subject to the approval of changes detailed in Section 4.10. I also think this makes sense: Only if every TD is assessed the reporting and approval of changes makes sense. If only samples are assessed, then what would the "approved device" in section 4.10 refer to? The whole generic device group / category of devices that the sample stands for or only the TDs that happen to have been sampled? If the NB does not know the TD the manufacturer is changing (because it has not yet been sampled), how are they assumed to assess changes? Would they have to do an additional full TD assessment?

And so far the MDR certificates I have seen are comparable to the ones under MDD, a certificate according to Annex IX Chapter I (and III) with the additional remark that in case of class III and IIb implantable devices an additional Annex IX Section 4 (or Chapter II) certificate is required (but not for class IIa/IIb).

However, I admit this is less clearly expressed than it was under the MDD and there might be NBs that interpret this differently and hand out Annex IX Chapter II certificates for class IIa/IIb (non implantable) devices. If you have examples for this, I would be interested to know them.

Best regards, Christoph

Under both the outgoing Directives (e.g., MDD 93/42/EEC) and the new regulations (e.g., MDR 2017/745), whenever a chosen conformity assessment route requires Notified Body (NB) assessment of the quality management system (QMS), it is always required to be paired with NB assessment of the Technical Documentation (TD).  This always results in two certificates issued in order to gain Notified Body (NB) approval for placing the corresponding devices on the market: One certificate for the quality management system conformity, and one certificate for the device conformity. This two-certificate ensemble is a most fundamental certification requirement that is core to any successful and real CE marking experience where the manufacturer chooses a route involving QMS conformity.

For example, if the Article 52(6) first paragraph Class IIa conformity assessment route is chosen, then, by definition, that route involves two types of conformity assessments and certificates by the NB: One for the QMS plus one for the device TD. This is the way it's always been under the outgoing Directives (e.g., MDD 93/42/EEC), and this has been maintained under the new regulations (e.g., MDR 2017/745).  Specifically, for example under the MDR per Article 52(6) first paragraph, the Annex IX Chapters I & III QMS conformity assessment is performed by the NB resulting in an EU QMS assessment certificate per Annex IX section 2.3 fourth paragraph PLUS an Annex IX Chapter II section 4 TD assessment is performed by the NB resulting in an EU TD assessment certificate per Annex IX section 4.9.

As further examples, the same type of two-certificate requirement applies to the Article 52(6) second paragraph Class IIa conformity assessment route (and others for other device classes).

Hi Anon,

as Kevin says, the MDR does not use the expression "design certificate".

Annex XII that deals with the certificates issued by a notified body lists: "EU technical documentation assessment certificates, EU type-examination certificates and EU product verification certificates" in section 4(a) and "EU quality management system certificates and EU quality assurance certificates" in section 4(b). You could argue that the former are to some degree "device certificates", because they refer to an individual technical documentation, device type or even single devices, while the latter are "quality management certificates", but there is no such formulation in the MDR.

For a class IIa SaMD you usually would need either a "EU quality management system certificate" (Annex IX, chapters I and III) or a "EU quality assurence certificate" (Annex XI, Part A, Section 10). Both procedures would include the assessment of a technical documentation for at least one representative device sample for each device category, but as far as I know this assessment does not result in a separate certificate (as opposed to e.g. the full Annex IX assessment for class III devices or a type examination according to Annex X). If you only have one device, then of course the sampling in the end means that every device is assessed this way, but again this to my knowledge is not reflected in its own certificate. In theory Annex XI, Part B, Section 18, would be a possible procedure, but testing each device does not make sense for software and I do not think any notified body has this in their scope.

So in a loose sense, you do not obtain a design certificate for a class IIa software, but a quality management certificate (either full quality management system or product assurance). However, using the language of the MDR in describing the certificate is the better idea.

Best regards, Christoph

Dear RAPS Community,

What is CE design certificate and it is mandatory to obtain from NB for class IIa device (SaMD)?

Thank you for any info!

Hi Kevin,

yes, I think I got a little bit carried away by the topic. Thank you for discussing this, it definitely brought up some points I would like to try to clarify.

I will do some research if there is some kind of consensus between NBs or an MDCG opinion on this and see if it makes sense to post the results in a separate thread.

Best regards, Christoph

Thank you, Christoph. NBOG-3 (addressing certificate practices for the outgoing MDD) doesn't directly foresee a two-certificate approach yet does so indirectly in my experience.  Specifically, if the manufacturer, like most do, has chosen to employ an ISO standard as its means for meeting the Annex II Full Quality Assurance QMS requirements, then there is required to be a separate QMS certificate that is the basis for the Annex II Full Quality Assurance EC Certificate.

Also of note is NB-MED/2.5.1/Rec4 stating that the matter of interlinking of certificates has been considered and is recognized to be an administrative responsibility of the NB and should be traceable through the file of the NB. Inclusion of a cross-reference on the certificates themselves, however, would lead to practical difficulties.  Thus, the lack of such cross-references (like in the MDD certificates your cited) certainly doesn't necessarily mean that the certificate stands on its own.

Regarding the various additional questions you've raised (i.e., handling of significant changes for class IIa devices; NB practices for TD sampling after significant changes), those seem to be well beyond the scope of Anon's question.  I certainly have answers for you on those peripheral questions; but I think it's most appropriate to raise them in a separate threat so that this current thread can stay on track with Anon's intended question.

Hi Kevin,

thank you for providing the examples, that was definitely something I had not realised before. Just for others reading this here are two examples for publicly available current MDR certificates by the mentioned TÜV Süd and BSI that either declare that TD assessment is part of this certificate or that an Annex IX Chapter II certificate is only required for class III and IIb implantable devices (where every TD must be assessed).

With respect to the MDD: At least the old NBOG Best Practice Guide (2010-3) does not foresee a combination of Annex II excluding (4) with Annex II (4) ( for class IIa and IIb, but exclusively for class III. And in the corresponding MDD certificates to the examples you provided DQS used the singe-certificate approach under MDD (see here and here, where an additional Annex II section 4 certificate is limited to class III devices).

Obviously there is some freedom in the interpretaton of the MDR and the corresponding use of certificates. However I would really prefer if there was a common approach by the NBs. I will try to reach out to Team NB and see if they could imagine a common approach for their members.

But aside from the number of certificates, the main impact if Annex IX Chapter II is fully applicable to class IIa and IIb non-implantable devices (aside from the sampling aspect) in my opinion would be that this would also mean that significant design changes must be approved by the Notified Body and lead to a supplement to the (then existing) TD-certificate (of course also under the single-certificate approach each NB might have additional requirements on what needs to be reported by the manufacturer that might be more extensive, but I still think this would be different from a legal requirement with mandatory consequences for the certificates). At least in my experience that was handled differently under MDD for Annex II excluding (4) (at least if you go with the single-certificate approach and the conformity assessment routes given in NBOG 2010-3, see NBOG 2014-3).

There currently is no MDCG guidance on the interpretation of the need for approval of significant changes (with the exception of the specific case of Article 120), so their position is currently not clear to me. But how would this be handled by the NB in the case of TD sampling? If the manufacturer makes a significant change to a device within a group with a TD that has not yet been assessed as a sample, how does the NB assess the change? Based on the change notification only without any knowledge about the previous state of the TD? Do they perform an additional full TD assessment for this device? Or is the assessment only necessary if the TD already had been sampled? Maybe I am wrong but I think this would require a fair amount of additional resources and time, both currently in rather short supply.

Best regards, Christoph

Christoph,

I can't show actual clients' examples; but I'll be creative to show a couple Class IIa examples that are publicly/readily available among others on the internet:

• Example 1
• Example 2

Therein on these Class IIa device Annex IX Chapter II EU Technical Documentation Assessment Certificates is stated, "For placing devices listed in the Annex on the market, an additional certificate according to Annex IX, Chapter I and III is required."

Along these lines during my 28-year tenure with medical device CE marking, I've seen many instances of the MDD Article 11(2) second paragraph Class IIa Annex II full quality assurance route generating a QMS certificate plus an EC certificate for the chosen assessment route through the most well-known NBs.

All this said, your insights have educated me to the fact that there does seem to be more NBs employing the single-certificate route for Class IIa devices. For example, it hadn't yet dawned on me that BSI and TUV seem to be sticking to that single-certificate approach for Class IIa Annex IX devices.

My understanding is that the NBs have freedom regarding the particular documented format/way in which they answer the aforesaid two separate, distinct, concomitant Class IIa Annex IX certificate requirements (QMS certificate + TD certificate).  Thus, this liberty may be why we are indeed seeing some use a single-certificate Class IIa Annex IX approach where that one certificate is in fact entitled "EU Quality Management System Assessment Certificate".  But I don't believe that is in clear alignment with the aforesaid two-certificate requirement.  Given how much trouble NBs have had in recent years issuing timely certificates, perhaps they are stretching this issue to reduce the administrative burden rather than following the approach in the examples above which seems to more clearly match the intent of the MDR.

Hi Kevin,

I respectfully disagree.

Under the MDD e.g. Annex II section 7 made it quite clear that for class IIa and IIb devices section 4 of Annex II (that contains provisions for design examination and a corresponding certificate) does not apply and instead TDs are sampled as part of the audit procedure (Section 3.3 of Annex II). The certificates for the quality management system (Annex II without section 4) that I know then stated that the validity for class III devices would depend on an additional design examination certificate (Annex II section 4) but I know of no additional certificate for class IIa/IIb devices where the TDs were sampled.

The MDR (of course) makes things less clear, but I think a not unusual interpretation of the additional TD assessment of samples in the case of IIa and IIb devices for e.g. Annex IX is that the assessment again is done as part of the audit activities (Chapter I of Annex IX, where also the additional TD sampling during surveillance in described) and while it is generally performed as described in Section 4 of Annex IX, the results are a part of the Annex IX chapter I (and III) certificate. Only class III and class IIb implantable devices where every TD is assessed receive an additional certificate in line with Section 4.9 and are subject to the approval of changes detailed in Section 4.10. I also think this makes sense: Only if every TD is assessed the reporting and approval of changes makes sense. If only samples are assessed, then what would the "approved device" in section 4.10 refer to? The whole generic device group / category of devices that the sample stands for or only the TDs that happen to have been sampled? If the NB does not know the TD the manufacturer is changing (because it has not yet been sampled), how are they assumed to assess changes? Would they have to do an additional full TD assessment?

And so far the MDR certificates I have seen are comparable to the ones under MDD, a certificate according to Annex IX Chapter I (and III) with the additional remark that in case of class III and IIb implantable devices an additional Annex IX Section 4 (or Chapter II) certificate is required (but not for class IIa/IIb).

However, I admit this is less clearly expressed than it was under the MDD and there might be NBs that interpret this differently and hand out Annex IX Chapter II certificates for class IIa/IIb (non implantable) devices. If you have examples for this, I would be interested to know them.

Best regards, Christoph

Under both the outgoing Directives (e.g., MDD 93/42/EEC) and the new regulations (e.g., MDR 2017/745), whenever a chosen conformity assessment route requires Notified Body (NB) assessment of the quality management system (QMS), it is always required to be paired with NB assessment of the Technical Documentation (TD).  This always results in two certificates issued in order to gain Notified Body (NB) approval for placing the corresponding devices on the market: One certificate for the quality management system conformity, and one certificate for the device conformity. This two-certificate ensemble is a most fundamental certification requirement that is core to any successful and real CE marking experience where the manufacturer chooses a route involving QMS conformity.

For example, if the Article 52(6) first paragraph Class IIa conformity assessment route is chosen, then, by definition, that route involves two types of conformity assessments and certificates by the NB: One for the QMS plus one for the device TD. This is the way it's always been under the outgoing Directives (e.g., MDD 93/42/EEC), and this has been maintained under the new regulations (e.g., MDR 2017/745).  Specifically, for example under the MDR per Article 52(6) first paragraph, the Annex IX Chapters I & III QMS conformity assessment is performed by the NB resulting in an EU QMS assessment certificate per Annex IX section 2.3 fourth paragraph PLUS an Annex IX Chapter II section 4 TD assessment is performed by the NB resulting in an EU TD assessment certificate per Annex IX section 4.9.

As further examples, the same type of two-certificate requirement applies to the Article 52(6) second paragraph Class IIa conformity assessment route (and others for other device classes).

Hi Anon,

as Kevin says, the MDR does not use the expression "design certificate".

Annex XII that deals with the certificates issued by a notified body lists: "EU technical documentation assessment certificates, EU type-examination certificates and EU product verification certificates" in section 4(a) and "EU quality management system certificates and EU quality assurance certificates" in section 4(b). You could argue that the former are to some degree "device certificates", because they refer to an individual technical documentation, device type or even single devices, while the latter are "quality management certificates", but there is no such formulation in the MDR.

For a class IIa SaMD you usually would need either a "EU quality management system certificate" (Annex IX, chapters I and III) or a "EU quality assurence certificate" (Annex XI, Part A, Section 10). Both procedures would include the assessment of a technical documentation for at least one representative device sample for each device category, but as far as I know this assessment does not result in a separate certificate (as opposed to e.g. the full Annex IX assessment for class III devices or a type examination according to Annex X). If you only have one device, then of course the sampling in the end means that every device is assessed this way, but again this to my knowledge is not reflected in its own certificate. In theory Annex XI, Part B, Section 18, would be a possible procedure, but testing each device does not make sense for software and I do not think any notified body has this in their scope.

So in a loose sense, you do not obtain a design certificate for a class IIa software, but a quality management certificate (either full quality management system or product assurance). However, using the language of the MDR in describing the certificate is the better idea.

Best regards, Christoph

Dear RAPS Community,

What is CE design certificate and it is mandatory to obtain from NB for class IIa device (SaMD)?

Thank you for any info!

I know we've gone some way from answering the initial question, but I thought I would just make a couple of comments.

I agree with Christoph's interpretation of both MDD and MDR, in that you generally only need a QMS certificate (MDD Annex II/MDR Annex IX Parts I and III) unless you have a Class III device, or there is a special category of Class IIb device involved.

It seems to me that the current confusion is caused by an ambiguity in the MDR (what a surprise!). If you read MDR Annex IX para 3.5, it says:

"In the case of class IIa and class IIb devices, the surveillance assessment shall also include an assessment of the technical documentation as specified in Section 4 for the device or devices concerned on the basis of further representative samples chosen in accordance with the rationale documented by the notified body in accordance with the third paragraph of Section 2.3."

The only problem is that when you go to Section 4, it is apparent that this entire section applies only to Class III and certain Class IIb devices, as it states in the very first line:

"Assessment of the technical documentation applicable to class III devices and to the class IIb devices referred to in the second subparagraph of Article 52(4)."

So, unfortunately, the referral to Annex IX Section 4 in Section 3.5 does not make sense, as Section 4 apparently applies ONLY to Class III and certain Class IIb devices, meaning that it has no application to Class IIa devices. There is literally zero mention of 'Class IIa' anywhere in Section 4, so I am very confused by DQS issuing an 'Annex IX Chapter II' certificate for a Class IIa device, citing "Conformity assessment based on a Quality Management System and an Assessment of Technical Documentation". 

My interpretation of the MDR is that for a Class IIa device "Conformity assessment based on a Quality Management System and an Assessment of Technical Documentation". would require either:

• Annex IX Parts I and III
• Annex XI Section 10

Note that while the use of Annex IX Parts I and III would result in just one 'Quality Management System' certificate from the notified body, using Annex XI Section 10 would result in two certificates from the notified body, these being a 'Quality Assurance Certificate' plus an 'Annex XI Part A Technical Documentation Certificate'.

Just for fun, I went back to the final draft of the MDR pre-publication, copy dated March 2017, and this is what Annex IX Section 3.5 stated at that time:

"Moreover, in the case of class IIa and class IIb devices, the quality management system assessment shall be accompanied by the assessment of technical documentation for devices selected on a representative basis in accordance with Sections 4.4 to 4.8."

If this wording had been maintained, then the reference to Section 4 would have made sense, because if you just apply Sections 4.4 to 4.8 to Class IIa devices, you completely avoid the possible need for a Technical Documentation Certificate, because this is mentioned only in Section 4.9.

So not for the first time, in moving from an earlier version to a later version, the authoring team have created an ambiguity where one didn't previously exist, thus causing us all these interpretational headaches.

------------------------------
Roger Gray
VP Quality and Regulatory
Donawa Lifescience
Clinical Studies - Regulatory - Quality Systems
AR/UKRP services
Italy and UK
rgray@donawa.com
www.donawa.com

Original Message:
Sent: 25-Jan-2024 01:56
From: Christoph Kiesselbach
Subject: CE design certificate

Hi Kevin,

yes, I think I got a little bit carried away by the topic. Thank you for discussing this, it definitely brought up some points I would like to try to clarify.

I will do some research if there is some kind of consensus between NBs or an MDCG opinion on this and see if it makes sense to post the results in a separate thread.

Best regards, Christoph

------------------------------
Christoph Kiesselbach
Schrack & Partner
Reutlingen
Germany

Original Message:
Sent: 24-Jan-2024 11:47
From: Kevin Randall
Subject: CE design certificate

Thank you, Christoph. NBOG-3 (addressing certificate practices for the outgoing MDD) doesn't directly foresee a two-certificate approach yet does so indirectly in my experience.  Specifically, if the manufacturer, like most do, has chosen to employ an ISO standard as its means for meeting the Annex II Full Quality Assurance QMS requirements, then there is required to be a separate QMS certificate that is the basis for the Annex II Full Quality Assurance EC Certificate.

Also of note is NB-MED/2.5.1/Rec4 stating that the matter of interlinking of certificates has been considered and is recognized to be an administrative responsibility of the NB and should be traceable through the file of the NB. Inclusion of a cross-reference on the certificates themselves, however, would lead to practical difficulties.  Thus, the lack of such cross-references (like in the MDD certificates your cited) certainly doesn't necessarily mean that the certificate stands on its own.

Regarding the various additional questions you've raised (i.e., handling of significant changes for class IIa devices; NB practices for TD sampling after significant changes), those seem to be well beyond the scope of Anon's question.  I certainly have answers for you on those peripheral questions; but I think it's most appropriate to raise them in a separate threat so that this current thread can stay on track with Anon's intended question.

------------------------------
Kevin Randall, ASQ CQA, RAC (Europe, U.S., Canada)
Principal Consultant
Ridgway, CO
United States
© Copyright by ComplianceAcuity, Inc. All rights reserved.

Original Message:
Sent: 24-Jan-2024 03:48
From: Christoph Kiesselbach
Subject: CE design certificate

Hi Kevin,

thank you for providing the examples, that was definitely something I had not realised before. Just for others reading this here are two examples for publicly available current MDR certificates by the mentioned TÜV Süd and BSI that either declare that TD assessment is part of this certificate or that an Annex IX Chapter II certificate is only required for class III and IIb implantable devices (where every TD must be assessed).

With respect to the MDD: At least the old NBOG Best Practice Guide (2010-3) does not foresee a combination of Annex II excluding (4) with Annex II (4) ( for class IIa and IIb, but exclusively for class III. And in the corresponding MDD certificates to the examples you provided DQS used the singe-certificate approach under MDD (see here and here, where an additional Annex II section 4 certificate is limited to class III devices).

Obviously there is some freedom in the interpretaton of the MDR and the corresponding use of certificates. However I would really prefer if there was a common approach by the NBs. I will try to reach out to Team NB and see if they could imagine a common approach for their members.

But aside from the number of certificates, the main impact if Annex IX Chapter II is fully applicable to class IIa and IIb non-implantable devices (aside from the sampling aspect) in my opinion would be that this would also mean that significant design changes must be approved by the Notified Body and lead to a supplement to the (then existing) TD-certificate (of course also under the single-certificate approach each NB might have additional requirements on what needs to be reported by the manufacturer that might be more extensive, but I still think this would be different from a legal requirement with mandatory consequences for the certificates). At least in my experience that was handled differently under MDD for Annex II excluding (4) (at least if you go with the single-certificate approach and the conformity assessment routes given in NBOG 2010-3, see NBOG 2014-3).

There currently is no MDCG guidance on the interpretation of the need for approval of significant changes (with the exception of the specific case of Article 120), so their position is currently not clear to me. But how would this be handled by the NB in the case of TD sampling? If the manufacturer makes a significant change to a device within a group with a TD that has not yet been assessed as a sample, how does the NB assess the change? Based on the change notification only without any knowledge about the previous state of the TD? Do they perform an additional full TD assessment for this device? Or is the assessment only necessary if the TD already had been sampled? Maybe I am wrong but I think this would require a fair amount of additional resources and time, both currently in rather short supply.

Best regards, Christoph

------------------------------
Christoph Kiesselbach
Schrack & Partner
Reutlingen
Germany

Original Message:
Sent: 23-Jan-2024 17:00
From: Kevin Randall
Subject: CE design certificate

Christoph,

I can't show actual clients' examples; but I'll be creative to show a couple Class IIa examples that are publicly/readily available among others on the internet:

• Example 1
• Example 2

Therein on these Class IIa device Annex IX Chapter II EU Technical Documentation Assessment Certificates is stated, "For placing devices listed in the Annex on the market, an additional certificate according to Annex IX, Chapter I and III is required."

Along these lines during my 28-year tenure with medical device CE marking, I've seen many instances of the MDD Article 11(2) second paragraph Class IIa Annex II full quality assurance route generating a QMS certificate plus an EC certificate for the chosen assessment route through the most well-known NBs.

All this said, your insights have educated me to the fact that there does seem to be more NBs employing the single-certificate route for Class IIa devices. For example, it hadn't yet dawned on me that BSI and TUV seem to be sticking to that single-certificate approach for Class IIa Annex IX devices.

My understanding is that the NBs have freedom regarding the particular documented format/way in which they answer the aforesaid two separate, distinct, concomitant Class IIa Annex IX certificate requirements (QMS certificate + TD certificate).  Thus, this liberty may be why we are indeed seeing some use a single-certificate Class IIa Annex IX approach where that one certificate is in fact entitled "EU Quality Management System Assessment Certificate".  But I don't believe that is in clear alignment with the aforesaid two-certificate requirement.  Given how much trouble NBs have had in recent years issuing timely certificates, perhaps they are stretching this issue to reduce the administrative burden rather than following the approach in the examples above which seems to more clearly match the intent of the MDR.

------------------------------
Kevin Randall, ASQ CQA, RAC (Europe, U.S., Canada)
Principal Consultant
Ridgway, CO
United States
© Copyright by ComplianceAcuity, Inc. All rights reserved.

Original Message:
Sent: 23-Jan-2024 12:22
From: Christoph Kiesselbach
Subject: CE design certificate

Hi Kevin,

I respectfully disagree.

Under the MDD e.g. Annex II section 7 made it quite clear that for class IIa and IIb devices section 4 of Annex II (that contains provisions for design examination and a corresponding certificate) does not apply and instead TDs are sampled as part of the audit procedure (Section 3.3 of Annex II). The certificates for the quality management system (Annex II without section 4) that I know then stated that the validity for class III devices would depend on an additional design examination certificate (Annex II section 4) but I know of no additional certificate for class IIa/IIb devices where the TDs were sampled.

The MDR (of course) makes things less clear, but I think a not unusual interpretation of the additional TD assessment of samples in the case of IIa and IIb devices for e.g. Annex IX is that the assessment again is done as part of the audit activities (Chapter I of Annex IX, where also the additional TD sampling during surveillance in described) and while it is generally performed as described in Section 4 of Annex IX, the results are a part of the Annex IX chapter I (and III) certificate. Only class III and class IIb implantable devices where every TD is assessed receive an additional certificate in line with Section 4.9 and are subject to the approval of changes detailed in Section 4.10. I also think this makes sense: Only if every TD is assessed the reporting and approval of changes makes sense. If only samples are assessed, then what would the "approved device" in section 4.10 refer to? The whole generic device group / category of devices that the sample stands for or only the TDs that happen to have been sampled? If the NB does not know the TD the manufacturer is changing (because it has not yet been sampled), how are they assumed to assess changes? Would they have to do an additional full TD assessment?

And so far the MDR certificates I have seen are comparable to the ones under MDD, a certificate according to Annex IX Chapter I (and III) with the additional remark that in case of class III and IIb implantable devices an additional Annex IX Section 4 (or Chapter II) certificate is required (but not for class IIa/IIb).

However, I admit this is less clearly expressed than it was under the MDD and there might be NBs that interpret this differently and hand out Annex IX Chapter II certificates for class IIa/IIb (non implantable) devices. If you have examples for this, I would be interested to know them.

Best regards, Christoph

------------------------------
Christoph Kiesselbach
Schrack & Partner
Reutlingen
Germany

Original Message:
Sent: 23-Jan-2024 09:27
From: Kevin Randall
Subject: CE design certificate

Under both the outgoing Directives (e.g., MDD 93/42/EEC) and the new regulations (e.g., MDR 2017/745), whenever a chosen conformity assessment route requires Notified Body (NB) assessment of the quality management system (QMS), it is always required to be paired with NB assessment of the Technical Documentation (TD).  This always results in two certificates issued in order to gain Notified Body (NB) approval for placing the corresponding devices on the market: One certificate for the quality management system conformity, and one certificate for the device conformity. This two-certificate ensemble is a most fundamental certification requirement that is core to any successful and real CE marking experience where the manufacturer chooses a route involving QMS conformity.

For example, if the Article 52(6) first paragraph Class IIa conformity assessment route is chosen, then, by definition, that route involves two types of conformity assessments and certificates by the NB: One for the QMS plus one for the device TD. This is the way it's always been under the outgoing Directives (e.g., MDD 93/42/EEC), and this has been maintained under the new regulations (e.g., MDR 2017/745).  Specifically, for example under the MDR per Article 52(6) first paragraph, the Annex IX Chapters I & III QMS conformity assessment is performed by the NB resulting in an EU QMS assessment certificate per Annex IX section 2.3 fourth paragraph PLUS an Annex IX Chapter II section 4 TD assessment is performed by the NB resulting in an EU TD assessment certificate per Annex IX section 4.9.

As further examples, the same type of two-certificate requirement applies to the Article 52(6) second paragraph Class IIa conformity assessment route (and others for other device classes).

------------------------------
Kevin Randall, ASQ CQA, RAC (Europe, U.S., Canada)
Principal Consultant
Ridgway, CO
United States
© Copyright by ComplianceAcuity, Inc. All rights reserved.

Original Message:
Sent: 23-Jan-2024 04:11
From: Christoph Kiesselbach
Subject: CE design certificate

Hi Anon,

as Kevin says, the MDR does not use the expression "design certificate".

Annex XII that deals with the certificates issued by a notified body lists: "EU technical documentation assessment certificates, EU type-examination certificates and EU product verification certificates" in section 4(a) and "EU quality management system certificates and EU quality assurance certificates" in section 4(b). You could argue that the former are to some degree "device certificates", because they refer to an individual technical documentation, device type or even single devices, while the latter are "quality management certificates", but there is no such formulation in the MDR.

For a class IIa SaMD you usually would need either a "EU quality management system certificate" (Annex IX, chapters I and III) or a "EU quality assurence certificate" (Annex XI, Part A, Section 10). Both procedures would include the assessment of a technical documentation for at least one representative device sample for each device category, but as far as I know this assessment does not result in a separate certificate (as opposed to e.g. the full Annex IX assessment for class III devices or a type examination according to Annex X). If you only have one device, then of course the sampling in the end means that every device is assessed this way, but again this to my knowledge is not reflected in its own certificate. In theory Annex XI, Part B, Section 18, would be a possible procedure, but testing each device does not make sense for software and I do not think any notified body has this in their scope.

So in a loose sense, you do not obtain a design certificate for a class IIa software, but a quality management certificate (either full quality management system or product assurance). However, using the language of the MDR in describing the certificate is the better idea.

Best regards, Christoph

------------------------------
Christoph Kiesselbach
Schrack & Partner
Reutlingen
Germany

Original Message:
Sent: 21-Jan-2024 07:29
From: Anonymous Member
Subject: CE design certificate

This message was posted by a user wishing to remain anonymous

Dear RAPS Community,

What is CE design certificate and it is mandatory to obtain from NB for class IIa device (SaMD)?

Thank you for any info!

Thank you, Roger for your additional insights.  When I pressed the Commission about this early on in the MDR's tenure, the Directorate-General (DG) for Health and Food Safety maintained that the potential contradiction ("Section 4.4 to 4.8" vs. "Section 4) was corrected by Corrigendum, OJ L 117, 3.5.2019, p. 9 (2017/745).  The DG maintained that now, the text of Annex IX, Section 2.3, third subparagraph [and Section 3.5], is "Moreover, in the case of class IIa and class IIb devices, the quality management system assessment shall be accompanied by the assessment of technical documentation for devices selected on a representative basis as specified in Section 4", therefore including the assessment of the entire technical documentation.

I find it interesting that when I pressed the DG on this, the DG still maintained the MDR's very clear and unambiguous requirement for Class IIa device Annex IX Section 4 TD review and TD certification.

Accordingly, I'm not sure this is an issue of "interpretation", since the MDR is so crystal clear and unambiguous about the aforesaid two-certificate approach when the Class IIa Annex IX route is chosen.  Instead, I'd say that if we claim a one-certificate approach, then it is instead a decision to ignore and/or disagree with this very unambiguous MDR requirement that the DG maintained when I pressed.

Maybe there's merit, maybe there's not, for arguing against what the MDR so clearly says in this case.  To know for sure whether there is merit for such opposition, then we need to make a case using something other than claiming an issue of interpretation.  Specifically, we need to make a case explaining why Class IIa devices employing the Annex IX route (traditionally called the "full quality assurance" route) shouldn't be subject to the very basic TD controls of Section 4.  In so doing, we would need to acknowledge that the "design dossier" emphasis of Section 4 from the MDD has been generalized into a more generic/holistic TD assessment rather than just a design dossier review.

I think we need to be sensitive to the possibility that the Commission did this intentionally in order to prescribe additional substance to the assessment of a Class IIa full QMS TD assessment.  Indeed, that same additional substance principle existed in the MDD at Annex II.7, though there it excluded Section 4 because Section 4 was indeed at that time reserved for the design dossier approach.  But as mentioned, that has now been generalized into a more holistic TD assessment providing for either the traditional design dossier level of review, but also to allow use for a more substantive TD assessment for devices (i.e., Class IIa full QMS devices) where some additional substance is of value for public health.  The MDR does after all intend to increase assurance of safety and performance relative to the MDD.

And I don't' think we can make a valid argument against a regulation's multi-purposing of existing legislative infrastructure like when multi-purposing Section 4 for Class IIa devices.  Regulations routinely do that kind of thing in order to avoid replicating/duplicating language in multiple places.  Indeed, incorporation by reference is a commonplace tactic used when writing regulations.

Moreover, the basic title of Annex IX is "Conformity Assessment Based on Quality Management System and On Assessment of Technical Documentation", not, "or Assessment of Technical Documentation".  Accordingly, while we may argue against the inconvenience of issuing two certificates, I'm not yet seeing grounds for arguing against Section 4's additional parameters for TD assessment of Class IIa devices undergoing the full QMS route.

Ultimately again, if we aim to ignore and/or disagree with the use of the Section 4 substance for a Class IIa device full QMS TD assessment, then we need to make a case for why that additional substance is unnecessary.  I haven't heard a good argument for that yet, nor can I think of one myself so far.

------------------------------
Kevin Randall, ASQ CQA, RAC (Europe, U.S., Canada)
Principal Consultant
Ridgway, CO
United States
© Copyright by ComplianceAcuity, Inc. All rights reserved.

Original Message:
Sent: 26-Jan-2024 08:05
From: Roger Gray
Subject: CE design certificate

I know we've gone some way from answering the initial question, but I thought I would just make a couple of comments.

I agree with Christoph's interpretation of both MDD and MDR, in that you generally only need a QMS certificate (MDD Annex II/MDR Annex IX Parts I and III) unless you have a Class III device, or there is a special category of Class IIb device involved.

It seems to me that the current confusion is caused by an ambiguity in the MDR (what a surprise!). If you read MDR Annex IX para 3.5, it says:

"In the case of class IIa and class IIb devices, the surveillance assessment shall also include an assessment of the technical documentation as specified in Section 4 for the device or devices concerned on the basis of further representative samples chosen in accordance with the rationale documented by the notified body in accordance with the third paragraph of Section 2.3."

The only problem is that when you go to Section 4, it is apparent that this entire section applies only to Class III and certain Class IIb devices, as it states in the very first line:

"Assessment of the technical documentation applicable to class III devices and to the class IIb devices referred to in the second subparagraph of Article 52(4)."

So, unfortunately, the referral to Annex IX Section 4 in Section 3.5 does not make sense, as Section 4 apparently applies ONLY to Class III and certain Class IIb devices, meaning that it has no application to Class IIa devices. There is literally zero mention of 'Class IIa' anywhere in Section 4, so I am very confused by DQS issuing an 'Annex IX Chapter II' certificate for a Class IIa device, citing "Conformity assessment based on a Quality Management System and an Assessment of Technical Documentation". 

My interpretation of the MDR is that for a Class IIa device "Conformity assessment based on a Quality Management System and an Assessment of Technical Documentation". would require either:

• Annex IX Parts I and III
• Annex XI Section 10

Note that while the use of Annex IX Parts I and III would result in just one 'Quality Management System' certificate from the notified body, using Annex XI Section 10 would result in two certificates from the notified body, these being a 'Quality Assurance Certificate' plus an 'Annex XI Part A Technical Documentation Certificate'.

Just for fun, I went back to the final draft of the MDR pre-publication, copy dated March 2017, and this is what Annex IX Section 3.5 stated at that time:

"Moreover, in the case of class IIa and class IIb devices, the quality management system assessment shall be accompanied by the assessment of technical documentation for devices selected on a representative basis in accordance with Sections 4.4 to 4.8."

If this wording had been maintained, then the reference to Section 4 would have made sense, because if you just apply Sections 4.4 to 4.8 to Class IIa devices, you completely avoid the possible need for a Technical Documentation Certificate, because this is mentioned only in Section 4.9.

So not for the first time, in moving from an earlier version to a later version, the authoring team have created an ambiguity where one didn't previously exist, thus causing us all these interpretational headaches.

------------------------------
Roger Gray
VP Quality and Regulatory
Donawa Lifescience
Clinical Studies - Regulatory - Quality Systems
AR/UKRP services
Italy and UK
rgray@donawa.com
www.donawa.com

Original Message:
Sent: 25-Jan-2024 01:56
From: Christoph Kiesselbach
Subject: CE design certificate

Hi Kevin,

yes, I think I got a little bit carried away by the topic. Thank you for discussing this, it definitely brought up some points I would like to try to clarify.

I will do some research if there is some kind of consensus between NBs or an MDCG opinion on this and see if it makes sense to post the results in a separate thread.

Best regards, Christoph

------------------------------
Christoph Kiesselbach
Schrack & Partner
Reutlingen
Germany

Original Message:
Sent: 24-Jan-2024 11:47
From: Kevin Randall
Subject: CE design certificate

Thank you, Christoph. NBOG-3 (addressing certificate practices for the outgoing MDD) doesn't directly foresee a two-certificate approach yet does so indirectly in my experience.  Specifically, if the manufacturer, like most do, has chosen to employ an ISO standard as its means for meeting the Annex II Full Quality Assurance QMS requirements, then there is required to be a separate QMS certificate that is the basis for the Annex II Full Quality Assurance EC Certificate.

Also of note is NB-MED/2.5.1/Rec4 stating that the matter of interlinking of certificates has been considered and is recognized to be an administrative responsibility of the NB and should be traceable through the file of the NB. Inclusion of a cross-reference on the certificates themselves, however, would lead to practical difficulties.  Thus, the lack of such cross-references (like in the MDD certificates your cited) certainly doesn't necessarily mean that the certificate stands on its own.

Regarding the various additional questions you've raised (i.e., handling of significant changes for class IIa devices; NB practices for TD sampling after significant changes), those seem to be well beyond the scope of Anon's question.  I certainly have answers for you on those peripheral questions; but I think it's most appropriate to raise them in a separate threat so that this current thread can stay on track with Anon's intended question.

------------------------------
Kevin Randall, ASQ CQA, RAC (Europe, U.S., Canada)
Principal Consultant
Ridgway, CO
United States
© Copyright by ComplianceAcuity, Inc. All rights reserved.

Original Message:
Sent: 24-Jan-2024 03:48
From: Christoph Kiesselbach
Subject: CE design certificate

Hi Kevin,

thank you for providing the examples, that was definitely something I had not realised before. Just for others reading this here are two examples for publicly available current MDR certificates by the mentioned TÜV Süd and BSI that either declare that TD assessment is part of this certificate or that an Annex IX Chapter II certificate is only required for class III and IIb implantable devices (where every TD must be assessed).

With respect to the MDD: At least the old NBOG Best Practice Guide (2010-3) does not foresee a combination of Annex II excluding (4) with Annex II (4) ( for class IIa and IIb, but exclusively for class III. And in the corresponding MDD certificates to the examples you provided DQS used the singe-certificate approach under MDD (see here and here, where an additional Annex II section 4 certificate is limited to class III devices).

Obviously there is some freedom in the interpretaton of the MDR and the corresponding use of certificates. However I would really prefer if there was a common approach by the NBs. I will try to reach out to Team NB and see if they could imagine a common approach for their members.

But aside from the number of certificates, the main impact if Annex IX Chapter II is fully applicable to class IIa and IIb non-implantable devices (aside from the sampling aspect) in my opinion would be that this would also mean that significant design changes must be approved by the Notified Body and lead to a supplement to the (then existing) TD-certificate (of course also under the single-certificate approach each NB might have additional requirements on what needs to be reported by the manufacturer that might be more extensive, but I still think this would be different from a legal requirement with mandatory consequences for the certificates). At least in my experience that was handled differently under MDD for Annex II excluding (4) (at least if you go with the single-certificate approach and the conformity assessment routes given in NBOG 2010-3, see NBOG 2014-3).

There currently is no MDCG guidance on the interpretation of the need for approval of significant changes (with the exception of the specific case of Article 120), so their position is currently not clear to me. But how would this be handled by the NB in the case of TD sampling? If the manufacturer makes a significant change to a device within a group with a TD that has not yet been assessed as a sample, how does the NB assess the change? Based on the change notification only without any knowledge about the previous state of the TD? Do they perform an additional full TD assessment for this device? Or is the assessment only necessary if the TD already had been sampled? Maybe I am wrong but I think this would require a fair amount of additional resources and time, both currently in rather short supply.

Best regards, Christoph

------------------------------
Christoph Kiesselbach
Schrack & Partner
Reutlingen
Germany

Original Message:
Sent: 23-Jan-2024 17:00
From: Kevin Randall
Subject: CE design certificate

Christoph,

I can't show actual clients' examples; but I'll be creative to show a couple Class IIa examples that are publicly/readily available among others on the internet:

• Example 1
• Example 2

Therein on these Class IIa device Annex IX Chapter II EU Technical Documentation Assessment Certificates is stated, "For placing devices listed in the Annex on the market, an additional certificate according to Annex IX, Chapter I and III is required."

Along these lines during my 28-year tenure with medical device CE marking, I've seen many instances of the MDD Article 11(2) second paragraph Class IIa Annex II full quality assurance route generating a QMS certificate plus an EC certificate for the chosen assessment route through the most well-known NBs.

All this said, your insights have educated me to the fact that there does seem to be more NBs employing the single-certificate route for Class IIa devices. For example, it hadn't yet dawned on me that BSI and TUV seem to be sticking to that single-certificate approach for Class IIa Annex IX devices.

My understanding is that the NBs have freedom regarding the particular documented format/way in which they answer the aforesaid two separate, distinct, concomitant Class IIa Annex IX certificate requirements (QMS certificate + TD certificate).  Thus, this liberty may be why we are indeed seeing some use a single-certificate Class IIa Annex IX approach where that one certificate is in fact entitled "EU Quality Management System Assessment Certificate".  But I don't believe that is in clear alignment with the aforesaid two-certificate requirement.  Given how much trouble NBs have had in recent years issuing timely certificates, perhaps they are stretching this issue to reduce the administrative burden rather than following the approach in the examples above which seems to more clearly match the intent of the MDR.

------------------------------
Kevin Randall, ASQ CQA, RAC (Europe, U.S., Canada)
Principal Consultant
Ridgway, CO
United States
© Copyright by ComplianceAcuity, Inc. All rights reserved.

Original Message:
Sent: 23-Jan-2024 12:22
From: Christoph Kiesselbach
Subject: CE design certificate

Hi Kevin,

I respectfully disagree.

Under the MDD e.g. Annex II section 7 made it quite clear that for class IIa and IIb devices section 4 of Annex II (that contains provisions for design examination and a corresponding certificate) does not apply and instead TDs are sampled as part of the audit procedure (Section 3.3 of Annex II). The certificates for the quality management system (Annex II without section 4) that I know then stated that the validity for class III devices would depend on an additional design examination certificate (Annex II section 4) but I know of no additional certificate for class IIa/IIb devices where the TDs were sampled.

The MDR (of course) makes things less clear, but I think a not unusual interpretation of the additional TD assessment of samples in the case of IIa and IIb devices for e.g. Annex IX is that the assessment again is done as part of the audit activities (Chapter I of Annex IX, where also the additional TD sampling during surveillance in described) and while it is generally performed as described in Section 4 of Annex IX, the results are a part of the Annex IX chapter I (and III) certificate. Only class III and class IIb implantable devices where every TD is assessed receive an additional certificate in line with Section 4.9 and are subject to the approval of changes detailed in Section 4.10. I also think this makes sense: Only if every TD is assessed the reporting and approval of changes makes sense. If only samples are assessed, then what would the "approved device" in section 4.10 refer to? The whole generic device group / category of devices that the sample stands for or only the TDs that happen to have been sampled? If the NB does not know the TD the manufacturer is changing (because it has not yet been sampled), how are they assumed to assess changes? Would they have to do an additional full TD assessment?

And so far the MDR certificates I have seen are comparable to the ones under MDD, a certificate according to Annex IX Chapter I (and III) with the additional remark that in case of class III and IIb implantable devices an additional Annex IX Section 4 (or Chapter II) certificate is required (but not for class IIa/IIb).

However, I admit this is less clearly expressed than it was under the MDD and there might be NBs that interpret this differently and hand out Annex IX Chapter II certificates for class IIa/IIb (non implantable) devices. If you have examples for this, I would be interested to know them.

Best regards, Christoph

------------------------------
Christoph Kiesselbach
Schrack & Partner
Reutlingen
Germany

Original Message:
Sent: 23-Jan-2024 09:27
From: Kevin Randall
Subject: CE design certificate

Under both the outgoing Directives (e.g., MDD 93/42/EEC) and the new regulations (e.g., MDR 2017/745), whenever a chosen conformity assessment route requires Notified Body (NB) assessment of the quality management system (QMS), it is always required to be paired with NB assessment of the Technical Documentation (TD).  This always results in two certificates issued in order to gain Notified Body (NB) approval for placing the corresponding devices on the market: One certificate for the quality management system conformity, and one certificate for the device conformity. This two-certificate ensemble is a most fundamental certification requirement that is core to any successful and real CE marking experience where the manufacturer chooses a route involving QMS conformity.

For example, if the Article 52(6) first paragraph Class IIa conformity assessment route is chosen, then, by definition, that route involves two types of conformity assessments and certificates by the NB: One for the QMS plus one for the device TD. This is the way it's always been under the outgoing Directives (e.g., MDD 93/42/EEC), and this has been maintained under the new regulations (e.g., MDR 2017/745).  Specifically, for example under the MDR per Article 52(6) first paragraph, the Annex IX Chapters I & III QMS conformity assessment is performed by the NB resulting in an EU QMS assessment certificate per Annex IX section 2.3 fourth paragraph PLUS an Annex IX Chapter II section 4 TD assessment is performed by the NB resulting in an EU TD assessment certificate per Annex IX section 4.9.

As further examples, the same type of two-certificate requirement applies to the Article 52(6) second paragraph Class IIa conformity assessment route (and others for other device classes).

------------------------------
Kevin Randall, ASQ CQA, RAC (Europe, U.S., Canada)
Principal Consultant
Ridgway, CO
United States
© Copyright by ComplianceAcuity, Inc. All rights reserved.

Original Message:
Sent: 23-Jan-2024 04:11
From: Christoph Kiesselbach
Subject: CE design certificate

Hi Anon,

as Kevin says, the MDR does not use the expression "design certificate".

Annex XII that deals with the certificates issued by a notified body lists: "EU technical documentation assessment certificates, EU type-examination certificates and EU product verification certificates" in section 4(a) and "EU quality management system certificates and EU quality assurance certificates" in section 4(b). You could argue that the former are to some degree "device certificates", because they refer to an individual technical documentation, device type or even single devices, while the latter are "quality management certificates", but there is no such formulation in the MDR.

For a class IIa SaMD you usually would need either a "EU quality management system certificate" (Annex IX, chapters I and III) or a "EU quality assurence certificate" (Annex XI, Part A, Section 10). Both procedures would include the assessment of a technical documentation for at least one representative device sample for each device category, but as far as I know this assessment does not result in a separate certificate (as opposed to e.g. the full Annex IX assessment for class III devices or a type examination according to Annex X). If you only have one device, then of course the sampling in the end means that every device is assessed this way, but again this to my knowledge is not reflected in its own certificate. In theory Annex XI, Part B, Section 18, would be a possible procedure, but testing each device does not make sense for software and I do not think any notified body has this in their scope.

So in a loose sense, you do not obtain a design certificate for a class IIa software, but a quality management certificate (either full quality management system or product assurance). However, using the language of the MDR in describing the certificate is the better idea.

Best regards, Christoph

------------------------------
Christoph Kiesselbach
Schrack & Partner
Reutlingen
Germany

Original Message:
Sent: 21-Jan-2024 07:29
From: Anonymous Member
Subject: CE design certificate

This message was posted by a user wishing to remain anonymous

Dear RAPS Community,

What is CE design certificate and it is mandatory to obtain from NB for class IIa device (SaMD)?

Thank you for any info!

Thanks for the additional information regarding your interaction with the Commission, Kevin, and for reminder that the change to Annex IX Section 3.5 was made in Corrigendum 1..

Unfortunately, the Commission has made the classic error of making one change without understanding the consequences elsewhere in the document. Even with this revision to Annex IX Section 3.5, Section 4 still introduces itself by stating that it applies only to Class III and certain Class II devices, when it no longer does. Clearly, this should have been amended at the same time as the change to Section 3.5. Only then would the current wording be "crystal clear and unambiguous". What is the point of having a section stating that it applies only to Class III and certain Class IIb devices, when it now apparently applies equally to Class IIa and all Class IIb?

It seems to me that the original intention of the MDR was to maintain the single QMS certificate approach of the MDD for Class IIa devices, but since initial publication, it was decided to make a very significant change, but then not implement it properly.

Well it was always said that CE stood for Confusion Everywhere!

Thank you, Roger for your additional insights.  When I pressed the Commission about this early on in the MDR's tenure, the Directorate-General (DG) for Health and Food Safety maintained that the potential contradiction ("Section 4.4 to 4.8" vs. "Section 4) was corrected by Corrigendum, OJ L 117, 3.5.2019, p. 9 (2017/745).  The DG maintained that now, the text of Annex IX, Section 2.3, third subparagraph [and Section 3.5], is "Moreover, in the case of class IIa and class IIb devices, the quality management system assessment shall be accompanied by the assessment of technical documentation for devices selected on a representative basis as specified in Section 4", therefore including the assessment of the entire technical documentation.

I find it interesting that when I pressed the DG on this, the DG still maintained the MDR's very clear and unambiguous requirement for Class IIa device Annex IX Section 4 TD review and TD certification.

Accordingly, I'm not sure this is an issue of "interpretation", since the MDR is so crystal clear and unambiguous about the aforesaid two-certificate approach when the Class IIa Annex IX route is chosen.  Instead, I'd say that if we claim a one-certificate approach, then it is instead a decision to ignore and/or disagree with this very unambiguous MDR requirement that the DG maintained when I pressed.

Maybe there's merit, maybe there's not, for arguing against what the MDR so clearly says in this case.  To know for sure whether there is merit for such opposition, then we need to make a case using something other than claiming an issue of interpretation.  Specifically, we need to make a case explaining why Class IIa devices employing the Annex IX route (traditionally called the "full quality assurance" route) shouldn't be subject to the very basic TD controls of Section 4.  In so doing, we would need to acknowledge that the "design dossier" emphasis of Section 4 from the MDD has been generalized into a more generic/holistic TD assessment rather than just a design dossier review.

I think we need to be sensitive to the possibility that the Commission did this intentionally in order to prescribe additional substance to the assessment of a Class IIa full QMS TD assessment.  Indeed, that same additional substance principle existed in the MDD at Annex II.7, though there it excluded Section 4 because Section 4 was indeed at that time reserved for the design dossier approach.  But as mentioned, that has now been generalized into a more holistic TD assessment providing for either the traditional design dossier level of review, but also to allow use for a more substantive TD assessment for devices (i.e., Class IIa full QMS devices) where some additional substance is of value for public health.  The MDR does after all intend to increase assurance of safety and performance relative to the MDD.

And I don't' think we can make a valid argument against a regulation's multi-purposing of existing legislative infrastructure like when multi-purposing Section 4 for Class IIa devices.  Regulations routinely do that kind of thing in order to avoid replicating/duplicating language in multiple places.  Indeed, incorporation by reference is a commonplace tactic used when writing regulations.

Moreover, the basic title of Annex IX is "Conformity Assessment Based on Quality Management System and On Assessment of Technical Documentation", not, "or Assessment of Technical Documentation".  Accordingly, while we may argue against the inconvenience of issuing two certificates, I'm not yet seeing grounds for arguing against Section 4's additional parameters for TD assessment of Class IIa devices undergoing the full QMS route.

Ultimately again, if we aim to ignore and/or disagree with the use of the Section 4 substance for a Class IIa device full QMS TD assessment, then we need to make a case for why that additional substance is unnecessary.  I haven't heard a good argument for that yet, nor can I think of one myself so far.

------------------------------
Kevin Randall, ASQ CQA, RAC (Europe, U.S., Canada)
Principal Consultant
Ridgway, CO
United States
© Copyright by ComplianceAcuity, Inc. All rights reserved.

Original Message:
Sent: 26-Jan-2024 08:05
From: Roger Gray
Subject: CE design certificate

I know we've gone some way from answering the initial question, but I thought I would just make a couple of comments.

I agree with Christoph's interpretation of both MDD and MDR, in that you generally only need a QMS certificate (MDD Annex II/MDR Annex IX Parts I and III) unless you have a Class III device, or there is a special category of Class IIb device involved.

It seems to me that the current confusion is caused by an ambiguity in the MDR (what a surprise!). If you read MDR Annex IX para 3.5, it says:

"In the case of class IIa and class IIb devices, the surveillance assessment shall also include an assessment of the technical documentation as specified in Section 4 for the device or devices concerned on the basis of further representative samples chosen in accordance with the rationale documented by the notified body in accordance with the third paragraph of Section 2.3."

The only problem is that when you go to Section 4, it is apparent that this entire section applies only to Class III and certain Class IIb devices, as it states in the very first line:

"Assessment of the technical documentation applicable to class III devices and to the class IIb devices referred to in the second subparagraph of Article 52(4)."

So, unfortunately, the referral to Annex IX Section 4 in Section 3.5 does not make sense, as Section 4 apparently applies ONLY to Class III and certain Class IIb devices, meaning that it has no application to Class IIa devices. There is literally zero mention of 'Class IIa' anywhere in Section 4, so I am very confused by DQS issuing an 'Annex IX Chapter II' certificate for a Class IIa device, citing "Conformity assessment based on a Quality Management System and an Assessment of Technical Documentation". 

My interpretation of the MDR is that for a Class IIa device "Conformity assessment based on a Quality Management System and an Assessment of Technical Documentation". would require either:

• Annex IX Parts I and III
• Annex XI Section 10

Note that while the use of Annex IX Parts I and III would result in just one 'Quality Management System' certificate from the notified body, using Annex XI Section 10 would result in two certificates from the notified body, these being a 'Quality Assurance Certificate' plus an 'Annex XI Part A Technical Documentation Certificate'.

Just for fun, I went back to the final draft of the MDR pre-publication, copy dated March 2017, and this is what Annex IX Section 3.5 stated at that time:

"Moreover, in the case of class IIa and class IIb devices, the quality management system assessment shall be accompanied by the assessment of technical documentation for devices selected on a representative basis in accordance with Sections 4.4 to 4.8."

If this wording had been maintained, then the reference to Section 4 would have made sense, because if you just apply Sections 4.4 to 4.8 to Class IIa devices, you completely avoid the possible need for a Technical Documentation Certificate, because this is mentioned only in Section 4.9.

So not for the first time, in moving from an earlier version to a later version, the authoring team have created an ambiguity where one didn't previously exist, thus causing us all these interpretational headaches.

------------------------------
Roger Gray
VP Quality and Regulatory
Donawa Lifescience
Clinical Studies - Regulatory - Quality Systems
AR/UKRP services
Italy and UK
rgray@donawa.com
www.donawa.com

Original Message:
Sent: 25-Jan-2024 01:56
From: Christoph Kiesselbach
Subject: CE design certificate

Hi Kevin,

yes, I think I got a little bit carried away by the topic. Thank you for discussing this, it definitely brought up some points I would like to try to clarify.

I will do some research if there is some kind of consensus between NBs or an MDCG opinion on this and see if it makes sense to post the results in a separate thread.

Best regards, Christoph

------------------------------
Christoph Kiesselbach
Schrack & Partner
Reutlingen
Germany

Original Message:
Sent: 24-Jan-2024 11:47
From: Kevin Randall
Subject: CE design certificate

Thank you, Christoph. NBOG-3 (addressing certificate practices for the outgoing MDD) doesn't directly foresee a two-certificate approach yet does so indirectly in my experience.  Specifically, if the manufacturer, like most do, has chosen to employ an ISO standard as its means for meeting the Annex II Full Quality Assurance QMS requirements, then there is required to be a separate QMS certificate that is the basis for the Annex II Full Quality Assurance EC Certificate.

Also of note is NB-MED/2.5.1/Rec4 stating that the matter of interlinking of certificates has been considered and is recognized to be an administrative responsibility of the NB and should be traceable through the file of the NB. Inclusion of a cross-reference on the certificates themselves, however, would lead to practical difficulties.  Thus, the lack of such cross-references (like in the MDD certificates your cited) certainly doesn't necessarily mean that the certificate stands on its own.

Regarding the various additional questions you've raised (i.e., handling of significant changes for class IIa devices; NB practices for TD sampling after significant changes), those seem to be well beyond the scope of Anon's question.  I certainly have answers for you on those peripheral questions; but I think it's most appropriate to raise them in a separate threat so that this current thread can stay on track with Anon's intended question.

------------------------------
Kevin Randall, ASQ CQA, RAC (Europe, U.S., Canada)
Principal Consultant
Ridgway, CO
United States
© Copyright by ComplianceAcuity, Inc. All rights reserved.

Original Message:
Sent: 24-Jan-2024 03:48
From: Christoph Kiesselbach
Subject: CE design certificate

Hi Kevin,

thank you for providing the examples, that was definitely something I had not realised before. Just for others reading this here are two examples for publicly available current MDR certificates by the mentioned TÜV Süd and BSI that either declare that TD assessment is part of this certificate or that an Annex IX Chapter II certificate is only required for class III and IIb implantable devices (where every TD must be assessed).

With respect to the MDD: At least the old NBOG Best Practice Guide (2010-3) does not foresee a combination of Annex II excluding (4) with Annex II (4) ( for class IIa and IIb, but exclusively for class III. And in the corresponding MDD certificates to the examples you provided DQS used the singe-certificate approach under MDD (see here and here, where an additional Annex II section 4 certificate is limited to class III devices).

Obviously there is some freedom in the interpretaton of the MDR and the corresponding use of certificates. However I would really prefer if there was a common approach by the NBs. I will try to reach out to Team NB and see if they could imagine a common approach for their members.

But aside from the number of certificates, the main impact if Annex IX Chapter II is fully applicable to class IIa and IIb non-implantable devices (aside from the sampling aspect) in my opinion would be that this would also mean that significant design changes must be approved by the Notified Body and lead to a supplement to the (then existing) TD-certificate (of course also under the single-certificate approach each NB might have additional requirements on what needs to be reported by the manufacturer that might be more extensive, but I still think this would be different from a legal requirement with mandatory consequences for the certificates). At least in my experience that was handled differently under MDD for Annex II excluding (4) (at least if you go with the single-certificate approach and the conformity assessment routes given in NBOG 2010-3, see NBOG 2014-3).

There currently is no MDCG guidance on the interpretation of the need for approval of significant changes (with the exception of the specific case of Article 120), so their position is currently not clear to me. But how would this be handled by the NB in the case of TD sampling? If the manufacturer makes a significant change to a device within a group with a TD that has not yet been assessed as a sample, how does the NB assess the change? Based on the change notification only without any knowledge about the previous state of the TD? Do they perform an additional full TD assessment for this device? Or is the assessment only necessary if the TD already had been sampled? Maybe I am wrong but I think this would require a fair amount of additional resources and time, both currently in rather short supply.

Best regards, Christoph

------------------------------
Christoph Kiesselbach
Schrack & Partner
Reutlingen
Germany

Original Message:
Sent: 23-Jan-2024 17:00
From: Kevin Randall
Subject: CE design certificate

Christoph,

I can't show actual clients' examples; but I'll be creative to show a couple Class IIa examples that are publicly/readily available among others on the internet:

• Example 1
• Example 2

Therein on these Class IIa device Annex IX Chapter II EU Technical Documentation Assessment Certificates is stated, "For placing devices listed in the Annex on the market, an additional certificate according to Annex IX, Chapter I and III is required."

Along these lines during my 28-year tenure with medical device CE marking, I've seen many instances of the MDD Article 11(2) second paragraph Class IIa Annex II full quality assurance route generating a QMS certificate plus an EC certificate for the chosen assessment route through the most well-known NBs.

All this said, your insights have educated me to the fact that there does seem to be more NBs employing the single-certificate route for Class IIa devices. For example, it hadn't yet dawned on me that BSI and TUV seem to be sticking to that single-certificate approach for Class IIa Annex IX devices.

My understanding is that the NBs have freedom regarding the particular documented format/way in which they answer the aforesaid two separate, distinct, concomitant Class IIa Annex IX certificate requirements (QMS certificate + TD certificate).  Thus, this liberty may be why we are indeed seeing some use a single-certificate Class IIa Annex IX approach where that one certificate is in fact entitled "EU Quality Management System Assessment Certificate".  But I don't believe that is in clear alignment with the aforesaid two-certificate requirement.  Given how much trouble NBs have had in recent years issuing timely certificates, perhaps they are stretching this issue to reduce the administrative burden rather than following the approach in the examples above which seems to more clearly match the intent of the MDR.

------------------------------
Kevin Randall, ASQ CQA, RAC (Europe, U.S., Canada)
Principal Consultant
Ridgway, CO
United States
© Copyright by ComplianceAcuity, Inc. All rights reserved.

Original Message:
Sent: 23-Jan-2024 12:22
From: Christoph Kiesselbach
Subject: CE design certificate

Hi Kevin,

I respectfully disagree.

Under the MDD e.g. Annex II section 7 made it quite clear that for class IIa and IIb devices section 4 of Annex II (that contains provisions for design examination and a corresponding certificate) does not apply and instead TDs are sampled as part of the audit procedure (Section 3.3 of Annex II). The certificates for the quality management system (Annex II without section 4) that I know then stated that the validity for class III devices would depend on an additional design examination certificate (Annex II section 4) but I know of no additional certificate for class IIa/IIb devices where the TDs were sampled.

The MDR (of course) makes things less clear, but I think a not unusual interpretation of the additional TD assessment of samples in the case of IIa and IIb devices for e.g. Annex IX is that the assessment again is done as part of the audit activities (Chapter I of Annex IX, where also the additional TD sampling during surveillance in described) and while it is generally performed as described in Section 4 of Annex IX, the results are a part of the Annex IX chapter I (and III) certificate. Only class III and class IIb implantable devices where every TD is assessed receive an additional certificate in line with Section 4.9 and are subject to the approval of changes detailed in Section 4.10. I also think this makes sense: Only if every TD is assessed the reporting and approval of changes makes sense. If only samples are assessed, then what would the "approved device" in section 4.10 refer to? The whole generic device group / category of devices that the sample stands for or only the TDs that happen to have been sampled? If the NB does not know the TD the manufacturer is changing (because it has not yet been sampled), how are they assumed to assess changes? Would they have to do an additional full TD assessment?

And so far the MDR certificates I have seen are comparable to the ones under MDD, a certificate according to Annex IX Chapter I (and III) with the additional remark that in case of class III and IIb implantable devices an additional Annex IX Section 4 (or Chapter II) certificate is required (but not for class IIa/IIb).

However, I admit this is less clearly expressed than it was under the MDD and there might be NBs that interpret this differently and hand out Annex IX Chapter II certificates for class IIa/IIb (non implantable) devices. If you have examples for this, I would be interested to know them.

Best regards, Christoph

------------------------------
Christoph Kiesselbach
Schrack & Partner
Reutlingen
Germany

Original Message:
Sent: 23-Jan-2024 09:27
From: Kevin Randall
Subject: CE design certificate

Under both the outgoing Directives (e.g., MDD 93/42/EEC) and the new regulations (e.g., MDR 2017/745), whenever a chosen conformity assessment route requires Notified Body (NB) assessment of the quality management system (QMS), it is always required to be paired with NB assessment of the Technical Documentation (TD).  This always results in two certificates issued in order to gain Notified Body (NB) approval for placing the corresponding devices on the market: One certificate for the quality management system conformity, and one certificate for the device conformity. This two-certificate ensemble is a most fundamental certification requirement that is core to any successful and real CE marking experience where the manufacturer chooses a route involving QMS conformity.

For example, if the Article 52(6) first paragraph Class IIa conformity assessment route is chosen, then, by definition, that route involves two types of conformity assessments and certificates by the NB: One for the QMS plus one for the device TD. This is the way it's always been under the outgoing Directives (e.g., MDD 93/42/EEC), and this has been maintained under the new regulations (e.g., MDR 2017/745).  Specifically, for example under the MDR per Article 52(6) first paragraph, the Annex IX Chapters I & III QMS conformity assessment is performed by the NB resulting in an EU QMS assessment certificate per Annex IX section 2.3 fourth paragraph PLUS an Annex IX Chapter II section 4 TD assessment is performed by the NB resulting in an EU TD assessment certificate per Annex IX section 4.9.

As further examples, the same type of two-certificate requirement applies to the Article 52(6) second paragraph Class IIa conformity assessment route (and others for other device classes).

------------------------------
Kevin Randall, ASQ CQA, RAC (Europe, U.S., Canada)
Principal Consultant
Ridgway, CO
United States
© Copyright by ComplianceAcuity, Inc. All rights reserved.

Original Message:
Sent: 23-Jan-2024 04:11
From: Christoph Kiesselbach
Subject: CE design certificate

Hi Anon,

as Kevin says, the MDR does not use the expression "design certificate".

Annex XII that deals with the certificates issued by a notified body lists: "EU technical documentation assessment certificates, EU type-examination certificates and EU product verification certificates" in section 4(a) and "EU quality management system certificates and EU quality assurance certificates" in section 4(b). You could argue that the former are to some degree "device certificates", because they refer to an individual technical documentation, device type or even single devices, while the latter are "quality management certificates", but there is no such formulation in the MDR.

For a class IIa SaMD you usually would need either a "EU quality management system certificate" (Annex IX, chapters I and III) or a "EU quality assurence certificate" (Annex XI, Part A, Section 10). Both procedures would include the assessment of a technical documentation for at least one representative device sample for each device category, but as far as I know this assessment does not result in a separate certificate (as opposed to e.g. the full Annex IX assessment for class III devices or a type examination according to Annex X). If you only have one device, then of course the sampling in the end means that every device is assessed this way, but again this to my knowledge is not reflected in its own certificate. In theory Annex XI, Part B, Section 18, would be a possible procedure, but testing each device does not make sense for software and I do not think any notified body has this in their scope.

So in a loose sense, you do not obtain a design certificate for a class IIa software, but a quality management certificate (either full quality management system or product assurance). However, using the language of the MDR in describing the certificate is the better idea.

Best regards, Christoph

------------------------------
Christoph Kiesselbach
Schrack & Partner
Reutlingen
Germany

Original Message:
Sent: 21-Jan-2024 07:29
From: Anonymous Member
Subject: CE design certificate

This message was posted by a user wishing to remain anonymous

Dear RAPS Community,

What is CE design certificate and it is mandatory to obtain from NB for class IIa device (SaMD)?

Thank you for any info!

Thank you for your comedy relief of CE standing for "Confusion Everywhere". :)

------------------------------
Kevin Randall, ASQ CQA, RAC (Europe, U.S., Canada)
Principal Consultant
Ridgway, CO
United States
© Copyright by ComplianceAcuity, Inc. All rights reserved.

Original Message:
Sent: 26-Jan-2024 12:51
From: Roger Gray
Subject: CE design certificate

Thanks for the additional information regarding your interaction with the Commission, Kevin, and for reminder that the change to Annex IX Section 3.5 was made in Corrigendum 1..

Unfortunately, the Commission has made the classic error of making one change without understanding the consequences elsewhere in the document. Even with this revision to Annex IX Section 3.5, Section 4 still introduces itself by stating that it applies only to Class III and certain Class II devices, when it no longer does. Clearly, this should have been amended at the same time as the change to Section 3.5. Only then would the current wording be "crystal clear and unambiguous". What is the point of having a section stating that it applies only to Class III and certain Class IIb devices, when it now apparently applies equally to Class IIa and all Class IIb?

It seems to me that the original intention of the MDR was to maintain the single QMS certificate approach of the MDD for Class IIa devices, but since initial publication, it was decided to make a very significant change, but then not implement it properly.

Well it was always said that CE stood for Confusion Everywhere!

------------------------------
Roger Gray
VP Quality and Regulatory
Donawa Lifescience
Clinical Studies - Regulatory - Quality Systems
AR/UKRP services
Italy and UK
rgray@donawa.com
www.donawa.com

Original Message:
Sent: 26-Jan-2024 11:19
From: Kevin Randall
Subject: CE design certificate

Thank you, Roger for your additional insights.  When I pressed the Commission about this early on in the MDR's tenure, the Directorate-General (DG) for Health and Food Safety maintained that the potential contradiction ("Section 4.4 to 4.8" vs. "Section 4) was corrected by Corrigendum, OJ L 117, 3.5.2019, p. 9 (2017/745).  The DG maintained that now, the text of Annex IX, Section 2.3, third subparagraph [and Section 3.5], is "Moreover, in the case of class IIa and class IIb devices, the quality management system assessment shall be accompanied by the assessment of technical documentation for devices selected on a representative basis as specified in Section 4", therefore including the assessment of the entire technical documentation.

I find it interesting that when I pressed the DG on this, the DG still maintained the MDR's very clear and unambiguous requirement for Class IIa device Annex IX Section 4 TD review and TD certification.

Accordingly, I'm not sure this is an issue of "interpretation", since the MDR is so crystal clear and unambiguous about the aforesaid two-certificate approach when the Class IIa Annex IX route is chosen.  Instead, I'd say that if we claim a one-certificate approach, then it is instead a decision to ignore and/or disagree with this very unambiguous MDR requirement that the DG maintained when I pressed.

Maybe there's merit, maybe there's not, for arguing against what the MDR so clearly says in this case.  To know for sure whether there is merit for such opposition, then we need to make a case using something other than claiming an issue of interpretation.  Specifically, we need to make a case explaining why Class IIa devices employing the Annex IX route (traditionally called the "full quality assurance" route) shouldn't be subject to the very basic TD controls of Section 4.  In so doing, we would need to acknowledge that the "design dossier" emphasis of Section 4 from the MDD has been generalized into a more generic/holistic TD assessment rather than just a design dossier review.

I think we need to be sensitive to the possibility that the Commission did this intentionally in order to prescribe additional substance to the assessment of a Class IIa full QMS TD assessment.  Indeed, that same additional substance principle existed in the MDD at Annex II.7, though there it excluded Section 4 because Section 4 was indeed at that time reserved for the design dossier approach.  But as mentioned, that has now been generalized into a more holistic TD assessment providing for either the traditional design dossier level of review, but also to allow use for a more substantive TD assessment for devices (i.e., Class IIa full QMS devices) where some additional substance is of value for public health.  The MDR does after all intend to increase assurance of safety and performance relative to the MDD.

And I don't' think we can make a valid argument against a regulation's multi-purposing of existing legislative infrastructure like when multi-purposing Section 4 for Class IIa devices.  Regulations routinely do that kind of thing in order to avoid replicating/duplicating language in multiple places.  Indeed, incorporation by reference is a commonplace tactic used when writing regulations.

Moreover, the basic title of Annex IX is "Conformity Assessment Based on Quality Management System and On Assessment of Technical Documentation", not, "or Assessment of Technical Documentation".  Accordingly, while we may argue against the inconvenience of issuing two certificates, I'm not yet seeing grounds for arguing against Section 4's additional parameters for TD assessment of Class IIa devices undergoing the full QMS route.

Ultimately again, if we aim to ignore and/or disagree with the use of the Section 4 substance for a Class IIa device full QMS TD assessment, then we need to make a case for why that additional substance is unnecessary.  I haven't heard a good argument for that yet, nor can I think of one myself so far.

------------------------------
Kevin Randall, ASQ CQA, RAC (Europe, U.S., Canada)
Principal Consultant
Ridgway, CO
United States
© Copyright by ComplianceAcuity, Inc. All rights reserved.

Original Message:
Sent: 26-Jan-2024 08:05
From: Roger Gray
Subject: CE design certificate

I know we've gone some way from answering the initial question, but I thought I would just make a couple of comments.

I agree with Christoph's interpretation of both MDD and MDR, in that you generally only need a QMS certificate (MDD Annex II/MDR Annex IX Parts I and III) unless you have a Class III device, or there is a special category of Class IIb device involved.

It seems to me that the current confusion is caused by an ambiguity in the MDR (what a surprise!). If you read MDR Annex IX para 3.5, it says:

"In the case of class IIa and class IIb devices, the surveillance assessment shall also include an assessment of the technical documentation as specified in Section 4 for the device or devices concerned on the basis of further representative samples chosen in accordance with the rationale documented by the notified body in accordance with the third paragraph of Section 2.3."

The only problem is that when you go to Section 4, it is apparent that this entire section applies only to Class III and certain Class IIb devices, as it states in the very first line:

"Assessment of the technical documentation applicable to class III devices and to the class IIb devices referred to in the second subparagraph of Article 52(4)."

So, unfortunately, the referral to Annex IX Section 4 in Section 3.5 does not make sense, as Section 4 apparently applies ONLY to Class III and certain Class IIb devices, meaning that it has no application to Class IIa devices. There is literally zero mention of 'Class IIa' anywhere in Section 4, so I am very confused by DQS issuing an 'Annex IX Chapter II' certificate for a Class IIa device, citing "Conformity assessment based on a Quality Management System and an Assessment of Technical Documentation". 

My interpretation of the MDR is that for a Class IIa device "Conformity assessment based on a Quality Management System and an Assessment of Technical Documentation". would require either:

• Annex IX Parts I and III
• Annex XI Section 10

Note that while the use of Annex IX Parts I and III would result in just one 'Quality Management System' certificate from the notified body, using Annex XI Section 10 would result in two certificates from the notified body, these being a 'Quality Assurance Certificate' plus an 'Annex XI Part A Technical Documentation Certificate'.

Just for fun, I went back to the final draft of the MDR pre-publication, copy dated March 2017, and this is what Annex IX Section 3.5 stated at that time:

"Moreover, in the case of class IIa and class IIb devices, the quality management system assessment shall be accompanied by the assessment of technical documentation for devices selected on a representative basis in accordance with Sections 4.4 to 4.8."

If this wording had been maintained, then the reference to Section 4 would have made sense, because if you just apply Sections 4.4 to 4.8 to Class IIa devices, you completely avoid the possible need for a Technical Documentation Certificate, because this is mentioned only in Section 4.9.

So not for the first time, in moving from an earlier version to a later version, the authoring team have created an ambiguity where one didn't previously exist, thus causing us all these interpretational headaches.

------------------------------
Roger Gray
VP Quality and Regulatory
Donawa Lifescience
Clinical Studies - Regulatory - Quality Systems
AR/UKRP services
Italy and UK
rgray@donawa.com
www.donawa.com

Original Message:
Sent: 25-Jan-2024 01:56
From: Christoph Kiesselbach
Subject: CE design certificate

Hi Kevin,

yes, I think I got a little bit carried away by the topic. Thank you for discussing this, it definitely brought up some points I would like to try to clarify.

I will do some research if there is some kind of consensus between NBs or an MDCG opinion on this and see if it makes sense to post the results in a separate thread.

Best regards, Christoph

------------------------------
Christoph Kiesselbach
Schrack & Partner
Reutlingen
Germany

Original Message:
Sent: 24-Jan-2024 11:47
From: Kevin Randall
Subject: CE design certificate

Thank you, Christoph. NBOG-3 (addressing certificate practices for the outgoing MDD) doesn't directly foresee a two-certificate approach yet does so indirectly in my experience.  Specifically, if the manufacturer, like most do, has chosen to employ an ISO standard as its means for meeting the Annex II Full Quality Assurance QMS requirements, then there is required to be a separate QMS certificate that is the basis for the Annex II Full Quality Assurance EC Certificate.

Also of note is NB-MED/2.5.1/Rec4 stating that the matter of interlinking of certificates has been considered and is recognized to be an administrative responsibility of the NB and should be traceable through the file of the NB. Inclusion of a cross-reference on the certificates themselves, however, would lead to practical difficulties.  Thus, the lack of such cross-references (like in the MDD certificates your cited) certainly doesn't necessarily mean that the certificate stands on its own.

Regarding the various additional questions you've raised (i.e., handling of significant changes for class IIa devices; NB practices for TD sampling after significant changes), those seem to be well beyond the scope of Anon's question.  I certainly have answers for you on those peripheral questions; but I think it's most appropriate to raise them in a separate threat so that this current thread can stay on track with Anon's intended question.

------------------------------
Kevin Randall, ASQ CQA, RAC (Europe, U.S., Canada)
Principal Consultant
Ridgway, CO
United States
© Copyright by ComplianceAcuity, Inc. All rights reserved.

Original Message:
Sent: 24-Jan-2024 03:48
From: Christoph Kiesselbach
Subject: CE design certificate

Hi Kevin,

thank you for providing the examples, that was definitely something I had not realised before. Just for others reading this here are two examples for publicly available current MDR certificates by the mentioned TÜV Süd and BSI that either declare that TD assessment is part of this certificate or that an Annex IX Chapter II certificate is only required for class III and IIb implantable devices (where every TD must be assessed).

With respect to the MDD: At least the old NBOG Best Practice Guide (2010-3) does not foresee a combination of Annex II excluding (4) with Annex II (4) ( for class IIa and IIb, but exclusively for class III. And in the corresponding MDD certificates to the examples you provided DQS used the singe-certificate approach under MDD (see here and here, where an additional Annex II section 4 certificate is limited to class III devices).

Obviously there is some freedom in the interpretaton of the MDR and the corresponding use of certificates. However I would really prefer if there was a common approach by the NBs. I will try to reach out to Team NB and see if they could imagine a common approach for their members.

But aside from the number of certificates, the main impact if Annex IX Chapter II is fully applicable to class IIa and IIb non-implantable devices (aside from the sampling aspect) in my opinion would be that this would also mean that significant design changes must be approved by the Notified Body and lead to a supplement to the (then existing) TD-certificate (of course also under the single-certificate approach each NB might have additional requirements on what needs to be reported by the manufacturer that might be more extensive, but I still think this would be different from a legal requirement with mandatory consequences for the certificates). At least in my experience that was handled differently under MDD for Annex II excluding (4) (at least if you go with the single-certificate approach and the conformity assessment routes given in NBOG 2010-3, see NBOG 2014-3).

There currently is no MDCG guidance on the interpretation of the need for approval of significant changes (with the exception of the specific case of Article 120), so their position is currently not clear to me. But how would this be handled by the NB in the case of TD sampling? If the manufacturer makes a significant change to a device within a group with a TD that has not yet been assessed as a sample, how does the NB assess the change? Based on the change notification only without any knowledge about the previous state of the TD? Do they perform an additional full TD assessment for this device? Or is the assessment only necessary if the TD already had been sampled? Maybe I am wrong but I think this would require a fair amount of additional resources and time, both currently in rather short supply.

Best regards, Christoph

------------------------------
Christoph Kiesselbach
Schrack & Partner
Reutlingen
Germany

Original Message:
Sent: 23-Jan-2024 17:00
From: Kevin Randall
Subject: CE design certificate

Christoph,

I can't show actual clients' examples; but I'll be creative to show a couple Class IIa examples that are publicly/readily available among others on the internet:

• Example 1
• Example 2

Therein on these Class IIa device Annex IX Chapter II EU Technical Documentation Assessment Certificates is stated, "For placing devices listed in the Annex on the market, an additional certificate according to Annex IX, Chapter I and III is required."

Along these lines during my 28-year tenure with medical device CE marking, I've seen many instances of the MDD Article 11(2) second paragraph Class IIa Annex II full quality assurance route generating a QMS certificate plus an EC certificate for the chosen assessment route through the most well-known NBs.

All this said, your insights have educated me to the fact that there does seem to be more NBs employing the single-certificate route for Class IIa devices. For example, it hadn't yet dawned on me that BSI and TUV seem to be sticking to that single-certificate approach for Class IIa Annex IX devices.

My understanding is that the NBs have freedom regarding the particular documented format/way in which they answer the aforesaid two separate, distinct, concomitant Class IIa Annex IX certificate requirements (QMS certificate + TD certificate).  Thus, this liberty may be why we are indeed seeing some use a single-certificate Class IIa Annex IX approach where that one certificate is in fact entitled "EU Quality Management System Assessment Certificate".  But I don't believe that is in clear alignment with the aforesaid two-certificate requirement.  Given how much trouble NBs have had in recent years issuing timely certificates, perhaps they are stretching this issue to reduce the administrative burden rather than following the approach in the examples above which seems to more clearly match the intent of the MDR.

------------------------------
Kevin Randall, ASQ CQA, RAC (Europe, U.S., Canada)
Principal Consultant
Ridgway, CO
United States
© Copyright by ComplianceAcuity, Inc. All rights reserved.

Original Message:
Sent: 23-Jan-2024 12:22
From: Christoph Kiesselbach
Subject: CE design certificate

Hi Kevin,

I respectfully disagree.

Under the MDD e.g. Annex II section 7 made it quite clear that for class IIa and IIb devices section 4 of Annex II (that contains provisions for design examination and a corresponding certificate) does not apply and instead TDs are sampled as part of the audit procedure (Section 3.3 of Annex II). The certificates for the quality management system (Annex II without section 4) that I know then stated that the validity for class III devices would depend on an additional design examination certificate (Annex II section 4) but I know of no additional certificate for class IIa/IIb devices where the TDs were sampled.

The MDR (of course) makes things less clear, but I think a not unusual interpretation of the additional TD assessment of samples in the case of IIa and IIb devices for e.g. Annex IX is that the assessment again is done as part of the audit activities (Chapter I of Annex IX, where also the additional TD sampling during surveillance in described) and while it is generally performed as described in Section 4 of Annex IX, the results are a part of the Annex IX chapter I (and III) certificate. Only class III and class IIb implantable devices where every TD is assessed receive an additional certificate in line with Section 4.9 and are subject to the approval of changes detailed in Section 4.10. I also think this makes sense: Only if every TD is assessed the reporting and approval of changes makes sense. If only samples are assessed, then what would the "approved device" in section 4.10 refer to? The whole generic device group / category of devices that the sample stands for or only the TDs that happen to have been sampled? If the NB does not know the TD the manufacturer is changing (because it has not yet been sampled), how are they assumed to assess changes? Would they have to do an additional full TD assessment?

And so far the MDR certificates I have seen are comparable to the ones under MDD, a certificate according to Annex IX Chapter I (and III) with the additional remark that in case of class III and IIb implantable devices an additional Annex IX Section 4 (or Chapter II) certificate is required (but not for class IIa/IIb).

However, I admit this is less clearly expressed than it was under the MDD and there might be NBs that interpret this differently and hand out Annex IX Chapter II certificates for class IIa/IIb (non implantable) devices. If you have examples for this, I would be interested to know them.

Best regards, Christoph

------------------------------
Christoph Kiesselbach
Schrack & Partner
Reutlingen
Germany

Original Message:
Sent: 23-Jan-2024 09:27
From: Kevin Randall
Subject: CE design certificate

Under both the outgoing Directives (e.g., MDD 93/42/EEC) and the new regulations (e.g., MDR 2017/745), whenever a chosen conformity assessment route requires Notified Body (NB) assessment of the quality management system (QMS), it is always required to be paired with NB assessment of the Technical Documentation (TD).  This always results in two certificates issued in order to gain Notified Body (NB) approval for placing the corresponding devices on the market: One certificate for the quality management system conformity, and one certificate for the device conformity. This two-certificate ensemble is a most fundamental certification requirement that is core to any successful and real CE marking experience where the manufacturer chooses a route involving QMS conformity.

For example, if the Article 52(6) first paragraph Class IIa conformity assessment route is chosen, then, by definition, that route involves two types of conformity assessments and certificates by the NB: One for the QMS plus one for the device TD. This is the way it's always been under the outgoing Directives (e.g., MDD 93/42/EEC), and this has been maintained under the new regulations (e.g., MDR 2017/745).  Specifically, for example under the MDR per Article 52(6) first paragraph, the Annex IX Chapters I & III QMS conformity assessment is performed by the NB resulting in an EU QMS assessment certificate per Annex IX section 2.3 fourth paragraph PLUS an Annex IX Chapter II section 4 TD assessment is performed by the NB resulting in an EU TD assessment certificate per Annex IX section 4.9.

As further examples, the same type of two-certificate requirement applies to the Article 52(6) second paragraph Class IIa conformity assessment route (and others for other device classes).

------------------------------
Kevin Randall, ASQ CQA, RAC (Europe, U.S., Canada)
Principal Consultant
Ridgway, CO
United States
© Copyright by ComplianceAcuity, Inc. All rights reserved.

Original Message:
Sent: 23-Jan-2024 04:11
From: Christoph Kiesselbach
Subject: CE design certificate

Hi Anon,

as Kevin says, the MDR does not use the expression "design certificate".

Annex XII that deals with the certificates issued by a notified body lists: "EU technical documentation assessment certificates, EU type-examination certificates and EU product verification certificates" in section 4(a) and "EU quality management system certificates and EU quality assurance certificates" in section 4(b). You could argue that the former are to some degree "device certificates", because they refer to an individual technical documentation, device type or even single devices, while the latter are "quality management certificates", but there is no such formulation in the MDR.

For a class IIa SaMD you usually would need either a "EU quality management system certificate" (Annex IX, chapters I and III) or a "EU quality assurence certificate" (Annex XI, Part A, Section 10). Both procedures would include the assessment of a technical documentation for at least one representative device sample for each device category, but as far as I know this assessment does not result in a separate certificate (as opposed to e.g. the full Annex IX assessment for class III devices or a type examination according to Annex X). If you only have one device, then of course the sampling in the end means that every device is assessed this way, but again this to my knowledge is not reflected in its own certificate. In theory Annex XI, Part B, Section 18, would be a possible procedure, but testing each device does not make sense for software and I do not think any notified body has this in their scope.

So in a loose sense, you do not obtain a design certificate for a class IIa software, but a quality management certificate (either full quality management system or product assurance). However, using the language of the MDR in describing the certificate is the better idea.

Best regards, Christoph

------------------------------
Christoph Kiesselbach
Schrack & Partner
Reutlingen
Germany

Original Message:
Sent: 21-Jan-2024 07:29
From: Anonymous Member
Subject: CE design certificate

This message was posted by a user wishing to remain anonymous

Dear RAPS Community,

What is CE design certificate and it is mandatory to obtain from NB for class IIa device (SaMD)?

Thank you for any info!

Thank you both for the additional input. I knew I had seen a more restricted version of that reference to section 4 and that was indeed in Annex IX 2.3/3.5 before the EU aligned it with Article 52 (4)/(6) in the first Corrigendum, choosing with uncanny accuracy the more ambiguous version.

Something else I had forgotten was that MDCG 2019-13 talks about TD sampling and with respect to applicability of Chapter II, Section 4 of Annex IX it says in section 5.2:

"Taking into account the wording of articles 52(4) and 52(6) of the MDR and Article 48(7) and 48(9) of the IVDR, as combined with annexes VII and IX, the
tasks to be carried out by the notified body as part of the conformity assessment activities described in Chapter II, Section 4 of Annex IX comprise the complete review of the technical documentation in accordance with Annexes II and III.

In addition, the manufacturer will grant access to the technical documentation as referred to in Section 2.2 of Annex IX and the notified body will provide the manufacturer with a report on the technical documentation assessment. For Class IIa / IIb and Class B / Class C the notified body will neither require an application nor issue an EU technical documentation assessment certificate (see 5.3 for exceptions)."

Section 5.3 then for the MDR adds that for implantable IIb devices a respective certificat is issued.

The guidance came out in December 2019, some months after the first Corrigendum, so I assume they took the changed wording into account and recognized the need to clarify this. I just did not remember that this was clarified in a MDCG Guidance. I cannot rule out that somebody at the EU might interpret this differently, and maybe the general interpretation changed, but at the moment I would consider this to be highly unlikely.

Why DQS sometime in 2022 decided to ignore that guidance (there are some earlier DQS certificates, that handle this differently, see here),  remains a mystery to me. At the moment I have not seen any other NB handing out an Annex IX Chapter II certificate for class IIa/IIb non implantable, but if anyone has other examples, this would be interesting.

Anyway, so much for "crystal clear and unambiguous" formulations in the MDR - confusion everywhere, indeed.

Best regards, Christoph

(edited to include a link to MDCG 2019-13 for convenience)

------------------------------
Christoph Kiesselbach
Schrack & Partner
Reutlingen
Germany

Original Message:
Sent: 26-Jan-2024 12:58
From: Kevin Randall
Subject: CE design certificate

Thank you for your comedy relief of CE standing for "Confusion Everywhere". :)

------------------------------
Kevin Randall, ASQ CQA, RAC (Europe, U.S., Canada)
Principal Consultant
Ridgway, CO
United States
© Copyright by ComplianceAcuity, Inc. All rights reserved.

Original Message:
Sent: 26-Jan-2024 12:51
From: Roger Gray
Subject: CE design certificate

Thanks for the additional information regarding your interaction with the Commission, Kevin, and for reminder that the change to Annex IX Section 3.5 was made in Corrigendum 1..

Unfortunately, the Commission has made the classic error of making one change without understanding the consequences elsewhere in the document. Even with this revision to Annex IX Section 3.5, Section 4 still introduces itself by stating that it applies only to Class III and certain Class II devices, when it no longer does. Clearly, this should have been amended at the same time as the change to Section 3.5. Only then would the current wording be "crystal clear and unambiguous". What is the point of having a section stating that it applies only to Class III and certain Class IIb devices, when it now apparently applies equally to Class IIa and all Class IIb?

It seems to me that the original intention of the MDR was to maintain the single QMS certificate approach of the MDD for Class IIa devices, but since initial publication, it was decided to make a very significant change, but then not implement it properly.

Well it was always said that CE stood for Confusion Everywhere!

------------------------------
Roger Gray
VP Quality and Regulatory
Donawa Lifescience
Clinical Studies - Regulatory - Quality Systems
AR/UKRP services
Italy and UK
rgray@donawa.com
www.donawa.com

Original Message:
Sent: 26-Jan-2024 11:19
From: Kevin Randall
Subject: CE design certificate

Thank you, Roger for your additional insights.  When I pressed the Commission about this early on in the MDR's tenure, the Directorate-General (DG) for Health and Food Safety maintained that the potential contradiction ("Section 4.4 to 4.8" vs. "Section 4) was corrected by Corrigendum, OJ L 117, 3.5.2019, p. 9 (2017/745).  The DG maintained that now, the text of Annex IX, Section 2.3, third subparagraph [and Section 3.5], is "Moreover, in the case of class IIa and class IIb devices, the quality management system assessment shall be accompanied by the assessment of technical documentation for devices selected on a representative basis as specified in Section 4", therefore including the assessment of the entire technical documentation.

I find it interesting that when I pressed the DG on this, the DG still maintained the MDR's very clear and unambiguous requirement for Class IIa device Annex IX Section 4 TD review and TD certification.

Accordingly, I'm not sure this is an issue of "interpretation", since the MDR is so crystal clear and unambiguous about the aforesaid two-certificate approach when the Class IIa Annex IX route is chosen.  Instead, I'd say that if we claim a one-certificate approach, then it is instead a decision to ignore and/or disagree with this very unambiguous MDR requirement that the DG maintained when I pressed.

Maybe there's merit, maybe there's not, for arguing against what the MDR so clearly says in this case.  To know for sure whether there is merit for such opposition, then we need to make a case using something other than claiming an issue of interpretation.  Specifically, we need to make a case explaining why Class IIa devices employing the Annex IX route (traditionally called the "full quality assurance" route) shouldn't be subject to the very basic TD controls of Section 4.  In so doing, we would need to acknowledge that the "design dossier" emphasis of Section 4 from the MDD has been generalized into a more generic/holistic TD assessment rather than just a design dossier review.

I think we need to be sensitive to the possibility that the Commission did this intentionally in order to prescribe additional substance to the assessment of a Class IIa full QMS TD assessment.  Indeed, that same additional substance principle existed in the MDD at Annex II.7, though there it excluded Section 4 because Section 4 was indeed at that time reserved for the design dossier approach.  But as mentioned, that has now been generalized into a more holistic TD assessment providing for either the traditional design dossier level of review, but also to allow use for a more substantive TD assessment for devices (i.e., Class IIa full QMS devices) where some additional substance is of value for public health.  The MDR does after all intend to increase assurance of safety and performance relative to the MDD.

And I don't' think we can make a valid argument against a regulation's multi-purposing of existing legislative infrastructure like when multi-purposing Section 4 for Class IIa devices.  Regulations routinely do that kind of thing in order to avoid replicating/duplicating language in multiple places.  Indeed, incorporation by reference is a commonplace tactic used when writing regulations.

Moreover, the basic title of Annex IX is "Conformity Assessment Based on Quality Management System and On Assessment of Technical Documentation", not, "or Assessment of Technical Documentation".  Accordingly, while we may argue against the inconvenience of issuing two certificates, I'm not yet seeing grounds for arguing against Section 4's additional parameters for TD assessment of Class IIa devices undergoing the full QMS route.

Ultimately again, if we aim to ignore and/or disagree with the use of the Section 4 substance for a Class IIa device full QMS TD assessment, then we need to make a case for why that additional substance is unnecessary.  I haven't heard a good argument for that yet, nor can I think of one myself so far.

------------------------------
Kevin Randall, ASQ CQA, RAC (Europe, U.S., Canada)
Principal Consultant
Ridgway, CO
United States
© Copyright by ComplianceAcuity, Inc. All rights reserved.

Original Message:
Sent: 26-Jan-2024 08:05
From: Roger Gray
Subject: CE design certificate

I know we've gone some way from answering the initial question, but I thought I would just make a couple of comments.

I agree with Christoph's interpretation of both MDD and MDR, in that you generally only need a QMS certificate (MDD Annex II/MDR Annex IX Parts I and III) unless you have a Class III device, or there is a special category of Class IIb device involved.

It seems to me that the current confusion is caused by an ambiguity in the MDR (what a surprise!). If you read MDR Annex IX para 3.5, it says:

"In the case of class IIa and class IIb devices, the surveillance assessment shall also include an assessment of the technical documentation as specified in Section 4 for the device or devices concerned on the basis of further representative samples chosen in accordance with the rationale documented by the notified body in accordance with the third paragraph of Section 2.3."

The only problem is that when you go to Section 4, it is apparent that this entire section applies only to Class III and certain Class IIb devices, as it states in the very first line:

"Assessment of the technical documentation applicable to class III devices and to the class IIb devices referred to in the second subparagraph of Article 52(4)."

So, unfortunately, the referral to Annex IX Section 4 in Section 3.5 does not make sense, as Section 4 apparently applies ONLY to Class III and certain Class IIb devices, meaning that it has no application to Class IIa devices. There is literally zero mention of 'Class IIa' anywhere in Section 4, so I am very confused by DQS issuing an 'Annex IX Chapter II' certificate for a Class IIa device, citing "Conformity assessment based on a Quality Management System and an Assessment of Technical Documentation". 

My interpretation of the MDR is that for a Class IIa device "Conformity assessment based on a Quality Management System and an Assessment of Technical Documentation". would require either:

• Annex IX Parts I and III
• Annex XI Section 10

Note that while the use of Annex IX Parts I and III would result in just one 'Quality Management System' certificate from the notified body, using Annex XI Section 10 would result in two certificates from the notified body, these being a 'Quality Assurance Certificate' plus an 'Annex XI Part A Technical Documentation Certificate'.

Just for fun, I went back to the final draft of the MDR pre-publication, copy dated March 2017, and this is what Annex IX Section 3.5 stated at that time:

"Moreover, in the case of class IIa and class IIb devices, the quality management system assessment shall be accompanied by the assessment of technical documentation for devices selected on a representative basis in accordance with Sections 4.4 to 4.8."

If this wording had been maintained, then the reference to Section 4 would have made sense, because if you just apply Sections 4.4 to 4.8 to Class IIa devices, you completely avoid the possible need for a Technical Documentation Certificate, because this is mentioned only in Section 4.9.

So not for the first time, in moving from an earlier version to a later version, the authoring team have created an ambiguity where one didn't previously exist, thus causing us all these interpretational headaches.

------------------------------
Roger Gray
VP Quality and Regulatory
Donawa Lifescience
Clinical Studies - Regulatory - Quality Systems
AR/UKRP services
Italy and UK
rgray@donawa.com
www.donawa.com

Original Message:
Sent: 25-Jan-2024 01:56
From: Christoph Kiesselbach
Subject: CE design certificate

Hi Kevin,

yes, I think I got a little bit carried away by the topic. Thank you for discussing this, it definitely brought up some points I would like to try to clarify.

I will do some research if there is some kind of consensus between NBs or an MDCG opinion on this and see if it makes sense to post the results in a separate thread.

Best regards, Christoph

------------------------------
Christoph Kiesselbach
Schrack & Partner
Reutlingen
Germany

Original Message:
Sent: 24-Jan-2024 11:47
From: Kevin Randall
Subject: CE design certificate

Thank you, Christoph. NBOG-3 (addressing certificate practices for the outgoing MDD) doesn't directly foresee a two-certificate approach yet does so indirectly in my experience.  Specifically, if the manufacturer, like most do, has chosen to employ an ISO standard as its means for meeting the Annex II Full Quality Assurance QMS requirements, then there is required to be a separate QMS certificate that is the basis for the Annex II Full Quality Assurance EC Certificate.

Also of note is NB-MED/2.5.1/Rec4 stating that the matter of interlinking of certificates has been considered and is recognized to be an administrative responsibility of the NB and should be traceable through the file of the NB. Inclusion of a cross-reference on the certificates themselves, however, would lead to practical difficulties.  Thus, the lack of such cross-references (like in the MDD certificates your cited) certainly doesn't necessarily mean that the certificate stands on its own.

Regarding the various additional questions you've raised (i.e., handling of significant changes for class IIa devices; NB practices for TD sampling after significant changes), those seem to be well beyond the scope of Anon's question.  I certainly have answers for you on those peripheral questions; but I think it's most appropriate to raise them in a separate threat so that this current thread can stay on track with Anon's intended question.

------------------------------
Kevin Randall, ASQ CQA, RAC (Europe, U.S., Canada)
Principal Consultant
Ridgway, CO
United States
© Copyright by ComplianceAcuity, Inc. All rights reserved.

Original Message:
Sent: 24-Jan-2024 03:48
From: Christoph Kiesselbach
Subject: CE design certificate

Hi Kevin,

thank you for providing the examples, that was definitely something I had not realised before. Just for others reading this here are two examples for publicly available current MDR certificates by the mentioned TÜV Süd and BSI that either declare that TD assessment is part of this certificate or that an Annex IX Chapter II certificate is only required for class III and IIb implantable devices (where every TD must be assessed).

With respect to the MDD: At least the old NBOG Best Practice Guide (2010-3) does not foresee a combination of Annex II excluding (4) with Annex II (4) ( for class IIa and IIb, but exclusively for class III. And in the corresponding MDD certificates to the examples you provided DQS used the singe-certificate approach under MDD (see here and here, where an additional Annex II section 4 certificate is limited to class III devices).

Obviously there is some freedom in the interpretaton of the MDR and the corresponding use of certificates. However I would really prefer if there was a common approach by the NBs. I will try to reach out to Team NB and see if they could imagine a common approach for their members.

But aside from the number of certificates, the main impact if Annex IX Chapter II is fully applicable to class IIa and IIb non-implantable devices (aside from the sampling aspect) in my opinion would be that this would also mean that significant design changes must be approved by the Notified Body and lead to a supplement to the (then existing) TD-certificate (of course also under the single-certificate approach each NB might have additional requirements on what needs to be reported by the manufacturer that might be more extensive, but I still think this would be different from a legal requirement with mandatory consequences for the certificates). At least in my experience that was handled differently under MDD for Annex II excluding (4) (at least if you go with the single-certificate approach and the conformity assessment routes given in NBOG 2010-3, see NBOG 2014-3).

There currently is no MDCG guidance on the interpretation of the need for approval of significant changes (with the exception of the specific case of Article 120), so their position is currently not clear to me. But how would this be handled by the NB in the case of TD sampling? If the manufacturer makes a significant change to a device within a group with a TD that has not yet been assessed as a sample, how does the NB assess the change? Based on the change notification only without any knowledge about the previous state of the TD? Do they perform an additional full TD assessment for this device? Or is the assessment only necessary if the TD already had been sampled? Maybe I am wrong but I think this would require a fair amount of additional resources and time, both currently in rather short supply.

Best regards, Christoph

------------------------------
Christoph Kiesselbach
Schrack & Partner
Reutlingen
Germany

Original Message:
Sent: 23-Jan-2024 17:00
From: Kevin Randall
Subject: CE design certificate

Christoph,

I can't show actual clients' examples; but I'll be creative to show a couple Class IIa examples that are publicly/readily available among others on the internet:

• Example 1
• Example 2

Therein on these Class IIa device Annex IX Chapter II EU Technical Documentation Assessment Certificates is stated, "For placing devices listed in the Annex on the market, an additional certificate according to Annex IX, Chapter I and III is required."

Along these lines during my 28-year tenure with medical device CE marking, I've seen many instances of the MDD Article 11(2) second paragraph Class IIa Annex II full quality assurance route generating a QMS certificate plus an EC certificate for the chosen assessment route through the most well-known NBs.

All this said, your insights have educated me to the fact that there does seem to be more NBs employing the single-certificate route for Class IIa devices. For example, it hadn't yet dawned on me that BSI and TUV seem to be sticking to that single-certificate approach for Class IIa Annex IX devices.

My understanding is that the NBs have freedom regarding the particular documented format/way in which they answer the aforesaid two separate, distinct, concomitant Class IIa Annex IX certificate requirements (QMS certificate + TD certificate).  Thus, this liberty may be why we are indeed seeing some use a single-certificate Class IIa Annex IX approach where that one certificate is in fact entitled "EU Quality Management System Assessment Certificate".  But I don't believe that is in clear alignment with the aforesaid two-certificate requirement.  Given how much trouble NBs have had in recent years issuing timely certificates, perhaps they are stretching this issue to reduce the administrative burden rather than following the approach in the examples above which seems to more clearly match the intent of the MDR.

------------------------------
Kevin Randall, ASQ CQA, RAC (Europe, U.S., Canada)
Principal Consultant
Ridgway, CO
United States
© Copyright by ComplianceAcuity, Inc. All rights reserved.

Original Message:
Sent: 23-Jan-2024 12:22
From: Christoph Kiesselbach
Subject: CE design certificate

Hi Kevin,

I respectfully disagree.

Under the MDD e.g. Annex II section 7 made it quite clear that for class IIa and IIb devices section 4 of Annex II (that contains provisions for design examination and a corresponding certificate) does not apply and instead TDs are sampled as part of the audit procedure (Section 3.3 of Annex II). The certificates for the quality management system (Annex II without section 4) that I know then stated that the validity for class III devices would depend on an additional design examination certificate (Annex II section 4) but I know of no additional certificate for class IIa/IIb devices where the TDs were sampled.

The MDR (of course) makes things less clear, but I think a not unusual interpretation of the additional TD assessment of samples in the case of IIa and IIb devices for e.g. Annex IX is that the assessment again is done as part of the audit activities (Chapter I of Annex IX, where also the additional TD sampling during surveillance in described) and while it is generally performed as described in Section 4 of Annex IX, the results are a part of the Annex IX chapter I (and III) certificate. Only class III and class IIb implantable devices where every TD is assessed receive an additional certificate in line with Section 4.9 and are subject to the approval of changes detailed in Section 4.10. I also think this makes sense: Only if every TD is assessed the reporting and approval of changes makes sense. If only samples are assessed, then what would the "approved device" in section 4.10 refer to? The whole generic device group / category of devices that the sample stands for or only the TDs that happen to have been sampled? If the NB does not know the TD the manufacturer is changing (because it has not yet been sampled), how are they assumed to assess changes? Would they have to do an additional full TD assessment?

And so far the MDR certificates I have seen are comparable to the ones under MDD, a certificate according to Annex IX Chapter I (and III) with the additional remark that in case of class III and IIb implantable devices an additional Annex IX Section 4 (or Chapter II) certificate is required (but not for class IIa/IIb).

However, I admit this is less clearly expressed than it was under the MDD and there might be NBs that interpret this differently and hand out Annex IX Chapter II certificates for class IIa/IIb (non implantable) devices. If you have examples for this, I would be interested to know them.

Best regards, Christoph

------------------------------
Christoph Kiesselbach
Schrack & Partner
Reutlingen
Germany

Original Message:
Sent: 23-Jan-2024 09:27
From: Kevin Randall
Subject: CE design certificate

Under both the outgoing Directives (e.g., MDD 93/42/EEC) and the new regulations (e.g., MDR 2017/745), whenever a chosen conformity assessment route requires Notified Body (NB) assessment of the quality management system (QMS), it is always required to be paired with NB assessment of the Technical Documentation (TD).  This always results in two certificates issued in order to gain Notified Body (NB) approval for placing the corresponding devices on the market: One certificate for the quality management system conformity, and one certificate for the device conformity. This two-certificate ensemble is a most fundamental certification requirement that is core to any successful and real CE marking experience where the manufacturer chooses a route involving QMS conformity.

For example, if the Article 52(6) first paragraph Class IIa conformity assessment route is chosen, then, by definition, that route involves two types of conformity assessments and certificates by the NB: One for the QMS plus one for the device TD. This is the way it's always been under the outgoing Directives (e.g., MDD 93/42/EEC), and this has been maintained under the new regulations (e.g., MDR 2017/745).  Specifically, for example under the MDR per Article 52(6) first paragraph, the Annex IX Chapters I & III QMS conformity assessment is performed by the NB resulting in an EU QMS assessment certificate per Annex IX section 2.3 fourth paragraph PLUS an Annex IX Chapter II section 4 TD assessment is performed by the NB resulting in an EU TD assessment certificate per Annex IX section 4.9.

As further examples, the same type of two-certificate requirement applies to the Article 52(6) second paragraph Class IIa conformity assessment route (and others for other device classes).

------------------------------
Kevin Randall, ASQ CQA, RAC (Europe, U.S., Canada)
Principal Consultant
Ridgway, CO
United States
© Copyright by ComplianceAcuity, Inc. All rights reserved.

Original Message:
Sent: 23-Jan-2024 04:11
From: Christoph Kiesselbach
Subject: CE design certificate

Hi Anon,

as Kevin says, the MDR does not use the expression "design certificate".

Annex XII that deals with the certificates issued by a notified body lists: "EU technical documentation assessment certificates, EU type-examination certificates and EU product verification certificates" in section 4(a) and "EU quality management system certificates and EU quality assurance certificates" in section 4(b). You could argue that the former are to some degree "device certificates", because they refer to an individual technical documentation, device type or even single devices, while the latter are "quality management certificates", but there is no such formulation in the MDR.

For a class IIa SaMD you usually would need either a "EU quality management system certificate" (Annex IX, chapters I and III) or a "EU quality assurence certificate" (Annex XI, Part A, Section 10). Both procedures would include the assessment of a technical documentation for at least one representative device sample for each device category, but as far as I know this assessment does not result in a separate certificate (as opposed to e.g. the full Annex IX assessment for class III devices or a type examination according to Annex X). If you only have one device, then of course the sampling in the end means that every device is assessed this way, but again this to my knowledge is not reflected in its own certificate. In theory Annex XI, Part B, Section 18, would be a possible procedure, but testing each device does not make sense for software and I do not think any notified body has this in their scope.

So in a loose sense, you do not obtain a design certificate for a class IIa software, but a quality management certificate (either full quality management system or product assurance). However, using the language of the MDR in describing the certificate is the better idea.

Best regards, Christoph

------------------------------
Christoph Kiesselbach
Schrack & Partner
Reutlingen
Germany

Original Message:
Sent: 21-Jan-2024 07:29
From: Anonymous Member
Subject: CE design certificate

This message was posted by a user wishing to remain anonymous

Dear RAPS Community,

What is CE design certificate and it is mandatory to obtain from NB for class IIa device (SaMD)?

Thank you for any info!

Thank you, Christoph for remembering the MDCG 2019-3 narrative. I would say the reason DQS is currently using the two-certificate approach is because they are implementing the legislation in the way it is clearly and unambiguously written rather than succumbing to group-think or being overwhelmed by intensive reinterpretations.  MDCG 2019-3 is quite dated by now, having been written well before the actual application of the MDR; so perhaps actual experience has influenced DQS.  But ultimately, NBs (as do anyone) have the liberty to ignore guidance because guidance isn't legislation.  I think it would be interesting to hear from DQS about assertions that they are somehow out of bounds; I'll reach out to them for further explanation.

Thank you both for the additional input. I knew I had seen a more restricted version of that reference to section 4 and that was indeed in Annex IX 2.3/3.5 before the EU aligned it with Article 52 (4)/(6) in the first Corrigendum, choosing with uncanny accuracy the more ambiguous version.

Something else I had forgotten was that MDCG 2019-13 talks about TD sampling and with respect to applicability of Chapter II, Section 4 of Annex IX it says in section 5.2:

"Taking into account the wording of articles 52(4) and 52(6) of the MDR and Article 48(7) and 48(9) of the IVDR, as combined with annexes VII and IX, the
tasks to be carried out by the notified body as part of the conformity assessment activities described in Chapter II, Section 4 of Annex IX comprise the complete review of the technical documentation in accordance with Annexes II and III.

In addition, the manufacturer will grant access to the technical documentation as referred to in Section 2.2 of Annex IX and the notified body will provide the manufacturer with a report on the technical documentation assessment. For Class IIa / IIb and Class B / Class C the notified body will neither require an application nor issue an EU technical documentation assessment certificate (see 5.3 for exceptions)."

Section 5.3 then for the MDR adds that for implantable IIb devices a respective certificat is issued.

The guidance came out in December 2019, some months after the first Corrigendum, so I assume they took the changed wording into account and recognized the need to clarify this. I just did not remember that this was clarified in a MDCG Guidance. I cannot rule out that somebody at the EU might interpret this differently, and maybe the general interpretation changed, but at the moment I would consider this to be highly unlikely.

Why DQS sometime in 2022 decided to ignore that guidance (there are some earlier DQS certificates, that handle this differently, see here),  remains a mystery to me. At the moment I have not seen any other NB handing out an Annex IX Chapter II certificate for class IIa/IIb non implantable, but if anyone has other examples, this would be interesting.

Anyway, so much for "crystal clear and unambiguous" formulations in the MDR - confusion everywhere, indeed.

Best regards, Christoph

(edited to include a link to MDCG 2019-13 for convenience)

------------------------------
Christoph Kiesselbach
Schrack & Partner
Reutlingen
Germany

Original Message:
Sent: 26-Jan-2024 12:58
From: Kevin Randall
Subject: CE design certificate

Thank you for your comedy relief of CE standing for "Confusion Everywhere". :)

------------------------------
Kevin Randall, ASQ CQA, RAC (Europe, U.S., Canada)
Principal Consultant
Ridgway, CO
United States
© Copyright by ComplianceAcuity, Inc. All rights reserved.

Original Message:
Sent: 26-Jan-2024 12:51
From: Roger Gray
Subject: CE design certificate

Thanks for the additional information regarding your interaction with the Commission, Kevin, and for reminder that the change to Annex IX Section 3.5 was made in Corrigendum 1..

Unfortunately, the Commission has made the classic error of making one change without understanding the consequences elsewhere in the document. Even with this revision to Annex IX Section 3.5, Section 4 still introduces itself by stating that it applies only to Class III and certain Class II devices, when it no longer does. Clearly, this should have been amended at the same time as the change to Section 3.5. Only then would the current wording be "crystal clear and unambiguous". What is the point of having a section stating that it applies only to Class III and certain Class IIb devices, when it now apparently applies equally to Class IIa and all Class IIb?

It seems to me that the original intention of the MDR was to maintain the single QMS certificate approach of the MDD for Class IIa devices, but since initial publication, it was decided to make a very significant change, but then not implement it properly.

Well it was always said that CE stood for Confusion Everywhere!

------------------------------
Roger Gray
VP Quality and Regulatory
Donawa Lifescience
Clinical Studies - Regulatory - Quality Systems
AR/UKRP services
Italy and UK
rgray@donawa.com
www.donawa.com

Original Message:
Sent: 26-Jan-2024 11:19
From: Kevin Randall
Subject: CE design certificate

Thank you, Roger for your additional insights.  When I pressed the Commission about this early on in the MDR's tenure, the Directorate-General (DG) for Health and Food Safety maintained that the potential contradiction ("Section 4.4 to 4.8" vs. "Section 4) was corrected by Corrigendum, OJ L 117, 3.5.2019, p. 9 (2017/745).  The DG maintained that now, the text of Annex IX, Section 2.3, third subparagraph [and Section 3.5], is "Moreover, in the case of class IIa and class IIb devices, the quality management system assessment shall be accompanied by the assessment of technical documentation for devices selected on a representative basis as specified in Section 4", therefore including the assessment of the entire technical documentation.

I find it interesting that when I pressed the DG on this, the DG still maintained the MDR's very clear and unambiguous requirement for Class IIa device Annex IX Section 4 TD review and TD certification.

Accordingly, I'm not sure this is an issue of "interpretation", since the MDR is so crystal clear and unambiguous about the aforesaid two-certificate approach when the Class IIa Annex IX route is chosen.  Instead, I'd say that if we claim a one-certificate approach, then it is instead a decision to ignore and/or disagree with this very unambiguous MDR requirement that the DG maintained when I pressed.

Maybe there's merit, maybe there's not, for arguing against what the MDR so clearly says in this case.  To know for sure whether there is merit for such opposition, then we need to make a case using something other than claiming an issue of interpretation.  Specifically, we need to make a case explaining why Class IIa devices employing the Annex IX route (traditionally called the "full quality assurance" route) shouldn't be subject to the very basic TD controls of Section 4.  In so doing, we would need to acknowledge that the "design dossier" emphasis of Section 4 from the MDD has been generalized into a more generic/holistic TD assessment rather than just a design dossier review.

I think we need to be sensitive to the possibility that the Commission did this intentionally in order to prescribe additional substance to the assessment of a Class IIa full QMS TD assessment.  Indeed, that same additional substance principle existed in the MDD at Annex II.7, though there it excluded Section 4 because Section 4 was indeed at that time reserved for the design dossier approach.  But as mentioned, that has now been generalized into a more holistic TD assessment providing for either the traditional design dossier level of review, but also to allow use for a more substantive TD assessment for devices (i.e., Class IIa full QMS devices) where some additional substance is of value for public health.  The MDR does after all intend to increase assurance of safety and performance relative to the MDD.

And I don't' think we can make a valid argument against a regulation's multi-purposing of existing legislative infrastructure like when multi-purposing Section 4 for Class IIa devices.  Regulations routinely do that kind of thing in order to avoid replicating/duplicating language in multiple places.  Indeed, incorporation by reference is a commonplace tactic used when writing regulations.

Moreover, the basic title of Annex IX is "Conformity Assessment Based on Quality Management System and On Assessment of Technical Documentation", not, "or Assessment of Technical Documentation".  Accordingly, while we may argue against the inconvenience of issuing two certificates, I'm not yet seeing grounds for arguing against Section 4's additional parameters for TD assessment of Class IIa devices undergoing the full QMS route.

Ultimately again, if we aim to ignore and/or disagree with the use of the Section 4 substance for a Class IIa device full QMS TD assessment, then we need to make a case for why that additional substance is unnecessary.  I haven't heard a good argument for that yet, nor can I think of one myself so far.

------------------------------
Kevin Randall, ASQ CQA, RAC (Europe, U.S., Canada)
Principal Consultant
Ridgway, CO
United States
© Copyright by ComplianceAcuity, Inc. All rights reserved.

Original Message:
Sent: 26-Jan-2024 08:05
From: Roger Gray
Subject: CE design certificate

I know we've gone some way from answering the initial question, but I thought I would just make a couple of comments.

I agree with Christoph's interpretation of both MDD and MDR, in that you generally only need a QMS certificate (MDD Annex II/MDR Annex IX Parts I and III) unless you have a Class III device, or there is a special category of Class IIb device involved.

It seems to me that the current confusion is caused by an ambiguity in the MDR (what a surprise!). If you read MDR Annex IX para 3.5, it says:

"In the case of class IIa and class IIb devices, the surveillance assessment shall also include an assessment of the technical documentation as specified in Section 4 for the device or devices concerned on the basis of further representative samples chosen in accordance with the rationale documented by the notified body in accordance with the third paragraph of Section 2.3."

The only problem is that when you go to Section 4, it is apparent that this entire section applies only to Class III and certain Class IIb devices, as it states in the very first line:

"Assessment of the technical documentation applicable to class III devices and to the class IIb devices referred to in the second subparagraph of Article 52(4)."

So, unfortunately, the referral to Annex IX Section 4 in Section 3.5 does not make sense, as Section 4 apparently applies ONLY to Class III and certain Class IIb devices, meaning that it has no application to Class IIa devices. There is literally zero mention of 'Class IIa' anywhere in Section 4, so I am very confused by DQS issuing an 'Annex IX Chapter II' certificate for a Class IIa device, citing "Conformity assessment based on a Quality Management System and an Assessment of Technical Documentation". 

My interpretation of the MDR is that for a Class IIa device "Conformity assessment based on a Quality Management System and an Assessment of Technical Documentation". would require either:

• Annex IX Parts I and III
• Annex XI Section 10

Note that while the use of Annex IX Parts I and III would result in just one 'Quality Management System' certificate from the notified body, using Annex XI Section 10 would result in two certificates from the notified body, these being a 'Quality Assurance Certificate' plus an 'Annex XI Part A Technical Documentation Certificate'.

Just for fun, I went back to the final draft of the MDR pre-publication, copy dated March 2017, and this is what Annex IX Section 3.5 stated at that time:

"Moreover, in the case of class IIa and class IIb devices, the quality management system assessment shall be accompanied by the assessment of technical documentation for devices selected on a representative basis in accordance with Sections 4.4 to 4.8."

If this wording had been maintained, then the reference to Section 4 would have made sense, because if you just apply Sections 4.4 to 4.8 to Class IIa devices, you completely avoid the possible need for a Technical Documentation Certificate, because this is mentioned only in Section 4.9.

So not for the first time, in moving from an earlier version to a later version, the authoring team have created an ambiguity where one didn't previously exist, thus causing us all these interpretational headaches.

------------------------------
Roger Gray
VP Quality and Regulatory
Donawa Lifescience
Clinical Studies - Regulatory - Quality Systems
AR/UKRP services
Italy and UK
rgray@donawa.com
www.donawa.com

Original Message:
Sent: 25-Jan-2024 01:56
From: Christoph Kiesselbach
Subject: CE design certificate

Hi Kevin,

yes, I think I got a little bit carried away by the topic. Thank you for discussing this, it definitely brought up some points I would like to try to clarify.

I will do some research if there is some kind of consensus between NBs or an MDCG opinion on this and see if it makes sense to post the results in a separate thread.

Best regards, Christoph

------------------------------
Christoph Kiesselbach
Schrack & Partner
Reutlingen
Germany

Original Message:
Sent: 24-Jan-2024 11:47
From: Kevin Randall
Subject: CE design certificate

Thank you, Christoph. NBOG-3 (addressing certificate practices for the outgoing MDD) doesn't directly foresee a two-certificate approach yet does so indirectly in my experience.  Specifically, if the manufacturer, like most do, has chosen to employ an ISO standard as its means for meeting the Annex II Full Quality Assurance QMS requirements, then there is required to be a separate QMS certificate that is the basis for the Annex II Full Quality Assurance EC Certificate.

Also of note is NB-MED/2.5.1/Rec4 stating that the matter of interlinking of certificates has been considered and is recognized to be an administrative responsibility of the NB and should be traceable through the file of the NB. Inclusion of a cross-reference on the certificates themselves, however, would lead to practical difficulties.  Thus, the lack of such cross-references (like in the MDD certificates your cited) certainly doesn't necessarily mean that the certificate stands on its own.

Regarding the various additional questions you've raised (i.e., handling of significant changes for class IIa devices; NB practices for TD sampling after significant changes), those seem to be well beyond the scope of Anon's question.  I certainly have answers for you on those peripheral questions; but I think it's most appropriate to raise them in a separate threat so that this current thread can stay on track with Anon's intended question.

------------------------------
Kevin Randall, ASQ CQA, RAC (Europe, U.S., Canada)
Principal Consultant
Ridgway, CO
United States
© Copyright by ComplianceAcuity, Inc. All rights reserved.

Original Message:
Sent: 24-Jan-2024 03:48
From: Christoph Kiesselbach
Subject: CE design certificate

Hi Kevin,

thank you for providing the examples, that was definitely something I had not realised before. Just for others reading this here are two examples for publicly available current MDR certificates by the mentioned TÜV Süd and BSI that either declare that TD assessment is part of this certificate or that an Annex IX Chapter II certificate is only required for class III and IIb implantable devices (where every TD must be assessed).

With respect to the MDD: At least the old NBOG Best Practice Guide (2010-3) does not foresee a combination of Annex II excluding (4) with Annex II (4) ( for class IIa and IIb, but exclusively for class III. And in the corresponding MDD certificates to the examples you provided DQS used the singe-certificate approach under MDD (see here and here, where an additional Annex II section 4 certificate is limited to class III devices).

Obviously there is some freedom in the interpretaton of the MDR and the corresponding use of certificates. However I would really prefer if there was a common approach by the NBs. I will try to reach out to Team NB and see if they could imagine a common approach for their members.

But aside from the number of certificates, the main impact if Annex IX Chapter II is fully applicable to class IIa and IIb non-implantable devices (aside from the sampling aspect) in my opinion would be that this would also mean that significant design changes must be approved by the Notified Body and lead to a supplement to the (then existing) TD-certificate (of course also under the single-certificate approach each NB might have additional requirements on what needs to be reported by the manufacturer that might be more extensive, but I still think this would be different from a legal requirement with mandatory consequences for the certificates). At least in my experience that was handled differently under MDD for Annex II excluding (4) (at least if you go with the single-certificate approach and the conformity assessment routes given in NBOG 2010-3, see NBOG 2014-3).

There currently is no MDCG guidance on the interpretation of the need for approval of significant changes (with the exception of the specific case of Article 120), so their position is currently not clear to me. But how would this be handled by the NB in the case of TD sampling? If the manufacturer makes a significant change to a device within a group with a TD that has not yet been assessed as a sample, how does the NB assess the change? Based on the change notification only without any knowledge about the previous state of the TD? Do they perform an additional full TD assessment for this device? Or is the assessment only necessary if the TD already had been sampled? Maybe I am wrong but I think this would require a fair amount of additional resources and time, both currently in rather short supply.

Best regards, Christoph

------------------------------
Christoph Kiesselbach
Schrack & Partner
Reutlingen
Germany

Original Message:
Sent: 23-Jan-2024 17:00
From: Kevin Randall
Subject: CE design certificate

Christoph,

I can't show actual clients' examples; but I'll be creative to show a couple Class IIa examples that are publicly/readily available among others on the internet:

• Example 1
• Example 2

Therein on these Class IIa device Annex IX Chapter II EU Technical Documentation Assessment Certificates is stated, "For placing devices listed in the Annex on the market, an additional certificate according to Annex IX, Chapter I and III is required."

Along these lines during my 28-year tenure with medical device CE marking, I've seen many instances of the MDD Article 11(2) second paragraph Class IIa Annex II full quality assurance route generating a QMS certificate plus an EC certificate for the chosen assessment route through the most well-known NBs.

All this said, your insights have educated me to the fact that there does seem to be more NBs employing the single-certificate route for Class IIa devices. For example, it hadn't yet dawned on me that BSI and TUV seem to be sticking to that single-certificate approach for Class IIa Annex IX devices.

My understanding is that the NBs have freedom regarding the particular documented format/way in which they answer the aforesaid two separate, distinct, concomitant Class IIa Annex IX certificate requirements (QMS certificate + TD certificate).  Thus, this liberty may be why we are indeed seeing some use a single-certificate Class IIa Annex IX approach where that one certificate is in fact entitled "EU Quality Management System Assessment Certificate".  But I don't believe that is in clear alignment with the aforesaid two-certificate requirement.  Given how much trouble NBs have had in recent years issuing timely certificates, perhaps they are stretching this issue to reduce the administrative burden rather than following the approach in the examples above which seems to more clearly match the intent of the MDR.

------------------------------
Kevin Randall, ASQ CQA, RAC (Europe, U.S., Canada)
Principal Consultant
Ridgway, CO
United States
© Copyright by ComplianceAcuity, Inc. All rights reserved.

Original Message:
Sent: 23-Jan-2024 12:22
From: Christoph Kiesselbach
Subject: CE design certificate

Hi Kevin,

I respectfully disagree.

Under the MDD e.g. Annex II section 7 made it quite clear that for class IIa and IIb devices section 4 of Annex II (that contains provisions for design examination and a corresponding certificate) does not apply and instead TDs are sampled as part of the audit procedure (Section 3.3 of Annex II). The certificates for the quality management system (Annex II without section 4) that I know then stated that the validity for class III devices would depend on an additional design examination certificate (Annex II section 4) but I know of no additional certificate for class IIa/IIb devices where the TDs were sampled.

The MDR (of course) makes things less clear, but I think a not unusual interpretation of the additional TD assessment of samples in the case of IIa and IIb devices for e.g. Annex IX is that the assessment again is done as part of the audit activities (Chapter I of Annex IX, where also the additional TD sampling during surveillance in described) and while it is generally performed as described in Section 4 of Annex IX, the results are a part of the Annex IX chapter I (and III) certificate. Only class III and class IIb implantable devices where every TD is assessed receive an additional certificate in line with Section 4.9 and are subject to the approval of changes detailed in Section 4.10. I also think this makes sense: Only if every TD is assessed the reporting and approval of changes makes sense. If only samples are assessed, then what would the "approved device" in section 4.10 refer to? The whole generic device group / category of devices that the sample stands for or only the TDs that happen to have been sampled? If the NB does not know the TD the manufacturer is changing (because it has not yet been sampled), how are they assumed to assess changes? Would they have to do an additional full TD assessment?

And so far the MDR certificates I have seen are comparable to the ones under MDD, a certificate according to Annex IX Chapter I (and III) with the additional remark that in case of class III and IIb implantable devices an additional Annex IX Section 4 (or Chapter II) certificate is required (but not for class IIa/IIb).

However, I admit this is less clearly expressed than it was under the MDD and there might be NBs that interpret this differently and hand out Annex IX Chapter II certificates for class IIa/IIb (non implantable) devices. If you have examples for this, I would be interested to know them.

Best regards, Christoph

------------------------------
Christoph Kiesselbach
Schrack & Partner
Reutlingen
Germany

Original Message:
Sent: 23-Jan-2024 09:27
From: Kevin Randall
Subject: CE design certificate

Under both the outgoing Directives (e.g., MDD 93/42/EEC) and the new regulations (e.g., MDR 2017/745), whenever a chosen conformity assessment route requires Notified Body (NB) assessment of the quality management system (QMS), it is always required to be paired with NB assessment of the Technical Documentation (TD).  This always results in two certificates issued in order to gain Notified Body (NB) approval for placing the corresponding devices on the market: One certificate for the quality management system conformity, and one certificate for the device conformity. This two-certificate ensemble is a most fundamental certification requirement that is core to any successful and real CE marking experience where the manufacturer chooses a route involving QMS conformity.

For example, if the Article 52(6) first paragraph Class IIa conformity assessment route is chosen, then, by definition, that route involves two types of conformity assessments and certificates by the NB: One for the QMS plus one for the device TD. This is the way it's always been under the outgoing Directives (e.g., MDD 93/42/EEC), and this has been maintained under the new regulations (e.g., MDR 2017/745).  Specifically, for example under the MDR per Article 52(6) first paragraph, the Annex IX Chapters I & III QMS conformity assessment is performed by the NB resulting in an EU QMS assessment certificate per Annex IX section 2.3 fourth paragraph PLUS an Annex IX Chapter II section 4 TD assessment is performed by the NB resulting in an EU TD assessment certificate per Annex IX section 4.9.

As further examples, the same type of two-certificate requirement applies to the Article 52(6) second paragraph Class IIa conformity assessment route (and others for other device classes).

------------------------------
Kevin Randall, ASQ CQA, RAC (Europe, U.S., Canada)
Principal Consultant
Ridgway, CO
United States
© Copyright by ComplianceAcuity, Inc. All rights reserved.

Original Message:
Sent: 23-Jan-2024 04:11
From: Christoph Kiesselbach
Subject: CE design certificate

Hi Anon,

as Kevin says, the MDR does not use the expression "design certificate".

Annex XII that deals with the certificates issued by a notified body lists: "EU technical documentation assessment certificates, EU type-examination certificates and EU product verification certificates" in section 4(a) and "EU quality management system certificates and EU quality assurance certificates" in section 4(b). You could argue that the former are to some degree "device certificates", because they refer to an individual technical documentation, device type or even single devices, while the latter are "quality management certificates", but there is no such formulation in the MDR.

For a class IIa SaMD you usually would need either a "EU quality management system certificate" (Annex IX, chapters I and III) or a "EU quality assurence certificate" (Annex XI, Part A, Section 10). Both procedures would include the assessment of a technical documentation for at least one representative device sample for each device category, but as far as I know this assessment does not result in a separate certificate (as opposed to e.g. the full Annex IX assessment for class III devices or a type examination according to Annex X). If you only have one device, then of course the sampling in the end means that every device is assessed this way, but again this to my knowledge is not reflected in its own certificate. In theory Annex XI, Part B, Section 18, would be a possible procedure, but testing each device does not make sense for software and I do not think any notified body has this in their scope.

So in a loose sense, you do not obtain a design certificate for a class IIa software, but a quality management certificate (either full quality management system or product assurance). However, using the language of the MDR in describing the certificate is the better idea.

Best regards, Christoph

------------------------------
Christoph Kiesselbach
Schrack & Partner
Reutlingen
Germany

Original Message:
Sent: 21-Jan-2024 07:29
From: Anonymous Member
Subject: CE design certificate

This message was posted by a user wishing to remain anonymous

Dear RAPS Community,

What is CE design certificate and it is mandatory to obtain from NB for class IIa device (SaMD)?

Thank you for any info!

Thanks Kevin for reaching out to DQS, and please note the edit of my previous post: DQS seems to have been consistent and just have clarified the design of their certificates.

If you approach them: I would find it interesting if they consider Annex IX, Section 4.10 with respect to changes applicable for class IIa/IIb non-implantable devices, and if yes how they are handling this in case of sampling

Best regards, Christoph.

Thank you, Christoph for remembering the MDCG 2019-3 narrative. I would say the reason DQS is currently using the two-certificate approach is because they are implementing the legislation in the way it is clearly and unambiguously written rather than succumbing to group-think or being overwhelmed by intensive reinterpretations.  MDCG 2019-3 is quite dated by now, having been written well before the actual application of the MDR; so perhaps actual experience has influenced DQS.  But ultimately, NBs (as do anyone) have the liberty to ignore guidance because guidance isn't legislation.  I think it would be interesting to hear from DQS about assertions that they are somehow out of bounds; I'll reach out to them for further explanation.

Thank you both for the additional input. I knew I had seen a more restricted version of that reference to section 4 and that was indeed in Annex IX 2.3/3.5 before the EU aligned it with Article 52 (4)/(6) in the first Corrigendum, choosing with uncanny accuracy the more ambiguous version.

Something else I had forgotten was that MDCG 2019-13 talks about TD sampling and with respect to applicability of Chapter II, Section 4 of Annex IX it says in section 5.2:

"Taking into account the wording of articles 52(4) and 52(6) of the MDR and Article 48(7) and 48(9) of the IVDR, as combined with annexes VII and IX, the
tasks to be carried out by the notified body as part of the conformity assessment activities described in Chapter II, Section 4 of Annex IX comprise the complete review of the technical documentation in accordance with Annexes II and III.

In addition, the manufacturer will grant access to the technical documentation as referred to in Section 2.2 of Annex IX and the notified body will provide the manufacturer with a report on the technical documentation assessment. For Class IIa / IIb and Class B / Class C the notified body will neither require an application nor issue an EU technical documentation assessment certificate (see 5.3 for exceptions)."

Section 5.3 then for the MDR adds that for implantable IIb devices a respective certificat is issued.

The guidance came out in December 2019, some months after the first Corrigendum, so I assume they took the changed wording into account and recognized the need to clarify this. I just did not remember that this was clarified in a MDCG Guidance. I cannot rule out that somebody at the EU might interpret this differently, and maybe the general interpretation changed, but at the moment I would consider this to be highly unlikely.

Why DQS sometime in 2022 decided to ignore that guidance (there are some earlier DQS certificates, that handle this differently, see here),  remains a mystery to me. At the moment I have not seen any other NB handing out an Annex IX Chapter II certificate for class IIa/IIb non implantable, but if anyone has other examples, this would be interesting.

Anyway, so much for "crystal clear and unambiguous" formulations in the MDR - confusion everywhere, indeed.

Best regards, Christoph

(edited to include a link to MDCG 2019-13 for convenience)

------------------------------
Christoph Kiesselbach
Schrack & Partner
Reutlingen
Germany

Original Message:
Sent: 26-Jan-2024 12:58
From: Kevin Randall
Subject: CE design certificate

Thank you for your comedy relief of CE standing for "Confusion Everywhere". :)

------------------------------
Kevin Randall, ASQ CQA, RAC (Europe, U.S., Canada)
Principal Consultant
Ridgway, CO
United States
© Copyright by ComplianceAcuity, Inc. All rights reserved.

Original Message:
Sent: 26-Jan-2024 12:51
From: Roger Gray
Subject: CE design certificate

Thanks for the additional information regarding your interaction with the Commission, Kevin, and for reminder that the change to Annex IX Section 3.5 was made in Corrigendum 1..

Unfortunately, the Commission has made the classic error of making one change without understanding the consequences elsewhere in the document. Even with this revision to Annex IX Section 3.5, Section 4 still introduces itself by stating that it applies only to Class III and certain Class II devices, when it no longer does. Clearly, this should have been amended at the same time as the change to Section 3.5. Only then would the current wording be "crystal clear and unambiguous". What is the point of having a section stating that it applies only to Class III and certain Class IIb devices, when it now apparently applies equally to Class IIa and all Class IIb?

It seems to me that the original intention of the MDR was to maintain the single QMS certificate approach of the MDD for Class IIa devices, but since initial publication, it was decided to make a very significant change, but then not implement it properly.

Well it was always said that CE stood for Confusion Everywhere!

------------------------------
Roger Gray
VP Quality and Regulatory
Donawa Lifescience
Clinical Studies - Regulatory - Quality Systems
AR/UKRP services
Italy and UK
rgray@donawa.com
www.donawa.com

Original Message:
Sent: 26-Jan-2024 11:19
From: Kevin Randall
Subject: CE design certificate

Thank you, Roger for your additional insights.  When I pressed the Commission about this early on in the MDR's tenure, the Directorate-General (DG) for Health and Food Safety maintained that the potential contradiction ("Section 4.4 to 4.8" vs. "Section 4) was corrected by Corrigendum, OJ L 117, 3.5.2019, p. 9 (2017/745).  The DG maintained that now, the text of Annex IX, Section 2.3, third subparagraph [and Section 3.5], is "Moreover, in the case of class IIa and class IIb devices, the quality management system assessment shall be accompanied by the assessment of technical documentation for devices selected on a representative basis as specified in Section 4", therefore including the assessment of the entire technical documentation.

I find it interesting that when I pressed the DG on this, the DG still maintained the MDR's very clear and unambiguous requirement for Class IIa device Annex IX Section 4 TD review and TD certification.

Accordingly, I'm not sure this is an issue of "interpretation", since the MDR is so crystal clear and unambiguous about the aforesaid two-certificate approach when the Class IIa Annex IX route is chosen.  Instead, I'd say that if we claim a one-certificate approach, then it is instead a decision to ignore and/or disagree with this very unambiguous MDR requirement that the DG maintained when I pressed.

Maybe there's merit, maybe there's not, for arguing against what the MDR so clearly says in this case.  To know for sure whether there is merit for such opposition, then we need to make a case using something other than claiming an issue of interpretation.  Specifically, we need to make a case explaining why Class IIa devices employing the Annex IX route (traditionally called the "full quality assurance" route) shouldn't be subject to the very basic TD controls of Section 4.  In so doing, we would need to acknowledge that the "design dossier" emphasis of Section 4 from the MDD has been generalized into a more generic/holistic TD assessment rather than just a design dossier review.

I think we need to be sensitive to the possibility that the Commission did this intentionally in order to prescribe additional substance to the assessment of a Class IIa full QMS TD assessment.  Indeed, that same additional substance principle existed in the MDD at Annex II.7, though there it excluded Section 4 because Section 4 was indeed at that time reserved for the design dossier approach.  But as mentioned, that has now been generalized into a more holistic TD assessment providing for either the traditional design dossier level of review, but also to allow use for a more substantive TD assessment for devices (i.e., Class IIa full QMS devices) where some additional substance is of value for public health.  The MDR does after all intend to increase assurance of safety and performance relative to the MDD.

And I don't' think we can make a valid argument against a regulation's multi-purposing of existing legislative infrastructure like when multi-purposing Section 4 for Class IIa devices.  Regulations routinely do that kind of thing in order to avoid replicating/duplicating language in multiple places.  Indeed, incorporation by reference is a commonplace tactic used when writing regulations.

Moreover, the basic title of Annex IX is "Conformity Assessment Based on Quality Management System and On Assessment of Technical Documentation", not, "or Assessment of Technical Documentation".  Accordingly, while we may argue against the inconvenience of issuing two certificates, I'm not yet seeing grounds for arguing against Section 4's additional parameters for TD assessment of Class IIa devices undergoing the full QMS route.

Ultimately again, if we aim to ignore and/or disagree with the use of the Section 4 substance for a Class IIa device full QMS TD assessment, then we need to make a case for why that additional substance is unnecessary.  I haven't heard a good argument for that yet, nor can I think of one myself so far.

------------------------------
Kevin Randall, ASQ CQA, RAC (Europe, U.S., Canada)
Principal Consultant
Ridgway, CO
United States
© Copyright by ComplianceAcuity, Inc. All rights reserved.

Original Message:
Sent: 26-Jan-2024 08:05
From: Roger Gray
Subject: CE design certificate

I know we've gone some way from answering the initial question, but I thought I would just make a couple of comments.

I agree with Christoph's interpretation of both MDD and MDR, in that you generally only need a QMS certificate (MDD Annex II/MDR Annex IX Parts I and III) unless you have a Class III device, or there is a special category of Class IIb device involved.

It seems to me that the current confusion is caused by an ambiguity in the MDR (what a surprise!). If you read MDR Annex IX para 3.5, it says:

"In the case of class IIa and class IIb devices, the surveillance assessment shall also include an assessment of the technical documentation as specified in Section 4 for the device or devices concerned on the basis of further representative samples chosen in accordance with the rationale documented by the notified body in accordance with the third paragraph of Section 2.3."

The only problem is that when you go to Section 4, it is apparent that this entire section applies only to Class III and certain Class IIb devices, as it states in the very first line:

"Assessment of the technical documentation applicable to class III devices and to the class IIb devices referred to in the second subparagraph of Article 52(4)."

So, unfortunately, the referral to Annex IX Section 4 in Section 3.5 does not make sense, as Section 4 apparently applies ONLY to Class III and certain Class IIb devices, meaning that it has no application to Class IIa devices. There is literally zero mention of 'Class IIa' anywhere in Section 4, so I am very confused by DQS issuing an 'Annex IX Chapter II' certificate for a Class IIa device, citing "Conformity assessment based on a Quality Management System and an Assessment of Technical Documentation". 

My interpretation of the MDR is that for a Class IIa device "Conformity assessment based on a Quality Management System and an Assessment of Technical Documentation". would require either:

• Annex IX Parts I and III
• Annex XI Section 10

Note that while the use of Annex IX Parts I and III would result in just one 'Quality Management System' certificate from the notified body, using Annex XI Section 10 would result in two certificates from the notified body, these being a 'Quality Assurance Certificate' plus an 'Annex XI Part A Technical Documentation Certificate'.

Just for fun, I went back to the final draft of the MDR pre-publication, copy dated March 2017, and this is what Annex IX Section 3.5 stated at that time:

"Moreover, in the case of class IIa and class IIb devices, the quality management system assessment shall be accompanied by the assessment of technical documentation for devices selected on a representative basis in accordance with Sections 4.4 to 4.8."

If this wording had been maintained, then the reference to Section 4 would have made sense, because if you just apply Sections 4.4 to 4.8 to Class IIa devices, you completely avoid the possible need for a Technical Documentation Certificate, because this is mentioned only in Section 4.9.

So not for the first time, in moving from an earlier version to a later version, the authoring team have created an ambiguity where one didn't previously exist, thus causing us all these interpretational headaches.

------------------------------
Roger Gray
VP Quality and Regulatory
Donawa Lifescience
Clinical Studies - Regulatory - Quality Systems
AR/UKRP services
Italy and UK
rgray@donawa.com
www.donawa.com

Original Message:
Sent: 25-Jan-2024 01:56
From: Christoph Kiesselbach
Subject: CE design certificate

Hi Kevin,

yes, I think I got a little bit carried away by the topic. Thank you for discussing this, it definitely brought up some points I would like to try to clarify.

I will do some research if there is some kind of consensus between NBs or an MDCG opinion on this and see if it makes sense to post the results in a separate thread.

Best regards, Christoph

------------------------------
Christoph Kiesselbach
Schrack & Partner
Reutlingen
Germany

Original Message:
Sent: 24-Jan-2024 11:47
From: Kevin Randall
Subject: CE design certificate

Thank you, Christoph. NBOG-3 (addressing certificate practices for the outgoing MDD) doesn't directly foresee a two-certificate approach yet does so indirectly in my experience.  Specifically, if the manufacturer, like most do, has chosen to employ an ISO standard as its means for meeting the Annex II Full Quality Assurance QMS requirements, then there is required to be a separate QMS certificate that is the basis for the Annex II Full Quality Assurance EC Certificate.

Also of note is NB-MED/2.5.1/Rec4 stating that the matter of interlinking of certificates has been considered and is recognized to be an administrative responsibility of the NB and should be traceable through the file of the NB. Inclusion of a cross-reference on the certificates themselves, however, would lead to practical difficulties.  Thus, the lack of such cross-references (like in the MDD certificates your cited) certainly doesn't necessarily mean that the certificate stands on its own.

Regarding the various additional questions you've raised (i.e., handling of significant changes for class IIa devices; NB practices for TD sampling after significant changes), those seem to be well beyond the scope of Anon's question.  I certainly have answers for you on those peripheral questions; but I think it's most appropriate to raise them in a separate threat so that this current thread can stay on track with Anon's intended question.

------------------------------
Kevin Randall, ASQ CQA, RAC (Europe, U.S., Canada)
Principal Consultant
Ridgway, CO
United States
© Copyright by ComplianceAcuity, Inc. All rights reserved.

Original Message:
Sent: 24-Jan-2024 03:48
From: Christoph Kiesselbach
Subject: CE design certificate

Hi Kevin,

thank you for providing the examples, that was definitely something I had not realised before. Just for others reading this here are two examples for publicly available current MDR certificates by the mentioned TÜV Süd and BSI that either declare that TD assessment is part of this certificate or that an Annex IX Chapter II certificate is only required for class III and IIb implantable devices (where every TD must be assessed).

With respect to the MDD: At least the old NBOG Best Practice Guide (2010-3) does not foresee a combination of Annex II excluding (4) with Annex II (4) ( for class IIa and IIb, but exclusively for class III. And in the corresponding MDD certificates to the examples you provided DQS used the singe-certificate approach under MDD (see here and here, where an additional Annex II section 4 certificate is limited to class III devices).

Obviously there is some freedom in the interpretaton of the MDR and the corresponding use of certificates. However I would really prefer if there was a common approach by the NBs. I will try to reach out to Team NB and see if they could imagine a common approach for their members.

But aside from the number of certificates, the main impact if Annex IX Chapter II is fully applicable to class IIa and IIb non-implantable devices (aside from the sampling aspect) in my opinion would be that this would also mean that significant design changes must be approved by the Notified Body and lead to a supplement to the (then existing) TD-certificate (of course also under the single-certificate approach each NB might have additional requirements on what needs to be reported by the manufacturer that might be more extensive, but I still think this would be different from a legal requirement with mandatory consequences for the certificates). At least in my experience that was handled differently under MDD for Annex II excluding (4) (at least if you go with the single-certificate approach and the conformity assessment routes given in NBOG 2010-3, see NBOG 2014-3).

There currently is no MDCG guidance on the interpretation of the need for approval of significant changes (with the exception of the specific case of Article 120), so their position is currently not clear to me. But how would this be handled by the NB in the case of TD sampling? If the manufacturer makes a significant change to a device within a group with a TD that has not yet been assessed as a sample, how does the NB assess the change? Based on the change notification only without any knowledge about the previous state of the TD? Do they perform an additional full TD assessment for this device? Or is the assessment only necessary if the TD already had been sampled? Maybe I am wrong but I think this would require a fair amount of additional resources and time, both currently in rather short supply.

Best regards, Christoph

------------------------------
Christoph Kiesselbach
Schrack & Partner
Reutlingen
Germany

Original Message:
Sent: 23-Jan-2024 17:00
From: Kevin Randall
Subject: CE design certificate

Christoph,

I can't show actual clients' examples; but I'll be creative to show a couple Class IIa examples that are publicly/readily available among others on the internet:

• Example 1
• Example 2

Therein on these Class IIa device Annex IX Chapter II EU Technical Documentation Assessment Certificates is stated, "For placing devices listed in the Annex on the market, an additional certificate according to Annex IX, Chapter I and III is required."

Along these lines during my 28-year tenure with medical device CE marking, I've seen many instances of the MDD Article 11(2) second paragraph Class IIa Annex II full quality assurance route generating a QMS certificate plus an EC certificate for the chosen assessment route through the most well-known NBs.

All this said, your insights have educated me to the fact that there does seem to be more NBs employing the single-certificate route for Class IIa devices. For example, it hadn't yet dawned on me that BSI and TUV seem to be sticking to that single-certificate approach for Class IIa Annex IX devices.

My understanding is that the NBs have freedom regarding the particular documented format/way in which they answer the aforesaid two separate, distinct, concomitant Class IIa Annex IX certificate requirements (QMS certificate + TD certificate).  Thus, this liberty may be why we are indeed seeing some use a single-certificate Class IIa Annex IX approach where that one certificate is in fact entitled "EU Quality Management System Assessment Certificate".  But I don't believe that is in clear alignment with the aforesaid two-certificate requirement.  Given how much trouble NBs have had in recent years issuing timely certificates, perhaps they are stretching this issue to reduce the administrative burden rather than following the approach in the examples above which seems to more clearly match the intent of the MDR.

------------------------------
Kevin Randall, ASQ CQA, RAC (Europe, U.S., Canada)
Principal Consultant
Ridgway, CO
United States
© Copyright by ComplianceAcuity, Inc. All rights reserved.

Original Message:
Sent: 23-Jan-2024 12:22
From: Christoph Kiesselbach
Subject: CE design certificate

Hi Kevin,

I respectfully disagree.

Under the MDD e.g. Annex II section 7 made it quite clear that for class IIa and IIb devices section 4 of Annex II (that contains provisions for design examination and a corresponding certificate) does not apply and instead TDs are sampled as part of the audit procedure (Section 3.3 of Annex II). The certificates for the quality management system (Annex II without section 4) that I know then stated that the validity for class III devices would depend on an additional design examination certificate (Annex II section 4) but I know of no additional certificate for class IIa/IIb devices where the TDs were sampled.

The MDR (of course) makes things less clear, but I think a not unusual interpretation of the additional TD assessment of samples in the case of IIa and IIb devices for e.g. Annex IX is that the assessment again is done as part of the audit activities (Chapter I of Annex IX, where also the additional TD sampling during surveillance in described) and while it is generally performed as described in Section 4 of Annex IX, the results are a part of the Annex IX chapter I (and III) certificate. Only class III and class IIb implantable devices where every TD is assessed receive an additional certificate in line with Section 4.9 and are subject to the approval of changes detailed in Section 4.10. I also think this makes sense: Only if every TD is assessed the reporting and approval of changes makes sense. If only samples are assessed, then what would the "approved device" in section 4.10 refer to? The whole generic device group / category of devices that the sample stands for or only the TDs that happen to have been sampled? If the NB does not know the TD the manufacturer is changing (because it has not yet been sampled), how are they assumed to assess changes? Would they have to do an additional full TD assessment?

And so far the MDR certificates I have seen are comparable to the ones under MDD, a certificate according to Annex IX Chapter I (and III) with the additional remark that in case of class III and IIb implantable devices an additional Annex IX Section 4 (or Chapter II) certificate is required (but not for class IIa/IIb).

However, I admit this is less clearly expressed than it was under the MDD and there might be NBs that interpret this differently and hand out Annex IX Chapter II certificates for class IIa/IIb (non implantable) devices. If you have examples for this, I would be interested to know them.

Best regards, Christoph

------------------------------
Christoph Kiesselbach
Schrack & Partner
Reutlingen
Germany

Original Message:
Sent: 23-Jan-2024 09:27
From: Kevin Randall
Subject: CE design certificate

Under both the outgoing Directives (e.g., MDD 93/42/EEC) and the new regulations (e.g., MDR 2017/745), whenever a chosen conformity assessment route requires Notified Body (NB) assessment of the quality management system (QMS), it is always required to be paired with NB assessment of the Technical Documentation (TD).  This always results in two certificates issued in order to gain Notified Body (NB) approval for placing the corresponding devices on the market: One certificate for the quality management system conformity, and one certificate for the device conformity. This two-certificate ensemble is a most fundamental certification requirement that is core to any successful and real CE marking experience where the manufacturer chooses a route involving QMS conformity.

For example, if the Article 52(6) first paragraph Class IIa conformity assessment route is chosen, then, by definition, that route involves two types of conformity assessments and certificates by the NB: One for the QMS plus one for the device TD. This is the way it's always been under the outgoing Directives (e.g., MDD 93/42/EEC), and this has been maintained under the new regulations (e.g., MDR 2017/745).  Specifically, for example under the MDR per Article 52(6) first paragraph, the Annex IX Chapters I & III QMS conformity assessment is performed by the NB resulting in an EU QMS assessment certificate per Annex IX section 2.3 fourth paragraph PLUS an Annex IX Chapter II section 4 TD assessment is performed by the NB resulting in an EU TD assessment certificate per Annex IX section 4.9.

As further examples, the same type of two-certificate requirement applies to the Article 52(6) second paragraph Class IIa conformity assessment route (and others for other device classes).

------------------------------
Kevin Randall, ASQ CQA, RAC (Europe, U.S., Canada)
Principal Consultant
Ridgway, CO
United States
© Copyright by ComplianceAcuity, Inc. All rights reserved.

Original Message:
Sent: 23-Jan-2024 04:11
From: Christoph Kiesselbach
Subject: CE design certificate

Hi Anon,

as Kevin says, the MDR does not use the expression "design certificate".

Annex XII that deals with the certificates issued by a notified body lists: "EU technical documentation assessment certificates, EU type-examination certificates and EU product verification certificates" in section 4(a) and "EU quality management system certificates and EU quality assurance certificates" in section 4(b). You could argue that the former are to some degree "device certificates", because they refer to an individual technical documentation, device type or even single devices, while the latter are "quality management certificates", but there is no such formulation in the MDR.

For a class IIa SaMD you usually would need either a "EU quality management system certificate" (Annex IX, chapters I and III) or a "EU quality assurence certificate" (Annex XI, Part A, Section 10). Both procedures would include the assessment of a technical documentation for at least one representative device sample for each device category, but as far as I know this assessment does not result in a separate certificate (as opposed to e.g. the full Annex IX assessment for class III devices or a type examination according to Annex X). If you only have one device, then of course the sampling in the end means that every device is assessed this way, but again this to my knowledge is not reflected in its own certificate. In theory Annex XI, Part B, Section 18, would be a possible procedure, but testing each device does not make sense for software and I do not think any notified body has this in their scope.

So in a loose sense, you do not obtain a design certificate for a class IIa software, but a quality management certificate (either full quality management system or product assurance). However, using the language of the MDR in describing the certificate is the better idea.

Best regards, Christoph

------------------------------
Christoph Kiesselbach
Schrack & Partner
Reutlingen
Germany

Original Message:
Sent: 21-Jan-2024 07:29
From: Anonymous Member
Subject: CE design certificate

This message was posted by a user wishing to remain anonymous

Dear RAPS Community,

What is CE design certificate and it is mandatory to obtain from NB for class IIa device (SaMD)?

Thank you for any info!

Hello Christoph,

I was just informed by DQS that they are doing away with the two-certificate approach.  In other words, they are adjusting their practices to instead align with what we saw the other NBs doing.  Specifically, they are adjusting so as to not issue both an EU Quality Management System Certificate plus an EU Technical Documentation Assessment Certificate.  Rather, it will be as you and Roger were saying (one type of certificate or the other but not both).

Yet the frequent issuance of a supporting QMS certificate is still in place for what is recognized to be the most common scenario where the manufacturer elects to formally employ certification to a standard like EN ISO 13485 as its means for meeting the MDR's Article 10(9) and Annex IX quality management system requirements such as in this public example here.

Thank you for your persistence on this topic.

Thanks Kevin for reaching out to DQS, and please note the edit of my previous post: DQS seems to have been consistent and just have clarified the design of their certificates.

If you approach them: I would find it interesting if they consider Annex IX, Section 4.10 with respect to changes applicable for class IIa/IIb non-implantable devices, and if yes how they are handling this in case of sampling

Best regards, Christoph.

Thank you, Christoph for remembering the MDCG 2019-3 narrative. I would say the reason DQS is currently using the two-certificate approach is because they are implementing the legislation in the way it is clearly and unambiguously written rather than succumbing to group-think or being overwhelmed by intensive reinterpretations.  MDCG 2019-3 is quite dated by now, having been written well before the actual application of the MDR; so perhaps actual experience has influenced DQS.  But ultimately, NBs (as do anyone) have the liberty to ignore guidance because guidance isn't legislation.  I think it would be interesting to hear from DQS about assertions that they are somehow out of bounds; I'll reach out to them for further explanation.

Thank you both for the additional input. I knew I had seen a more restricted version of that reference to section 4 and that was indeed in Annex IX 2.3/3.5 before the EU aligned it with Article 52 (4)/(6) in the first Corrigendum, choosing with uncanny accuracy the more ambiguous version.

Something else I had forgotten was that MDCG 2019-13 talks about TD sampling and with respect to applicability of Chapter II, Section 4 of Annex IX it says in section 5.2:

"Taking into account the wording of articles 52(4) and 52(6) of the MDR and Article 48(7) and 48(9) of the IVDR, as combined with annexes VII and IX, the
tasks to be carried out by the notified body as part of the conformity assessment activities described in Chapter II, Section 4 of Annex IX comprise the complete review of the technical documentation in accordance with Annexes II and III.

In addition, the manufacturer will grant access to the technical documentation as referred to in Section 2.2 of Annex IX and the notified body will provide the manufacturer with a report on the technical documentation assessment. For Class IIa / IIb and Class B / Class C the notified body will neither require an application nor issue an EU technical documentation assessment certificate (see 5.3 for exceptions)."

Section 5.3 then for the MDR adds that for implantable IIb devices a respective certificat is issued.

The guidance came out in December 2019, some months after the first Corrigendum, so I assume they took the changed wording into account and recognized the need to clarify this. I just did not remember that this was clarified in a MDCG Guidance. I cannot rule out that somebody at the EU might interpret this differently, and maybe the general interpretation changed, but at the moment I would consider this to be highly unlikely.

Why DQS sometime in 2022 decided to ignore that guidance (there are some earlier DQS certificates, that handle this differently, see here),  remains a mystery to me. At the moment I have not seen any other NB handing out an Annex IX Chapter II certificate for class IIa/IIb non implantable, but if anyone has other examples, this would be interesting.

Anyway, so much for "crystal clear and unambiguous" formulations in the MDR - confusion everywhere, indeed.

Best regards, Christoph

(edited to include a link to MDCG 2019-13 for convenience)

------------------------------
Christoph Kiesselbach
Schrack & Partner
Reutlingen
Germany

Original Message:
Sent: 26-Jan-2024 12:58
From: Kevin Randall
Subject: CE design certificate

Thank you for your comedy relief of CE standing for "Confusion Everywhere". :)

------------------------------
Kevin Randall, ASQ CQA, RAC (Europe, U.S., Canada)
Principal Consultant
Ridgway, CO
United States
© Copyright by ComplianceAcuity, Inc. All rights reserved.

Original Message:
Sent: 26-Jan-2024 12:51
From: Roger Gray
Subject: CE design certificate

Thanks for the additional information regarding your interaction with the Commission, Kevin, and for reminder that the change to Annex IX Section 3.5 was made in Corrigendum 1..

Unfortunately, the Commission has made the classic error of making one change without understanding the consequences elsewhere in the document. Even with this revision to Annex IX Section 3.5, Section 4 still introduces itself by stating that it applies only to Class III and certain Class II devices, when it no longer does. Clearly, this should have been amended at the same time as the change to Section 3.5. Only then would the current wording be "crystal clear and unambiguous". What is the point of having a section stating that it applies only to Class III and certain Class IIb devices, when it now apparently applies equally to Class IIa and all Class IIb?

It seems to me that the original intention of the MDR was to maintain the single QMS certificate approach of the MDD for Class IIa devices, but since initial publication, it was decided to make a very significant change, but then not implement it properly.

Well it was always said that CE stood for Confusion Everywhere!

------------------------------
Roger Gray
VP Quality and Regulatory
Donawa Lifescience
Clinical Studies - Regulatory - Quality Systems
AR/UKRP services
Italy and UK
rgray@donawa.com
www.donawa.com

Original Message:
Sent: 26-Jan-2024 11:19
From: Kevin Randall
Subject: CE design certificate

Thank you, Roger for your additional insights.  When I pressed the Commission about this early on in the MDR's tenure, the Directorate-General (DG) for Health and Food Safety maintained that the potential contradiction ("Section 4.4 to 4.8" vs. "Section 4) was corrected by Corrigendum, OJ L 117, 3.5.2019, p. 9 (2017/745).  The DG maintained that now, the text of Annex IX, Section 2.3, third subparagraph [and Section 3.5], is "Moreover, in the case of class IIa and class IIb devices, the quality management system assessment shall be accompanied by the assessment of technical documentation for devices selected on a representative basis as specified in Section 4", therefore including the assessment of the entire technical documentation.

I find it interesting that when I pressed the DG on this, the DG still maintained the MDR's very clear and unambiguous requirement for Class IIa device Annex IX Section 4 TD review and TD certification.

Accordingly, I'm not sure this is an issue of "interpretation", since the MDR is so crystal clear and unambiguous about the aforesaid two-certificate approach when the Class IIa Annex IX route is chosen.  Instead, I'd say that if we claim a one-certificate approach, then it is instead a decision to ignore and/or disagree with this very unambiguous MDR requirement that the DG maintained when I pressed.

Maybe there's merit, maybe there's not, for arguing against what the MDR so clearly says in this case.  To know for sure whether there is merit for such opposition, then we need to make a case using something other than claiming an issue of interpretation.  Specifically, we need to make a case explaining why Class IIa devices employing the Annex IX route (traditionally called the "full quality assurance" route) shouldn't be subject to the very basic TD controls of Section 4.  In so doing, we would need to acknowledge that the "design dossier" emphasis of Section 4 from the MDD has been generalized into a more generic/holistic TD assessment rather than just a design dossier review.

I think we need to be sensitive to the possibility that the Commission did this intentionally in order to prescribe additional substance to the assessment of a Class IIa full QMS TD assessment.  Indeed, that same additional substance principle existed in the MDD at Annex II.7, though there it excluded Section 4 because Section 4 was indeed at that time reserved for the design dossier approach.  But as mentioned, that has now been generalized into a more holistic TD assessment providing for either the traditional design dossier level of review, but also to allow use for a more substantive TD assessment for devices (i.e., Class IIa full QMS devices) where some additional substance is of value for public health.  The MDR does after all intend to increase assurance of safety and performance relative to the MDD.

And I don't' think we can make a valid argument against a regulation's multi-purposing of existing legislative infrastructure like when multi-purposing Section 4 for Class IIa devices.  Regulations routinely do that kind of thing in order to avoid replicating/duplicating language in multiple places.  Indeed, incorporation by reference is a commonplace tactic used when writing regulations.

Moreover, the basic title of Annex IX is "Conformity Assessment Based on Quality Management System and On Assessment of Technical Documentation", not, "or Assessment of Technical Documentation".  Accordingly, while we may argue against the inconvenience of issuing two certificates, I'm not yet seeing grounds for arguing against Section 4's additional parameters for TD assessment of Class IIa devices undergoing the full QMS route.

Ultimately again, if we aim to ignore and/or disagree with the use of the Section 4 substance for a Class IIa device full QMS TD assessment, then we need to make a case for why that additional substance is unnecessary.  I haven't heard a good argument for that yet, nor can I think of one myself so far.

------------------------------
Kevin Randall, ASQ CQA, RAC (Europe, U.S., Canada)
Principal Consultant
Ridgway, CO
United States
© Copyright by ComplianceAcuity, Inc. All rights reserved.

Original Message:
Sent: 26-Jan-2024 08:05
From: Roger Gray
Subject: CE design certificate

I know we've gone some way from answering the initial question, but I thought I would just make a couple of comments.

I agree with Christoph's interpretation of both MDD and MDR, in that you generally only need a QMS certificate (MDD Annex II/MDR Annex IX Parts I and III) unless you have a Class III device, or there is a special category of Class IIb device involved.

It seems to me that the current confusion is caused by an ambiguity in the MDR (what a surprise!). If you read MDR Annex IX para 3.5, it says:

"In the case of class IIa and class IIb devices, the surveillance assessment shall also include an assessment of the technical documentation as specified in Section 4 for the device or devices concerned on the basis of further representative samples chosen in accordance with the rationale documented by the notified body in accordance with the third paragraph of Section 2.3."

The only problem is that when you go to Section 4, it is apparent that this entire section applies only to Class III and certain Class IIb devices, as it states in the very first line:

"Assessment of the technical documentation applicable to class III devices and to the class IIb devices referred to in the second subparagraph of Article 52(4)."

So, unfortunately, the referral to Annex IX Section 4 in Section 3.5 does not make sense, as Section 4 apparently applies ONLY to Class III and certain Class IIb devices, meaning that it has no application to Class IIa devices. There is literally zero mention of 'Class IIa' anywhere in Section 4, so I am very confused by DQS issuing an 'Annex IX Chapter II' certificate for a Class IIa device, citing "Conformity assessment based on a Quality Management System and an Assessment of Technical Documentation". 

My interpretation of the MDR is that for a Class IIa device "Conformity assessment based on a Quality Management System and an Assessment of Technical Documentation". would require either:

• Annex IX Parts I and III
• Annex XI Section 10

Note that while the use of Annex IX Parts I and III would result in just one 'Quality Management System' certificate from the notified body, using Annex XI Section 10 would result in two certificates from the notified body, these being a 'Quality Assurance Certificate' plus an 'Annex XI Part A Technical Documentation Certificate'.

Just for fun, I went back to the final draft of the MDR pre-publication, copy dated March 2017, and this is what Annex IX Section 3.5 stated at that time:

"Moreover, in the case of class IIa and class IIb devices, the quality management system assessment shall be accompanied by the assessment of technical documentation for devices selected on a representative basis in accordance with Sections 4.4 to 4.8."

If this wording had been maintained, then the reference to Section 4 would have made sense, because if you just apply Sections 4.4 to 4.8 to Class IIa devices, you completely avoid the possible need for a Technical Documentation Certificate, because this is mentioned only in Section 4.9.

So not for the first time, in moving from an earlier version to a later version, the authoring team have created an ambiguity where one didn't previously exist, thus causing us all these interpretational headaches.

------------------------------
Roger Gray
VP Quality and Regulatory
Donawa Lifescience
Clinical Studies - Regulatory - Quality Systems
AR/UKRP services
Italy and UK
rgray@donawa.com
www.donawa.com

Original Message:
Sent: 25-Jan-2024 01:56
From: Christoph Kiesselbach
Subject: CE design certificate

Hi Kevin,

yes, I think I got a little bit carried away by the topic. Thank you for discussing this, it definitely brought up some points I would like to try to clarify.

I will do some research if there is some kind of consensus between NBs or an MDCG opinion on this and see if it makes sense to post the results in a separate thread.

Best regards, Christoph

------------------------------
Christoph Kiesselbach
Schrack & Partner
Reutlingen
Germany

Original Message:
Sent: 24-Jan-2024 11:47
From: Kevin Randall
Subject: CE design certificate

Thank you, Christoph. NBOG-3 (addressing certificate practices for the outgoing MDD) doesn't directly foresee a two-certificate approach yet does so indirectly in my experience.  Specifically, if the manufacturer, like most do, has chosen to employ an ISO standard as its means for meeting the Annex II Full Quality Assurance QMS requirements, then there is required to be a separate QMS certificate that is the basis for the Annex II Full Quality Assurance EC Certificate.

Also of note is NB-MED/2.5.1/Rec4 stating that the matter of interlinking of certificates has been considered and is recognized to be an administrative responsibility of the NB and should be traceable through the file of the NB. Inclusion of a cross-reference on the certificates themselves, however, would lead to practical difficulties.  Thus, the lack of such cross-references (like in the MDD certificates your cited) certainly doesn't necessarily mean that the certificate stands on its own.

Regarding the various additional questions you've raised (i.e., handling of significant changes for class IIa devices; NB practices for TD sampling after significant changes), those seem to be well beyond the scope of Anon's question.  I certainly have answers for you on those peripheral questions; but I think it's most appropriate to raise them in a separate threat so that this current thread can stay on track with Anon's intended question.

------------------------------
Kevin Randall, ASQ CQA, RAC (Europe, U.S., Canada)
Principal Consultant
Ridgway, CO
United States
© Copyright by ComplianceAcuity, Inc. All rights reserved.

Original Message:
Sent: 24-Jan-2024 03:48
From: Christoph Kiesselbach
Subject: CE design certificate

Hi Kevin,

thank you for providing the examples, that was definitely something I had not realised before. Just for others reading this here are two examples for publicly available current MDR certificates by the mentioned TÜV Süd and BSI that either declare that TD assessment is part of this certificate or that an Annex IX Chapter II certificate is only required for class III and IIb implantable devices (where every TD must be assessed).

With respect to the MDD: At least the old NBOG Best Practice Guide (2010-3) does not foresee a combination of Annex II excluding (4) with Annex II (4) ( for class IIa and IIb, but exclusively for class III. And in the corresponding MDD certificates to the examples you provided DQS used the singe-certificate approach under MDD (see here and here, where an additional Annex II section 4 certificate is limited to class III devices).

Obviously there is some freedom in the interpretaton of the MDR and the corresponding use of certificates. However I would really prefer if there was a common approach by the NBs. I will try to reach out to Team NB and see if they could imagine a common approach for their members.

But aside from the number of certificates, the main impact if Annex IX Chapter II is fully applicable to class IIa and IIb non-implantable devices (aside from the sampling aspect) in my opinion would be that this would also mean that significant design changes must be approved by the Notified Body and lead to a supplement to the (then existing) TD-certificate (of course also under the single-certificate approach each NB might have additional requirements on what needs to be reported by the manufacturer that might be more extensive, but I still think this would be different from a legal requirement with mandatory consequences for the certificates). At least in my experience that was handled differently under MDD for Annex II excluding (4) (at least if you go with the single-certificate approach and the conformity assessment routes given in NBOG 2010-3, see NBOG 2014-3).

There currently is no MDCG guidance on the interpretation of the need for approval of significant changes (with the exception of the specific case of Article 120), so their position is currently not clear to me. But how would this be handled by the NB in the case of TD sampling? If the manufacturer makes a significant change to a device within a group with a TD that has not yet been assessed as a sample, how does the NB assess the change? Based on the change notification only without any knowledge about the previous state of the TD? Do they perform an additional full TD assessment for this device? Or is the assessment only necessary if the TD already had been sampled? Maybe I am wrong but I think this would require a fair amount of additional resources and time, both currently in rather short supply.

Best regards, Christoph

------------------------------
Christoph Kiesselbach
Schrack & Partner
Reutlingen
Germany

Original Message:
Sent: 23-Jan-2024 17:00
From: Kevin Randall
Subject: CE design certificate

Christoph,

I can't show actual clients' examples; but I'll be creative to show a couple Class IIa examples that are publicly/readily available among others on the internet:

• Example 1
• Example 2

Therein on these Class IIa device Annex IX Chapter II EU Technical Documentation Assessment Certificates is stated, "For placing devices listed in the Annex on the market, an additional certificate according to Annex IX, Chapter I and III is required."

Along these lines during my 28-year tenure with medical device CE marking, I've seen many instances of the MDD Article 11(2) second paragraph Class IIa Annex II full quality assurance route generating a QMS certificate plus an EC certificate for the chosen assessment route through the most well-known NBs.

All this said, your insights have educated me to the fact that there does seem to be more NBs employing the single-certificate route for Class IIa devices. For example, it hadn't yet dawned on me that BSI and TUV seem to be sticking to that single-certificate approach for Class IIa Annex IX devices.

My understanding is that the NBs have freedom regarding the particular documented format/way in which they answer the aforesaid two separate, distinct, concomitant Class IIa Annex IX certificate requirements (QMS certificate + TD certificate).  Thus, this liberty may be why we are indeed seeing some use a single-certificate Class IIa Annex IX approach where that one certificate is in fact entitled "EU Quality Management System Assessment Certificate".  But I don't believe that is in clear alignment with the aforesaid two-certificate requirement.  Given how much trouble NBs have had in recent years issuing timely certificates, perhaps they are stretching this issue to reduce the administrative burden rather than following the approach in the examples above which seems to more clearly match the intent of the MDR.

------------------------------
Kevin Randall, ASQ CQA, RAC (Europe, U.S., Canada)
Principal Consultant
Ridgway, CO
United States
© Copyright by ComplianceAcuity, Inc. All rights reserved.

Original Message:
Sent: 23-Jan-2024 12:22
From: Christoph Kiesselbach
Subject: CE design certificate

Hi Kevin,

I respectfully disagree.

Under the MDD e.g. Annex II section 7 made it quite clear that for class IIa and IIb devices section 4 of Annex II (that contains provisions for design examination and a corresponding certificate) does not apply and instead TDs are sampled as part of the audit procedure (Section 3.3 of Annex II). The certificates for the quality management system (Annex II without section 4) that I know then stated that the validity for class III devices would depend on an additional design examination certificate (Annex II section 4) but I know of no additional certificate for class IIa/IIb devices where the TDs were sampled.

The MDR (of course) makes things less clear, but I think a not unusual interpretation of the additional TD assessment of samples in the case of IIa and IIb devices for e.g. Annex IX is that the assessment again is done as part of the audit activities (Chapter I of Annex IX, where also the additional TD sampling during surveillance in described) and while it is generally performed as described in Section 4 of Annex IX, the results are a part of the Annex IX chapter I (and III) certificate. Only class III and class IIb implantable devices where every TD is assessed receive an additional certificate in line with Section 4.9 and are subject to the approval of changes detailed in Section 4.10. I also think this makes sense: Only if every TD is assessed the reporting and approval of changes makes sense. If only samples are assessed, then what would the "approved device" in section 4.10 refer to? The whole generic device group / category of devices that the sample stands for or only the TDs that happen to have been sampled? If the NB does not know the TD the manufacturer is changing (because it has not yet been sampled), how are they assumed to assess changes? Would they have to do an additional full TD assessment?

And so far the MDR certificates I have seen are comparable to the ones under MDD, a certificate according to Annex IX Chapter I (and III) with the additional remark that in case of class III and IIb implantable devices an additional Annex IX Section 4 (or Chapter II) certificate is required (but not for class IIa/IIb).

However, I admit this is less clearly expressed than it was under the MDD and there might be NBs that interpret this differently and hand out Annex IX Chapter II certificates for class IIa/IIb (non implantable) devices. If you have examples for this, I would be interested to know them.

Best regards, Christoph

------------------------------
Christoph Kiesselbach
Schrack & Partner
Reutlingen
Germany

Original Message:
Sent: 23-Jan-2024 09:27
From: Kevin Randall
Subject: CE design certificate

Under both the outgoing Directives (e.g., MDD 93/42/EEC) and the new regulations (e.g., MDR 2017/745), whenever a chosen conformity assessment route requires Notified Body (NB) assessment of the quality management system (QMS), it is always required to be paired with NB assessment of the Technical Documentation (TD).  This always results in two certificates issued in order to gain Notified Body (NB) approval for placing the corresponding devices on the market: One certificate for the quality management system conformity, and one certificate for the device conformity. This two-certificate ensemble is a most fundamental certification requirement that is core to any successful and real CE marking experience where the manufacturer chooses a route involving QMS conformity.

For example, if the Article 52(6) first paragraph Class IIa conformity assessment route is chosen, then, by definition, that route involves two types of conformity assessments and certificates by the NB: One for the QMS plus one for the device TD. This is the way it's always been under the outgoing Directives (e.g., MDD 93/42/EEC), and this has been maintained under the new regulations (e.g., MDR 2017/745).  Specifically, for example under the MDR per Article 52(6) first paragraph, the Annex IX Chapters I & III QMS conformity assessment is performed by the NB resulting in an EU QMS assessment certificate per Annex IX section 2.3 fourth paragraph PLUS an Annex IX Chapter II section 4 TD assessment is performed by the NB resulting in an EU TD assessment certificate per Annex IX section 4.9.

As further examples, the same type of two-certificate requirement applies to the Article 52(6) second paragraph Class IIa conformity assessment route (and others for other device classes).

------------------------------
Kevin Randall, ASQ CQA, RAC (Europe, U.S., Canada)
Principal Consultant
Ridgway, CO
United States
© Copyright by ComplianceAcuity, Inc. All rights reserved.

Original Message:
Sent: 23-Jan-2024 04:11
From: Christoph Kiesselbach
Subject: CE design certificate

Hi Anon,

as Kevin says, the MDR does not use the expression "design certificate".

Annex XII that deals with the certificates issued by a notified body lists: "EU technical documentation assessment certificates, EU type-examination certificates and EU product verification certificates" in section 4(a) and "EU quality management system certificates and EU quality assurance certificates" in section 4(b). You could argue that the former are to some degree "device certificates", because they refer to an individual technical documentation, device type or even single devices, while the latter are "quality management certificates", but there is no such formulation in the MDR.

For a class IIa SaMD you usually would need either a "EU quality management system certificate" (Annex IX, chapters I and III) or a "EU quality assurence certificate" (Annex XI, Part A, Section 10). Both procedures would include the assessment of a technical documentation for at least one representative device sample for each device category, but as far as I know this assessment does not result in a separate certificate (as opposed to e.g. the full Annex IX assessment for class III devices or a type examination according to Annex X). If you only have one device, then of course the sampling in the end means that every device is assessed this way, but again this to my knowledge is not reflected in its own certificate. In theory Annex XI, Part B, Section 18, would be a possible procedure, but testing each device does not make sense for software and I do not think any notified body has this in their scope.

So in a loose sense, you do not obtain a design certificate for a class IIa software, but a quality management certificate (either full quality management system or product assurance). However, using the language of the MDR in describing the certificate is the better idea.

Best regards, Christoph

------------------------------
Christoph Kiesselbach
Schrack & Partner
Reutlingen
Germany

Original Message:
Sent: 21-Jan-2024 07:29
From: Anonymous Member
Subject: CE design certificate

This message was posted by a user wishing to remain anonymous

Dear RAPS Community,

What is CE design certificate and it is mandatory to obtain from NB for class IIa device (SaMD)?

Thank you for any info!

Hi all,

Thanks for the discussion - took me a while to read through everything :) but great reading.  I still just shake my head at the bumbling of the EU Commission for something as simple as providing direction - 1 certificate or 2 certificates?  When I was "growing up" in the EU MDD world, I was so used to 2 certificates.  In fact, discussions with Notified Bodies (NB) back then was the approach they used because many companies would get their QMS certification first (ISO 13485) and then their CE Certificate sometime later, sometimes even months or a year later.  It especially helped companies which were bringing new products to the market because they could get their QMS sorted, with design controls and manufacturing, and then submit the Technical File (including ye 'ole Design Dossier) sometime later when they were ready.

It appears for those Class I (NB reviewed ones), Class IIa, and some Class IIb, the single certificate approach is being used more commonly.  I recently saw a EU MDR certificate issued by TUV SUD which was a single certificate for the QMS (Titled as: EU Quality Management System Certificate (MDR) Pursuant to Regulation (EU) 2017/745 on Medical Devices, Annex IX Chapters I and III (Class IIa and Class IIb Devices) and then a following page with the classification and Device Group (the EMDN designation).  Sadly, they still can not even get those right because they messed up the dates when the certification was done.  And more importantly their own certificate said Intended Purpose which was blank ... :facepalm:.  This was a Class II SaMD software only product.  What I found boggling and mind-blowing is the QMS/CE Certificate had no Device Name, device designation such as part number, and no version indication - being a software only product.  I have no idea how companies are then going to use their "CE Certificate" to leverage registration in other countries which accepts CE mark when the "CE Certificate" does not even say the product name at all.

Hello Christoph,

I was just informed by DQS that they are doing away with the two-certificate approach.  In other words, they are adjusting their practices to instead align with what we saw the other NBs doing.  Specifically, they are adjusting so as to not issue both an EU Quality Management System Certificate plus an EU Technical Documentation Assessment Certificate.  Rather, it will be as you and Roger were saying (one type of certificate or the other but not both).

Yet the frequent issuance of a supporting QMS certificate is still in place for what is recognized to be the most common scenario where the manufacturer elects to formally employ certification to a standard like EN ISO 13485 as its means for meeting the MDR's Article 10(9) and Annex IX quality management system requirements such as in this public example here.

Thank you for your persistence on this topic.

Thanks Kevin for reaching out to DQS, and please note the edit of my previous post: DQS seems to have been consistent and just have clarified the design of their certificates.

If you approach them: I would find it interesting if they consider Annex IX, Section 4.10 with respect to changes applicable for class IIa/IIb non-implantable devices, and if yes how they are handling this in case of sampling

Best regards, Christoph.

Thank you, Christoph for remembering the MDCG 2019-3 narrative. I would say the reason DQS is currently using the two-certificate approach is because they are implementing the legislation in the way it is clearly and unambiguously written rather than succumbing to group-think or being overwhelmed by intensive reinterpretations.  MDCG 2019-3 is quite dated by now, having been written well before the actual application of the MDR; so perhaps actual experience has influenced DQS.  But ultimately, NBs (as do anyone) have the liberty to ignore guidance because guidance isn't legislation.  I think it would be interesting to hear from DQS about assertions that they are somehow out of bounds; I'll reach out to them for further explanation.

Thank you both for the additional input. I knew I had seen a more restricted version of that reference to section 4 and that was indeed in Annex IX 2.3/3.5 before the EU aligned it with Article 52 (4)/(6) in the first Corrigendum, choosing with uncanny accuracy the more ambiguous version.

Something else I had forgotten was that MDCG 2019-13 talks about TD sampling and with respect to applicability of Chapter II, Section 4 of Annex IX it says in section 5.2:

"Taking into account the wording of articles 52(4) and 52(6) of the MDR and Article 48(7) and 48(9) of the IVDR, as combined with annexes VII and IX, the
tasks to be carried out by the notified body as part of the conformity assessment activities described in Chapter II, Section 4 of Annex IX comprise the complete review of the technical documentation in accordance with Annexes II and III.

In addition, the manufacturer will grant access to the technical documentation as referred to in Section 2.2 of Annex IX and the notified body will provide the manufacturer with a report on the technical documentation assessment. For Class IIa / IIb and Class B / Class C the notified body will neither require an application nor issue an EU technical documentation assessment certificate (see 5.3 for exceptions)."

Section 5.3 then for the MDR adds that for implantable IIb devices a respective certificat is issued.

The guidance came out in December 2019, some months after the first Corrigendum, so I assume they took the changed wording into account and recognized the need to clarify this. I just did not remember that this was clarified in a MDCG Guidance. I cannot rule out that somebody at the EU might interpret this differently, and maybe the general interpretation changed, but at the moment I would consider this to be highly unlikely.

Why DQS sometime in 2022 decided to ignore that guidance (there are some earlier DQS certificates, that handle this differently, see here),  remains a mystery to me. At the moment I have not seen any other NB handing out an Annex IX Chapter II certificate for class IIa/IIb non implantable, but if anyone has other examples, this would be interesting.

Anyway, so much for "crystal clear and unambiguous" formulations in the MDR - confusion everywhere, indeed.

Best regards, Christoph

(edited to include a link to MDCG 2019-13 for convenience)

------------------------------
Christoph Kiesselbach
Schrack & Partner
Reutlingen
Germany

Original Message:
Sent: 26-Jan-2024 12:58
From: Kevin Randall
Subject: CE design certificate

Thank you for your comedy relief of CE standing for "Confusion Everywhere". :)

------------------------------
Kevin Randall, ASQ CQA, RAC (Europe, U.S., Canada)
Principal Consultant
Ridgway, CO
United States
© Copyright by ComplianceAcuity, Inc. All rights reserved.

Original Message:
Sent: 26-Jan-2024 12:51
From: Roger Gray
Subject: CE design certificate

Thanks for the additional information regarding your interaction with the Commission, Kevin, and for reminder that the change to Annex IX Section 3.5 was made in Corrigendum 1..

Unfortunately, the Commission has made the classic error of making one change without understanding the consequences elsewhere in the document. Even with this revision to Annex IX Section 3.5, Section 4 still introduces itself by stating that it applies only to Class III and certain Class II devices, when it no longer does. Clearly, this should have been amended at the same time as the change to Section 3.5. Only then would the current wording be "crystal clear and unambiguous". What is the point of having a section stating that it applies only to Class III and certain Class IIb devices, when it now apparently applies equally to Class IIa and all Class IIb?

It seems to me that the original intention of the MDR was to maintain the single QMS certificate approach of the MDD for Class IIa devices, but since initial publication, it was decided to make a very significant change, but then not implement it properly.

Well it was always said that CE stood for Confusion Everywhere!

------------------------------
Roger Gray
VP Quality and Regulatory
Donawa Lifescience
Clinical Studies - Regulatory - Quality Systems
AR/UKRP services
Italy and UK
rgray@donawa.com
www.donawa.com

Original Message:
Sent: 26-Jan-2024 11:19
From: Kevin Randall
Subject: CE design certificate

Thank you, Roger for your additional insights.  When I pressed the Commission about this early on in the MDR's tenure, the Directorate-General (DG) for Health and Food Safety maintained that the potential contradiction ("Section 4.4 to 4.8" vs. "Section 4) was corrected by Corrigendum, OJ L 117, 3.5.2019, p. 9 (2017/745).  The DG maintained that now, the text of Annex IX, Section 2.3, third subparagraph [and Section 3.5], is "Moreover, in the case of class IIa and class IIb devices, the quality management system assessment shall be accompanied by the assessment of technical documentation for devices selected on a representative basis as specified in Section 4", therefore including the assessment of the entire technical documentation.

I find it interesting that when I pressed the DG on this, the DG still maintained the MDR's very clear and unambiguous requirement for Class IIa device Annex IX Section 4 TD review and TD certification.

Accordingly, I'm not sure this is an issue of "interpretation", since the MDR is so crystal clear and unambiguous about the aforesaid two-certificate approach when the Class IIa Annex IX route is chosen.  Instead, I'd say that if we claim a one-certificate approach, then it is instead a decision to ignore and/or disagree with this very unambiguous MDR requirement that the DG maintained when I pressed.

Maybe there's merit, maybe there's not, for arguing against what the MDR so clearly says in this case.  To know for sure whether there is merit for such opposition, then we need to make a case using something other than claiming an issue of interpretation.  Specifically, we need to make a case explaining why Class IIa devices employing the Annex IX route (traditionally called the "full quality assurance" route) shouldn't be subject to the very basic TD controls of Section 4.  In so doing, we would need to acknowledge that the "design dossier" emphasis of Section 4 from the MDD has been generalized into a more generic/holistic TD assessment rather than just a design dossier review.

I think we need to be sensitive to the possibility that the Commission did this intentionally in order to prescribe additional substance to the assessment of a Class IIa full QMS TD assessment.  Indeed, that same additional substance principle existed in the MDD at Annex II.7, though there it excluded Section 4 because Section 4 was indeed at that time reserved for the design dossier approach.  But as mentioned, that has now been generalized into a more holistic TD assessment providing for either the traditional design dossier level of review, but also to allow use for a more substantive TD assessment for devices (i.e., Class IIa full QMS devices) where some additional substance is of value for public health.  The MDR does after all intend to increase assurance of safety and performance relative to the MDD.

And I don't' think we can make a valid argument against a regulation's multi-purposing of existing legislative infrastructure like when multi-purposing Section 4 for Class IIa devices.  Regulations routinely do that kind of thing in order to avoid replicating/duplicating language in multiple places.  Indeed, incorporation by reference is a commonplace tactic used when writing regulations.

Moreover, the basic title of Annex IX is "Conformity Assessment Based on Quality Management System and On Assessment of Technical Documentation", not, "or Assessment of Technical Documentation".  Accordingly, while we may argue against the inconvenience of issuing two certificates, I'm not yet seeing grounds for arguing against Section 4's additional parameters for TD assessment of Class IIa devices undergoing the full QMS route.

Ultimately again, if we aim to ignore and/or disagree with the use of the Section 4 substance for a Class IIa device full QMS TD assessment, then we need to make a case for why that additional substance is unnecessary.  I haven't heard a good argument for that yet, nor can I think of one myself so far.

------------------------------
Kevin Randall, ASQ CQA, RAC (Europe, U.S., Canada)
Principal Consultant
Ridgway, CO
United States
© Copyright by ComplianceAcuity, Inc. All rights reserved.

Original Message:
Sent: 26-Jan-2024 08:05
From: Roger Gray
Subject: CE design certificate

I know we've gone some way from answering the initial question, but I thought I would just make a couple of comments.

I agree with Christoph's interpretation of both MDD and MDR, in that you generally only need a QMS certificate (MDD Annex II/MDR Annex IX Parts I and III) unless you have a Class III device, or there is a special category of Class IIb device involved.

It seems to me that the current confusion is caused by an ambiguity in the MDR (what a surprise!). If you read MDR Annex IX para 3.5, it says:

"In the case of class IIa and class IIb devices, the surveillance assessment shall also include an assessment of the technical documentation as specified in Section 4 for the device or devices concerned on the basis of further representative samples chosen in accordance with the rationale documented by the notified body in accordance with the third paragraph of Section 2.3."

The only problem is that when you go to Section 4, it is apparent that this entire section applies only to Class III and certain Class IIb devices, as it states in the very first line:

"Assessment of the technical documentation applicable to class III devices and to the class IIb devices referred to in the second subparagraph of Article 52(4)."

So, unfortunately, the referral to Annex IX Section 4 in Section 3.5 does not make sense, as Section 4 apparently applies ONLY to Class III and certain Class IIb devices, meaning that it has no application to Class IIa devices. There is literally zero mention of 'Class IIa' anywhere in Section 4, so I am very confused by DQS issuing an 'Annex IX Chapter II' certificate for a Class IIa device, citing "Conformity assessment based on a Quality Management System and an Assessment of Technical Documentation". 

My interpretation of the MDR is that for a Class IIa device "Conformity assessment based on a Quality Management System and an Assessment of Technical Documentation". would require either:

• Annex IX Parts I and III
• Annex XI Section 10

Note that while the use of Annex IX Parts I and III would result in just one 'Quality Management System' certificate from the notified body, using Annex XI Section 10 would result in two certificates from the notified body, these being a 'Quality Assurance Certificate' plus an 'Annex XI Part A Technical Documentation Certificate'.

Just for fun, I went back to the final draft of the MDR pre-publication, copy dated March 2017, and this is what Annex IX Section 3.5 stated at that time:

"Moreover, in the case of class IIa and class IIb devices, the quality management system assessment shall be accompanied by the assessment of technical documentation for devices selected on a representative basis in accordance with Sections 4.4 to 4.8."

If this wording had been maintained, then the reference to Section 4 would have made sense, because if you just apply Sections 4.4 to 4.8 to Class IIa devices, you completely avoid the possible need for a Technical Documentation Certificate, because this is mentioned only in Section 4.9.

So not for the first time, in moving from an earlier version to a later version, the authoring team have created an ambiguity where one didn't previously exist, thus causing us all these interpretational headaches.

------------------------------
Roger Gray
VP Quality and Regulatory
Donawa Lifescience
Clinical Studies - Regulatory - Quality Systems
AR/UKRP services
Italy and UK
rgray@donawa.com
www.donawa.com

Original Message:
Sent: 25-Jan-2024 01:56
From: Christoph Kiesselbach
Subject: CE design certificate

Hi Kevin,

yes, I think I got a little bit carried away by the topic. Thank you for discussing this, it definitely brought up some points I would like to try to clarify.

I will do some research if there is some kind of consensus between NBs or an MDCG opinion on this and see if it makes sense to post the results in a separate thread.

Best regards, Christoph

------------------------------
Christoph Kiesselbach
Schrack & Partner
Reutlingen
Germany

Original Message:
Sent: 24-Jan-2024 11:47
From: Kevin Randall
Subject: CE design certificate

Thank you, Christoph. NBOG-3 (addressing certificate practices for the outgoing MDD) doesn't directly foresee a two-certificate approach yet does so indirectly in my experience.  Specifically, if the manufacturer, like most do, has chosen to employ an ISO standard as its means for meeting the Annex II Full Quality Assurance QMS requirements, then there is required to be a separate QMS certificate that is the basis for the Annex II Full Quality Assurance EC Certificate.

Also of note is NB-MED/2.5.1/Rec4 stating that the matter of interlinking of certificates has been considered and is recognized to be an administrative responsibility of the NB and should be traceable through the file of the NB. Inclusion of a cross-reference on the certificates themselves, however, would lead to practical difficulties.  Thus, the lack of such cross-references (like in the MDD certificates your cited) certainly doesn't necessarily mean that the certificate stands on its own.

Regarding the various additional questions you've raised (i.e., handling of significant changes for class IIa devices; NB practices for TD sampling after significant changes), those seem to be well beyond the scope of Anon's question.  I certainly have answers for you on those peripheral questions; but I think it's most appropriate to raise them in a separate threat so that this current thread can stay on track with Anon's intended question.

------------------------------
Kevin Randall, ASQ CQA, RAC (Europe, U.S., Canada)
Principal Consultant
Ridgway, CO
United States
© Copyright by ComplianceAcuity, Inc. All rights reserved.

Original Message:
Sent: 24-Jan-2024 03:48
From: Christoph Kiesselbach
Subject: CE design certificate

Hi Kevin,

thank you for providing the examples, that was definitely something I had not realised before. Just for others reading this here are two examples for publicly available current MDR certificates by the mentioned TÜV Süd and BSI that either declare that TD assessment is part of this certificate or that an Annex IX Chapter II certificate is only required for class III and IIb implantable devices (where every TD must be assessed).

With respect to the MDD: At least the old NBOG Best Practice Guide (2010-3) does not foresee a combination of Annex II excluding (4) with Annex II (4) ( for class IIa and IIb, but exclusively for class III. And in the corresponding MDD certificates to the examples you provided DQS used the singe-certificate approach under MDD (see here and here, where an additional Annex II section 4 certificate is limited to class III devices).

Obviously there is some freedom in the interpretaton of the MDR and the corresponding use of certificates. However I would really prefer if there was a common approach by the NBs. I will try to reach out to Team NB and see if they could imagine a common approach for their members.

But aside from the number of certificates, the main impact if Annex IX Chapter II is fully applicable to class IIa and IIb non-implantable devices (aside from the sampling aspect) in my opinion would be that this would also mean that significant design changes must be approved by the Notified Body and lead to a supplement to the (then existing) TD-certificate (of course also under the single-certificate approach each NB might have additional requirements on what needs to be reported by the manufacturer that might be more extensive, but I still think this would be different from a legal requirement with mandatory consequences for the certificates). At least in my experience that was handled differently under MDD for Annex II excluding (4) (at least if you go with the single-certificate approach and the conformity assessment routes given in NBOG 2010-3, see NBOG 2014-3).

There currently is no MDCG guidance on the interpretation of the need for approval of significant changes (with the exception of the specific case of Article 120), so their position is currently not clear to me. But how would this be handled by the NB in the case of TD sampling? If the manufacturer makes a significant change to a device within a group with a TD that has not yet been assessed as a sample, how does the NB assess the change? Based on the change notification only without any knowledge about the previous state of the TD? Do they perform an additional full TD assessment for this device? Or is the assessment only necessary if the TD already had been sampled? Maybe I am wrong but I think this would require a fair amount of additional resources and time, both currently in rather short supply.

Best regards, Christoph

------------------------------
Christoph Kiesselbach
Schrack & Partner
Reutlingen
Germany

Original Message:
Sent: 23-Jan-2024 17:00
From: Kevin Randall
Subject: CE design certificate

Christoph,

I can't show actual clients' examples; but I'll be creative to show a couple Class IIa examples that are publicly/readily available among others on the internet:

• Example 1
• Example 2

Therein on these Class IIa device Annex IX Chapter II EU Technical Documentation Assessment Certificates is stated, "For placing devices listed in the Annex on the market, an additional certificate according to Annex IX, Chapter I and III is required."

Along these lines during my 28-year tenure with medical device CE marking, I've seen many instances of the MDD Article 11(2) second paragraph Class IIa Annex II full quality assurance route generating a QMS certificate plus an EC certificate for the chosen assessment route through the most well-known NBs.

All this said, your insights have educated me to the fact that there does seem to be more NBs employing the single-certificate route for Class IIa devices. For example, it hadn't yet dawned on me that BSI and TUV seem to be sticking to that single-certificate approach for Class IIa Annex IX devices.

My understanding is that the NBs have freedom regarding the particular documented format/way in which they answer the aforesaid two separate, distinct, concomitant Class IIa Annex IX certificate requirements (QMS certificate + TD certificate).  Thus, this liberty may be why we are indeed seeing some use a single-certificate Class IIa Annex IX approach where that one certificate is in fact entitled "EU Quality Management System Assessment Certificate".  But I don't believe that is in clear alignment with the aforesaid two-certificate requirement.  Given how much trouble NBs have had in recent years issuing timely certificates, perhaps they are stretching this issue to reduce the administrative burden rather than following the approach in the examples above which seems to more clearly match the intent of the MDR.

------------------------------
Kevin Randall, ASQ CQA, RAC (Europe, U.S., Canada)
Principal Consultant
Ridgway, CO
United States
© Copyright by ComplianceAcuity, Inc. All rights reserved.

Original Message:
Sent: 23-Jan-2024 12:22
From: Christoph Kiesselbach
Subject: CE design certificate

Hi Kevin,

I respectfully disagree.

Under the MDD e.g. Annex II section 7 made it quite clear that for class IIa and IIb devices section 4 of Annex II (that contains provisions for design examination and a corresponding certificate) does not apply and instead TDs are sampled as part of the audit procedure (Section 3.3 of Annex II). The certificates for the quality management system (Annex II without section 4) that I know then stated that the validity for class III devices would depend on an additional design examination certificate (Annex II section 4) but I know of no additional certificate for class IIa/IIb devices where the TDs were sampled.

The MDR (of course) makes things less clear, but I think a not unusual interpretation of the additional TD assessment of samples in the case of IIa and IIb devices for e.g. Annex IX is that the assessment again is done as part of the audit activities (Chapter I of Annex IX, where also the additional TD sampling during surveillance in described) and while it is generally performed as described in Section 4 of Annex IX, the results are a part of the Annex IX chapter I (and III) certificate. Only class III and class IIb implantable devices where every TD is assessed receive an additional certificate in line with Section 4.9 and are subject to the approval of changes detailed in Section 4.10. I also think this makes sense: Only if every TD is assessed the reporting and approval of changes makes sense. If only samples are assessed, then what would the "approved device" in section 4.10 refer to? The whole generic device group / category of devices that the sample stands for or only the TDs that happen to have been sampled? If the NB does not know the TD the manufacturer is changing (because it has not yet been sampled), how are they assumed to assess changes? Would they have to do an additional full TD assessment?

And so far the MDR certificates I have seen are comparable to the ones under MDD, a certificate according to Annex IX Chapter I (and III) with the additional remark that in case of class III and IIb implantable devices an additional Annex IX Section 4 (or Chapter II) certificate is required (but not for class IIa/IIb).

However, I admit this is less clearly expressed than it was under the MDD and there might be NBs that interpret this differently and hand out Annex IX Chapter II certificates for class IIa/IIb (non implantable) devices. If you have examples for this, I would be interested to know them.

Best regards, Christoph

------------------------------
Christoph Kiesselbach
Schrack & Partner
Reutlingen
Germany

Original Message:
Sent: 23-Jan-2024 09:27
From: Kevin Randall
Subject: CE design certificate

Under both the outgoing Directives (e.g., MDD 93/42/EEC) and the new regulations (e.g., MDR 2017/745), whenever a chosen conformity assessment route requires Notified Body (NB) assessment of the quality management system (QMS), it is always required to be paired with NB assessment of the Technical Documentation (TD).  This always results in two certificates issued in order to gain Notified Body (NB) approval for placing the corresponding devices on the market: One certificate for the quality management system conformity, and one certificate for the device conformity. This two-certificate ensemble is a most fundamental certification requirement that is core to any successful and real CE marking experience where the manufacturer chooses a route involving QMS conformity.

For example, if the Article 52(6) first paragraph Class IIa conformity assessment route is chosen, then, by definition, that route involves two types of conformity assessments and certificates by the NB: One for the QMS plus one for the device TD. This is the way it's always been under the outgoing Directives (e.g., MDD 93/42/EEC), and this has been maintained under the new regulations (e.g., MDR 2017/745).  Specifically, for example under the MDR per Article 52(6) first paragraph, the Annex IX Chapters I & III QMS conformity assessment is performed by the NB resulting in an EU QMS assessment certificate per Annex IX section 2.3 fourth paragraph PLUS an Annex IX Chapter II section 4 TD assessment is performed by the NB resulting in an EU TD assessment certificate per Annex IX section 4.9.

As further examples, the same type of two-certificate requirement applies to the Article 52(6) second paragraph Class IIa conformity assessment route (and others for other device classes).

------------------------------
Kevin Randall, ASQ CQA, RAC (Europe, U.S., Canada)
Principal Consultant
Ridgway, CO
United States
© Copyright by ComplianceAcuity, Inc. All rights reserved.

Original Message:
Sent: 23-Jan-2024 04:11
From: Christoph Kiesselbach
Subject: CE design certificate

Hi Anon,

as Kevin says, the MDR does not use the expression "design certificate".

Annex XII that deals with the certificates issued by a notified body lists: "EU technical documentation assessment certificates, EU type-examination certificates and EU product verification certificates" in section 4(a) and "EU quality management system certificates and EU quality assurance certificates" in section 4(b). You could argue that the former are to some degree "device certificates", because they refer to an individual technical documentation, device type or even single devices, while the latter are "quality management certificates", but there is no such formulation in the MDR.

For a class IIa SaMD you usually would need either a "EU quality management system certificate" (Annex IX, chapters I and III) or a "EU quality assurence certificate" (Annex XI, Part A, Section 10). Both procedures would include the assessment of a technical documentation for at least one representative device sample for each device category, but as far as I know this assessment does not result in a separate certificate (as opposed to e.g. the full Annex IX assessment for class III devices or a type examination according to Annex X). If you only have one device, then of course the sampling in the end means that every device is assessed this way, but again this to my knowledge is not reflected in its own certificate. In theory Annex XI, Part B, Section 18, would be a possible procedure, but testing each device does not make sense for software and I do not think any notified body has this in their scope.

So in a loose sense, you do not obtain a design certificate for a class IIa software, but a quality management certificate (either full quality management system or product assurance). However, using the language of the MDR in describing the certificate is the better idea.

Best regards, Christoph

------------------------------
Christoph Kiesselbach
Schrack & Partner
Reutlingen
Germany

Original Message:
Sent: 21-Jan-2024 07:29
From: Anonymous Member
Subject: CE design certificate

This message was posted by a user wishing to remain anonymous

Dear RAPS Community,

What is CE design certificate and it is mandatory to obtain from NB for class IIa device (SaMD)?

Thank you for any info!

Hello Christoph,

I was just informed by DQS that they are doing away with the two-certificate approach.  In other words, they are adjusting their practices to instead align with what we saw the other NBs doing.  Specifically, they are adjusting so as to not issue both an EU Quality Management System Certificate plus an EU Technical Documentation Assessment Certificate.  Rather, it will be as you and Roger were saying (one type of certificate or the other but not both).

Yet the frequent issuance of a supporting QMS certificate is still in place for what is recognized to be the most common scenario where the manufacturer elects to formally employ certification to a standard like EN ISO 13485 as its means for meeting the MDR's Article 10(9) and Annex IX quality management system requirements such as in this public example here.

Thank you for your persistence on this topic.

Thanks Kevin for reaching out to DQS, and please note the edit of my previous post: DQS seems to have been consistent and just have clarified the design of their certificates.

If you approach them: I would find it interesting if they consider Annex IX, Section 4.10 with respect to changes applicable for class IIa/IIb non-implantable devices, and if yes how they are handling this in case of sampling

Best regards, Christoph.

Thank you, Christoph for remembering the MDCG 2019-3 narrative. I would say the reason DQS is currently using the two-certificate approach is because they are implementing the legislation in the way it is clearly and unambiguously written rather than succumbing to group-think or being overwhelmed by intensive reinterpretations.  MDCG 2019-3 is quite dated by now, having been written well before the actual application of the MDR; so perhaps actual experience has influenced DQS.  But ultimately, NBs (as do anyone) have the liberty to ignore guidance because guidance isn't legislation.  I think it would be interesting to hear from DQS about assertions that they are somehow out of bounds; I'll reach out to them for further explanation.

Thank you both for the additional input. I knew I had seen a more restricted version of that reference to section 4 and that was indeed in Annex IX 2.3/3.5 before the EU aligned it with Article 52 (4)/(6) in the first Corrigendum, choosing with uncanny accuracy the more ambiguous version.

Something else I had forgotten was that MDCG 2019-13 talks about TD sampling and with respect to applicability of Chapter II, Section 4 of Annex IX it says in section 5.2:

"Taking into account the wording of articles 52(4) and 52(6) of the MDR and Article 48(7) and 48(9) of the IVDR, as combined with annexes VII and IX, the
tasks to be carried out by the notified body as part of the conformity assessment activities described in Chapter II, Section 4 of Annex IX comprise the complete review of the technical documentation in accordance with Annexes II and III.

In addition, the manufacturer will grant access to the technical documentation as referred to in Section 2.2 of Annex IX and the notified body will provide the manufacturer with a report on the technical documentation assessment. For Class IIa / IIb and Class B / Class C the notified body will neither require an application nor issue an EU technical documentation assessment certificate (see 5.3 for exceptions)."

Section 5.3 then for the MDR adds that for implantable IIb devices a respective certificat is issued.

The guidance came out in December 2019, some months after the first Corrigendum, so I assume they took the changed wording into account and recognized the need to clarify this. I just did not remember that this was clarified in a MDCG Guidance. I cannot rule out that somebody at the EU might interpret this differently, and maybe the general interpretation changed, but at the moment I would consider this to be highly unlikely.

Why DQS sometime in 2022 decided to ignore that guidance (there are some earlier DQS certificates, that handle this differently, see here),  remains a mystery to me. At the moment I have not seen any other NB handing out an Annex IX Chapter II certificate for class IIa/IIb non implantable, but if anyone has other examples, this would be interesting.

Anyway, so much for "crystal clear and unambiguous" formulations in the MDR - confusion everywhere, indeed.

Best regards, Christoph

(edited to include a link to MDCG 2019-13 for convenience)

------------------------------
Christoph Kiesselbach
Schrack & Partner
Reutlingen
Germany

Original Message:
Sent: 26-Jan-2024 12:58
From: Kevin Randall
Subject: CE design certificate

Thank you for your comedy relief of CE standing for "Confusion Everywhere". :)

------------------------------
Kevin Randall, ASQ CQA, RAC (Europe, U.S., Canada)
Principal Consultant
Ridgway, CO
United States
© Copyright by ComplianceAcuity, Inc. All rights reserved.

Original Message:
Sent: 26-Jan-2024 12:51
From: Roger Gray
Subject: CE design certificate

Thanks for the additional information regarding your interaction with the Commission, Kevin, and for reminder that the change to Annex IX Section 3.5 was made in Corrigendum 1..

Unfortunately, the Commission has made the classic error of making one change without understanding the consequences elsewhere in the document. Even with this revision to Annex IX Section 3.5, Section 4 still introduces itself by stating that it applies only to Class III and certain Class II devices, when it no longer does. Clearly, this should have been amended at the same time as the change to Section 3.5. Only then would the current wording be "crystal clear and unambiguous". What is the point of having a section stating that it applies only to Class III and certain Class IIb devices, when it now apparently applies equally to Class IIa and all Class IIb?

It seems to me that the original intention of the MDR was to maintain the single QMS certificate approach of the MDD for Class IIa devices, but since initial publication, it was decided to make a very significant change, but then not implement it properly.

Well it was always said that CE stood for Confusion Everywhere!

------------------------------
Roger Gray
VP Quality and Regulatory
Donawa Lifescience
Clinical Studies - Regulatory - Quality Systems
AR/UKRP services
Italy and UK
rgray@donawa.com
www.donawa.com

Original Message:
Sent: 26-Jan-2024 11:19
From: Kevin Randall
Subject: CE design certificate

Thank you, Roger for your additional insights.  When I pressed the Commission about this early on in the MDR's tenure, the Directorate-General (DG) for Health and Food Safety maintained that the potential contradiction ("Section 4.4 to 4.8" vs. "Section 4) was corrected by Corrigendum, OJ L 117, 3.5.2019, p. 9 (2017/745).  The DG maintained that now, the text of Annex IX, Section 2.3, third subparagraph [and Section 3.5], is "Moreover, in the case of class IIa and class IIb devices, the quality management system assessment shall be accompanied by the assessment of technical documentation for devices selected on a representative basis as specified in Section 4", therefore including the assessment of the entire technical documentation.

I find it interesting that when I pressed the DG on this, the DG still maintained the MDR's very clear and unambiguous requirement for Class IIa device Annex IX Section 4 TD review and TD certification.

Accordingly, I'm not sure this is an issue of "interpretation", since the MDR is so crystal clear and unambiguous about the aforesaid two-certificate approach when the Class IIa Annex IX route is chosen.  Instead, I'd say that if we claim a one-certificate approach, then it is instead a decision to ignore and/or disagree with this very unambiguous MDR requirement that the DG maintained when I pressed.

Maybe there's merit, maybe there's not, for arguing against what the MDR so clearly says in this case.  To know for sure whether there is merit for such opposition, then we need to make a case using something other than claiming an issue of interpretation.  Specifically, we need to make a case explaining why Class IIa devices employing the Annex IX route (traditionally called the "full quality assurance" route) shouldn't be subject to the very basic TD controls of Section 4.  In so doing, we would need to acknowledge that the "design dossier" emphasis of Section 4 from the MDD has been generalized into a more generic/holistic TD assessment rather than just a design dossier review.

I think we need to be sensitive to the possibility that the Commission did this intentionally in order to prescribe additional substance to the assessment of a Class IIa full QMS TD assessment.  Indeed, that same additional substance principle existed in the MDD at Annex II.7, though there it excluded Section 4 because Section 4 was indeed at that time reserved for the design dossier approach.  But as mentioned, that has now been generalized into a more holistic TD assessment providing for either the traditional design dossier level of review, but also to allow use for a more substantive TD assessment for devices (i.e., Class IIa full QMS devices) where some additional substance is of value for public health.  The MDR does after all intend to increase assurance of safety and performance relative to the MDD.

And I don't' think we can make a valid argument against a regulation's multi-purposing of existing legislative infrastructure like when multi-purposing Section 4 for Class IIa devices.  Regulations routinely do that kind of thing in order to avoid replicating/duplicating language in multiple places.  Indeed, incorporation by reference is a commonplace tactic used when writing regulations.

Moreover, the basic title of Annex IX is "Conformity Assessment Based on Quality Management System and On Assessment of Technical Documentation", not, "or Assessment of Technical Documentation".  Accordingly, while we may argue against the inconvenience of issuing two certificates, I'm not yet seeing grounds for arguing against Section 4's additional parameters for TD assessment of Class IIa devices undergoing the full QMS route.

Ultimately again, if we aim to ignore and/or disagree with the use of the Section 4 substance for a Class IIa device full QMS TD assessment, then we need to make a case for why that additional substance is unnecessary.  I haven't heard a good argument for that yet, nor can I think of one myself so far.

------------------------------
Kevin Randall, ASQ CQA, RAC (Europe, U.S., Canada)
Principal Consultant
Ridgway, CO
United States
© Copyright by ComplianceAcuity, Inc. All rights reserved.

Original Message:
Sent: 26-Jan-2024 08:05
From: Roger Gray
Subject: CE design certificate

I know we've gone some way from answering the initial question, but I thought I would just make a couple of comments.

I agree with Christoph's interpretation of both MDD and MDR, in that you generally only need a QMS certificate (MDD Annex II/MDR Annex IX Parts I and III) unless you have a Class III device, or there is a special category of Class IIb device involved.

It seems to me that the current confusion is caused by an ambiguity in the MDR (what a surprise!). If you read MDR Annex IX para 3.5, it says:

"In the case of class IIa and class IIb devices, the surveillance assessment shall also include an assessment of the technical documentation as specified in Section 4 for the device or devices concerned on the basis of further representative samples chosen in accordance with the rationale documented by the notified body in accordance with the third paragraph of Section 2.3."

The only problem is that when you go to Section 4, it is apparent that this entire section applies only to Class III and certain Class IIb devices, as it states in the very first line:

"Assessment of the technical documentation applicable to class III devices and to the class IIb devices referred to in the second subparagraph of Article 52(4)."

So, unfortunately, the referral to Annex IX Section 4 in Section 3.5 does not make sense, as Section 4 apparently applies ONLY to Class III and certain Class IIb devices, meaning that it has no application to Class IIa devices. There is literally zero mention of 'Class IIa' anywhere in Section 4, so I am very confused by DQS issuing an 'Annex IX Chapter II' certificate for a Class IIa device, citing "Conformity assessment based on a Quality Management System and an Assessment of Technical Documentation". 

My interpretation of the MDR is that for a Class IIa device "Conformity assessment based on a Quality Management System and an Assessment of Technical Documentation". would require either:

• Annex IX Parts I and III
• Annex XI Section 10

Note that while the use of Annex IX Parts I and III would result in just one 'Quality Management System' certificate from the notified body, using Annex XI Section 10 would result in two certificates from the notified body, these being a 'Quality Assurance Certificate' plus an 'Annex XI Part A Technical Documentation Certificate'.

Just for fun, I went back to the final draft of the MDR pre-publication, copy dated March 2017, and this is what Annex IX Section 3.5 stated at that time:

"Moreover, in the case of class IIa and class IIb devices, the quality management system assessment shall be accompanied by the assessment of technical documentation for devices selected on a representative basis in accordance with Sections 4.4 to 4.8."

If this wording had been maintained, then the reference to Section 4 would have made sense, because if you just apply Sections 4.4 to 4.8 to Class IIa devices, you completely avoid the possible need for a Technical Documentation Certificate, because this is mentioned only in Section 4.9.

So not for the first time, in moving from an earlier version to a later version, the authoring team have created an ambiguity where one didn't previously exist, thus causing us all these interpretational headaches.

------------------------------
Roger Gray
VP Quality and Regulatory
Donawa Lifescience
Clinical Studies - Regulatory - Quality Systems
AR/UKRP services
Italy and UK
rgray@donawa.com
www.donawa.com

Original Message:
Sent: 25-Jan-2024 01:56
From: Christoph Kiesselbach
Subject: CE design certificate

Hi Kevin,

yes, I think I got a little bit carried away by the topic. Thank you for discussing this, it definitely brought up some points I would like to try to clarify.

I will do some research if there is some kind of consensus between NBs or an MDCG opinion on this and see if it makes sense to post the results in a separate thread.

Best regards, Christoph

------------------------------
Christoph Kiesselbach
Schrack & Partner
Reutlingen
Germany

Original Message:
Sent: 24-Jan-2024 11:47
From: Kevin Randall
Subject: CE design certificate

Thank you, Christoph. NBOG-3 (addressing certificate practices for the outgoing MDD) doesn't directly foresee a two-certificate approach yet does so indirectly in my experience.  Specifically, if the manufacturer, like most do, has chosen to employ an ISO standard as its means for meeting the Annex II Full Quality Assurance QMS requirements, then there is required to be a separate QMS certificate that is the basis for the Annex II Full Quality Assurance EC Certificate.

Also of note is NB-MED/2.5.1/Rec4 stating that the matter of interlinking of certificates has been considered and is recognized to be an administrative responsibility of the NB and should be traceable through the file of the NB. Inclusion of a cross-reference on the certificates themselves, however, would lead to practical difficulties.  Thus, the lack of such cross-references (like in the MDD certificates your cited) certainly doesn't necessarily mean that the certificate stands on its own.

Regarding the various additional questions you've raised (i.e., handling of significant changes for class IIa devices; NB practices for TD sampling after significant changes), those seem to be well beyond the scope of Anon's question.  I certainly have answers for you on those peripheral questions; but I think it's most appropriate to raise them in a separate threat so that this current thread can stay on track with Anon's intended question.

------------------------------
Kevin Randall, ASQ CQA, RAC (Europe, U.S., Canada)
Principal Consultant
Ridgway, CO
United States
© Copyright by ComplianceAcuity, Inc. All rights reserved.

Original Message:
Sent: 24-Jan-2024 03:48
From: Christoph Kiesselbach
Subject: CE design certificate

Hi Kevin,

thank you for providing the examples, that was definitely something I had not realised before. Just for others reading this here are two examples for publicly available current MDR certificates by the mentioned TÜV Süd and BSI that either declare that TD assessment is part of this certificate or that an Annex IX Chapter II certificate is only required for class III and IIb implantable devices (where every TD must be assessed).

With respect to the MDD: At least the old NBOG Best Practice Guide (2010-3) does not foresee a combination of Annex II excluding (4) with Annex II (4) ( for class IIa and IIb, but exclusively for class III. And in the corresponding MDD certificates to the examples you provided DQS used the singe-certificate approach under MDD (see here and here, where an additional Annex II section 4 certificate is limited to class III devices).

Obviously there is some freedom in the interpretaton of the MDR and the corresponding use of certificates. However I would really prefer if there was a common approach by the NBs. I will try to reach out to Team NB and see if they could imagine a common approach for their members.

But aside from the number of certificates, the main impact if Annex IX Chapter II is fully applicable to class IIa and IIb non-implantable devices (aside from the sampling aspect) in my opinion would be that this would also mean that significant design changes must be approved by the Notified Body and lead to a supplement to the (then existing) TD-certificate (of course also under the single-certificate approach each NB might have additional requirements on what needs to be reported by the manufacturer that might be more extensive, but I still think this would be different from a legal requirement with mandatory consequences for the certificates). At least in my experience that was handled differently under MDD for Annex II excluding (4) (at least if you go with the single-certificate approach and the conformity assessment routes given in NBOG 2010-3, see NBOG 2014-3).

There currently is no MDCG guidance on the interpretation of the need for approval of significant changes (with the exception of the specific case of Article 120), so their position is currently not clear to me. But how would this be handled by the NB in the case of TD sampling? If the manufacturer makes a significant change to a device within a group with a TD that has not yet been assessed as a sample, how does the NB assess the change? Based on the change notification only without any knowledge about the previous state of the TD? Do they perform an additional full TD assessment for this device? Or is the assessment only necessary if the TD already had been sampled? Maybe I am wrong but I think this would require a fair amount of additional resources and time, both currently in rather short supply.

Best regards, Christoph

------------------------------
Christoph Kiesselbach
Schrack & Partner
Reutlingen
Germany

Original Message:
Sent: 23-Jan-2024 17:00
From: Kevin Randall
Subject: CE design certificate

Christoph,

I can't show actual clients' examples; but I'll be creative to show a couple Class IIa examples that are publicly/readily available among others on the internet:

• Example 1
• Example 2

Therein on these Class IIa device Annex IX Chapter II EU Technical Documentation Assessment Certificates is stated, "For placing devices listed in the Annex on the market, an additional certificate according to Annex IX, Chapter I and III is required."

Along these lines during my 28-year tenure with medical device CE marking, I've seen many instances of the MDD Article 11(2) second paragraph Class IIa Annex II full quality assurance route generating a QMS certificate plus an EC certificate for the chosen assessment route through the most well-known NBs.

All this said, your insights have educated me to the fact that there does seem to be more NBs employing the single-certificate route for Class IIa devices. For example, it hadn't yet dawned on me that BSI and TUV seem to be sticking to that single-certificate approach for Class IIa Annex IX devices.

My understanding is that the NBs have freedom regarding the particular documented format/way in which they answer the aforesaid two separate, distinct, concomitant Class IIa Annex IX certificate requirements (QMS certificate + TD certificate).  Thus, this liberty may be why we are indeed seeing some use a single-certificate Class IIa Annex IX approach where that one certificate is in fact entitled "EU Quality Management System Assessment Certificate".  But I don't believe that is in clear alignment with the aforesaid two-certificate requirement.  Given how much trouble NBs have had in recent years issuing timely certificates, perhaps they are stretching this issue to reduce the administrative burden rather than following the approach in the examples above which seems to more clearly match the intent of the MDR.

------------------------------
Kevin Randall, ASQ CQA, RAC (Europe, U.S., Canada)
Principal Consultant
Ridgway, CO
United States
© Copyright by ComplianceAcuity, Inc. All rights reserved.

Original Message:
Sent: 23-Jan-2024 12:22
From: Christoph Kiesselbach
Subject: CE design certificate

Hi Kevin,

I respectfully disagree.

Under the MDD e.g. Annex II section 7 made it quite clear that for class IIa and IIb devices section 4 of Annex II (that contains provisions for design examination and a corresponding certificate) does not apply and instead TDs are sampled as part of the audit procedure (Section 3.3 of Annex II). The certificates for the quality management system (Annex II without section 4) that I know then stated that the validity for class III devices would depend on an additional design examination certificate (Annex II section 4) but I know of no additional certificate for class IIa/IIb devices where the TDs were sampled.

The MDR (of course) makes things less clear, but I think a not unusual interpretation of the additional TD assessment of samples in the case of IIa and IIb devices for e.g. Annex IX is that the assessment again is done as part of the audit activities (Chapter I of Annex IX, where also the additional TD sampling during surveillance in described) and while it is generally performed as described in Section 4 of Annex IX, the results are a part of the Annex IX chapter I (and III) certificate. Only class III and class IIb implantable devices where every TD is assessed receive an additional certificate in line with Section 4.9 and are subject to the approval of changes detailed in Section 4.10. I also think this makes sense: Only if every TD is assessed the reporting and approval of changes makes sense. If only samples are assessed, then what would the "approved device" in section 4.10 refer to? The whole generic device group / category of devices that the sample stands for or only the TDs that happen to have been sampled? If the NB does not know the TD the manufacturer is changing (because it has not yet been sampled), how are they assumed to assess changes? Would they have to do an additional full TD assessment?

And so far the MDR certificates I have seen are comparable to the ones under MDD, a certificate according to Annex IX Chapter I (and III) with the additional remark that in case of class III and IIb implantable devices an additional Annex IX Section 4 (or Chapter II) certificate is required (but not for class IIa/IIb).

However, I admit this is less clearly expressed than it was under the MDD and there might be NBs that interpret this differently and hand out Annex IX Chapter II certificates for class IIa/IIb (non implantable) devices. If you have examples for this, I would be interested to know them.

Best regards, Christoph

------------------------------
Christoph Kiesselbach
Schrack & Partner
Reutlingen
Germany

Original Message:
Sent: 23-Jan-2024 09:27
From: Kevin Randall
Subject: CE design certificate

Under both the outgoing Directives (e.g., MDD 93/42/EEC) and the new regulations (e.g., MDR 2017/745), whenever a chosen conformity assessment route requires Notified Body (NB) assessment of the quality management system (QMS), it is always required to be paired with NB assessment of the Technical Documentation (TD).  This always results in two certificates issued in order to gain Notified Body (NB) approval for placing the corresponding devices on the market: One certificate for the quality management system conformity, and one certificate for the device conformity. This two-certificate ensemble is a most fundamental certification requirement that is core to any successful and real CE marking experience where the manufacturer chooses a route involving QMS conformity.

For example, if the Article 52(6) first paragraph Class IIa conformity assessment route is chosen, then, by definition, that route involves two types of conformity assessments and certificates by the NB: One for the QMS plus one for the device TD. This is the way it's always been under the outgoing Directives (e.g., MDD 93/42/EEC), and this has been maintained under the new regulations (e.g., MDR 2017/745).  Specifically, for example under the MDR per Article 52(6) first paragraph, the Annex IX Chapters I & III QMS conformity assessment is performed by the NB resulting in an EU QMS assessment certificate per Annex IX section 2.3 fourth paragraph PLUS an Annex IX Chapter II section 4 TD assessment is performed by the NB resulting in an EU TD assessment certificate per Annex IX section 4.9.

As further examples, the same type of two-certificate requirement applies to the Article 52(6) second paragraph Class IIa conformity assessment route (and others for other device classes).

------------------------------
Kevin Randall, ASQ CQA, RAC (Europe, U.S., Canada)
Principal Consultant
Ridgway, CO
United States
© Copyright by ComplianceAcuity, Inc. All rights reserved.

Original Message:
Sent: 23-Jan-2024 04:11
From: Christoph Kiesselbach
Subject: CE design certificate

Hi Anon,

as Kevin says, the MDR does not use the expression "design certificate".

Annex XII that deals with the certificates issued by a notified body lists: "EU technical documentation assessment certificates, EU type-examination certificates and EU product verification certificates" in section 4(a) and "EU quality management system certificates and EU quality assurance certificates" in section 4(b). You could argue that the former are to some degree "device certificates", because they refer to an individual technical documentation, device type or even single devices, while the latter are "quality management certificates", but there is no such formulation in the MDR.

For a class IIa SaMD you usually would need either a "EU quality management system certificate" (Annex IX, chapters I and III) or a "EU quality assurence certificate" (Annex XI, Part A, Section 10). Both procedures would include the assessment of a technical documentation for at least one representative device sample for each device category, but as far as I know this assessment does not result in a separate certificate (as opposed to e.g. the full Annex IX assessment for class III devices or a type examination according to Annex X). If you only have one device, then of course the sampling in the end means that every device is assessed this way, but again this to my knowledge is not reflected in its own certificate. In theory Annex XI, Part B, Section 18, would be a possible procedure, but testing each device does not make sense for software and I do not think any notified body has this in their scope.

So in a loose sense, you do not obtain a design certificate for a class IIa software, but a quality management certificate (either full quality management system or product assurance). However, using the language of the MDR in describing the certificate is the better idea.

Best regards, Christoph

------------------------------
Christoph Kiesselbach
Schrack & Partner
Reutlingen
Germany

Original Message:
Sent: 21-Jan-2024 07:29
From: Anonymous Member
Subject: CE design certificate

This message was posted by a user wishing to remain anonymous

Dear RAPS Community,

What is CE design certificate and it is mandatory to obtain from NB for class IIa device (SaMD)?

Thank you for any info!