Regulatory Open Forum

-------------------------------------------
Bethany Hills JD
Attorney
Epstein Becker & Green PC
New York NY
United States
-------------------------------------------

820.70(a) applies to inspection, measuring, and test equipment (IM&TE). When you say they did validation, I infer you mean process validation under 820.75. In this case, one controls the process input parameters to provide a high degree of assurance that the process output, the product, is conforming. The reason for this approach is that the product cannot (or is not) fully verified by subsequent inspection or test.

Process validation set the acceptable values of the input parameters: time, temperature, and pressure in your example. Implicit is that one knows the actual value of these parameters during process validation, at setup for a production run, and during the run. 820.75(b) requires monitoring and recording of the data.

Consequently, these parameters must be measured, and so IM&TE is involved. Often, in injection molding, the IM&TE is built into the molding machine. For example there may be a pressure sensor with a display on the operating screen for the machine. The pressure sensor and display are IM&TE and subject to 820.70(a). The section equally applies to IM&TE built into production equipment and to stand-alone IM&TE. It is the existence of the measuring function that invokes 820.75(a).

FDA would not accept process validation or product verification in lieu of calibration, since they address very different issues. In the conventional process validation model, the IQ step would ensure that any IM&TE is in the calibration program. In addition, the OQ and PQ portions typically require that the IM&TE is within its calibration interval at the start and end. If not, you won't get valid results. In addition, the IM&TE must be in the calibration interval for any production runs.

Based on your description, the 483 citation is valid. You response should be a project to identify all IM&TE in the plant; establish the procedures to ensure that equipment is routinely calibrated, inspected, checked, and maintained; and bring all of the IM&TE under the aegis of the procedure. 

In my opinion, you should also expect a Warning Letter.
-------------------------------------------
Dan O'Leary
Swanzey NH
United States
-------------------------------------------


Original Message:
Sent: 06-25-2015 07:40
From: Bethany Hills
Subject: Calibration Procedures - Needed?

I am working with a client that uses an injection mold process and a sonic weld in manufacturing medical device.  They have an FDA 483 inspection report that is requiring them to calibration the instruments.  Since they haven't done calibration before (only validation and verification), they are worried that calibration would impact their validation.  They are looking for some help understanding what others are doing - verifying time and temp and pressure?  Would FDA accept verification plan in place of calibration?

I told them I'd reach out to my contacts that would have injection molding and get some thoughts.  The FDA is citing 820.72(a) and saying they need to establish procedures to ensure equipment is routinely calibrated, inspected, checked and maintained...Does this mean they must calibrate a piece of equipment or only aspects that need calibration (like gauges, etc).

-------------------------------------------
Bethany Hills JD
Attorney
Epstein Becker & Green PC
New York NY
United States
-------------------------------------------

While we are on this subject ... How about the need for a Master Validation Plan?
I recognize that an MVP is useful and broadly used, but is a 483 that cites the failure to have a MVP legit ?
Art
PS the auditor could not identify any missing validations, but still required a MVP
-------------------------------------------
A Arthur Rankis
President
A. A. Rankis & Associates, Inc.
Acton MA
United States
-------------------------------------------


Original Message:
Sent: 06-25-2015 12:25
From: Dan O'Leary
Subject: Calibration Procedures - Needed?

820.70(a) applies to inspection, measuring, and test equipment (IM&TE). When you say they did validation, I infer you mean process validation under 820.75. In this case, one controls the process input parameters to provide a high degree of assurance that the process output, the product, is conforming. The reason for this approach is that the product cannot (or is not) fully verified by subsequent inspection or test.

Process validation set the acceptable values of the input parameters: time, temperature, and pressure in your example. Implicit is that one knows the actual value of these parameters during process validation, at setup for a production run, and during the run. 820.75(b) requires monitoring and recording of the data.

Consequently, these parameters must be measured, and so IM&TE is involved. Often, in injection molding, the IM&TE is built into the molding machine. For example there may be a pressure sensor with a display on the operating screen for the machine. The pressure sensor and display are IM&TE and subject to 820.70(a). The section equally applies to IM&TE built into production equipment and to stand-alone IM&TE. It is the existence of the measuring function that invokes 820.75(a).

FDA would not accept process validation or product verification in lieu of calibration, since they address very different issues. In the conventional process validation model, the IQ step would ensure that any IM&TE is in the calibration program. In addition, the OQ and PQ portions typically require that the IM&TE is within its calibration interval at the start and end. If not, you won't get valid results. In addition, the IM&TE must be in the calibration interval for any production runs.

Based on your description, the 483 citation is valid. You response should be a project to identify all IM&TE in the plant; establish the procedures to ensure that equipment is routinely calibrated, inspected, checked, and maintained; and bring all of the IM&TE under the aegis of the procedure. 

In my opinion, you should also expect a Warning Letter.
-------------------------------------------
Dan O'Leary
Swanzey NH
United States
-------------------------------------------


Original Message:
Sent: 06-25-2015 07:40
From: Bethany Hills
Subject: Calibration Procedures - Needed?

I am working with a client that uses an injection mold process and a sonic weld in manufacturing medical device.  They have an FDA 483 inspection report that is requiring them to calibration the instruments.  Since they haven't done calibration before (only validation and verification), they are worried that calibration would impact their validation.  They are looking for some help understanding what others are doing - verifying time and temp and pressure?  Would FDA accept verification plan in place of calibration?

I told them I'd reach out to my contacts that would have injection molding and get some thoughts.  The FDA is citing 820.72(a) and saying they need to establish procedures to ensure equipment is routinely calibrated, inspected, checked and maintained...Does this mean they must calibrate a piece of equipment or only aspects that need calibration (like gauges, etc).

-------------------------------------------
Bethany Hills JD
Attorney
Epstein Becker & Green PC
New York NY
United States
-------------------------------------------

A hot button for me is auditors and Investigators who impose their own sense of best practice rather than the requirement. This is especially troubling in 483s and Warning Letters, since they are legal documents.

Having said that, many companies get themselves into trouble by writing procedures that they don't follow. Often these procedures call out the best practices. The key word here is "establish", which means to define, document, and implement. If a company's procedure calls for a Master Validation Plan and the company doesn't have one, it is a legitimate 483 observation.

820.75 doesn't require IQ, OQ, & PQ; a master validation plan; or any other specific method or tool. Short of a company's procedure calling for these approaches, a 483 that cites failure to use specific method is, in my opinion, incorrect.

-------------------------------------------
Dan O'Leary
Swanzey NH
United States
-------------------------------------------


Original Message:
Sent: 06-26-2015 10:02
From: A Arthur Rankis
Subject: Calibration Procedures - Needed?

While we are on this subject ... How about the need for a Master Validation Plan?
I recognize that an MVP is useful and broadly used, but is a 483 that cites the failure to have a MVP legit ?
Art
PS the auditor could not identify any missing validations, but still required a MVP
-------------------------------------------
A Arthur Rankis
President
A. A. Rankis & Associates, Inc.
Acton MA
United States
-------------------------------------------


Original Message:
Sent: 06-25-2015 12:25
From: Dan O'Leary
Subject: Calibration Procedures - Needed?

820.70(a) applies to inspection, measuring, and test equipment (IM&TE). When you say they did validation, I infer you mean process validation under 820.75. In this case, one controls the process input parameters to provide a high degree of assurance that the process output, the product, is conforming. The reason for this approach is that the product cannot (or is not) fully verified by subsequent inspection or test.

Process validation set the acceptable values of the input parameters: time, temperature, and pressure in your example. Implicit is that one knows the actual value of these parameters during process validation, at setup for a production run, and during the run. 820.75(b) requires monitoring and recording of the data.

Consequently, these parameters must be measured, and so IM&TE is involved. Often, in injection molding, the IM&TE is built into the molding machine. For example there may be a pressure sensor with a display on the operating screen for the machine. The pressure sensor and display are IM&TE and subject to 820.70(a). The section equally applies to IM&TE built into production equipment and to stand-alone IM&TE. It is the existence of the measuring function that invokes 820.75(a).

FDA would not accept process validation or product verification in lieu of calibration, since they address very different issues. In the conventional process validation model, the IQ step would ensure that any IM&TE is in the calibration program. In addition, the OQ and PQ portions typically require that the IM&TE is within its calibration interval at the start and end. If not, you won't get valid results. In addition, the IM&TE must be in the calibration interval for any production runs.

Based on your description, the 483 citation is valid. You response should be a project to identify all IM&TE in the plant; establish the procedures to ensure that equipment is routinely calibrated, inspected, checked, and maintained; and bring all of the IM&TE under the aegis of the procedure. 

In my opinion, you should also expect a Warning Letter.
-------------------------------------------
Dan O'Leary
Swanzey NH
United States
-------------------------------------------


Original Message:
Sent: 06-25-2015 07:40
From: Bethany Hills
Subject: Calibration Procedures - Needed?

I am working with a client that uses an injection mold process and a sonic weld in manufacturing medical device.  They have an FDA 483 inspection report that is requiring them to calibration the instruments.  Since they haven't done calibration before (only validation and verification), they are worried that calibration would impact their validation.  They are looking for some help understanding what others are doing - verifying time and temp and pressure?  Would FDA accept verification plan in place of calibration?

I told them I'd reach out to my contacts that would have injection molding and get some thoughts.  The FDA is citing 820.72(a) and saying they need to establish procedures to ensure equipment is routinely calibrated, inspected, checked and maintained...Does this mean they must calibrate a piece of equipment or only aspects that need calibration (like gauges, etc).

-------------------------------------------
Bethany Hills JD
Attorney
Epstein Becker & Green PC
New York NY
United States
-------------------------------------------

Hi folks.

I completely agree with Dan's assessment.  The most likely cause of the 483 observation is the fact that somewhere in some document the company references a master validation plan.  Mind you, it could be in something as far away from your equipment as a product validation SOP referencing a MVP.  Once the term is mentioned, if the company cannot supply the documentation to the auditor, it becomes a potential 483 issue.

That said, we have likely ALL had that auditor who comes in "expecting" certain things and finds fault when you do not have the ability to produce them - regardless of what your procedures actually state you will do.  I would probably have less of an issue if he had made the comment the no MVP existed but was necessary in order to provide the records needed to perform a full functional audit.  I also wonder (I am not sure anymore!) whether or not the FDA has established guidance on the issue where the "recommend" a MVP be created and maintained.  While we all know from guidance documents that the guidance does not establish any right or responsibility on the part of either the industry or the FDA, there does seem to be a hold in training of inspectors sometimes because they site guidance as the reason for the observation. 

I would suggest if you really believe your system is in compliance that you discuss the issues with someone at FDA.  The problem, as Dan notes, is that the document is a legal document that can be used as evidence against you later and more importantly can damage your corporate credibility with investors, physicians, and patients.  Additionally, in conjunction with the EIR the 483 is going to be reviewed to determine if the FDA issues a Warning Letter or takes further regulatory corrective actions to "ensure compliance" to something that is not necessarily a compliance fault - especially if you have all the necessary documentation to prove that you are in control.

Obviously, these are just my personal thoughts on the issue.

-------------------------------------------
Victor Mencarelli
Sr. Manager - Regulatory Affairs
United States
-------------------------------------------


Original Message:
Sent: 06-26-2015 10:22
From: Dan O'Leary
Subject: Calibration Procedures - Needed?

A hot button for me is auditors and Investigators who impose their own sense of best practice rather than the requirement. This is especially troubling in 483s and Warning Letters, since they are legal documents.

Having said that, many companies get themselves into trouble by writing procedures that they don't follow. Often these procedures call out the best practices. The key word here is "establish", which means to define, document, and implement. If a company's procedure calls for a Master Validation Plan and the company doesn't have one, it is a legitimate 483 observation.

820.75 doesn't require IQ, OQ, & PQ; a master validation plan; or any other specific method or tool. Short of a company's procedure calling for these approaches, a 483 that cites failure to use specific method is, in my opinion, incorrect.

-------------------------------------------
Dan O'Leary
Swanzey NH
United States
-------------------------------------------


Original Message:
Sent: 06-26-2015 10:02
From: A Arthur Rankis
Subject: Calibration Procedures - Needed?

While we are on this subject ... How about the need for a Master Validation Plan?
I recognize that an MVP is useful and broadly used, but is a 483 that cites the failure to have a MVP legit ?
Art
PS the auditor could not identify any missing validations, but still required a MVP
-------------------------------------------
A Arthur Rankis
President
A. A. Rankis & Associates, Inc.
Acton MA
United States
-------------------------------------------


Original Message:
Sent: 06-25-2015 12:25
From: Dan O'Leary
Subject: Calibration Procedures - Needed?

820.70(a) applies to inspection, measuring, and test equipment (IM&TE). When you say they did validation, I infer you mean process validation under 820.75. In this case, one controls the process input parameters to provide a high degree of assurance that the process output, the product, is conforming. The reason for this approach is that the product cannot (or is not) fully verified by subsequent inspection or test.

Process validation set the acceptable values of the input parameters: time, temperature, and pressure in your example. Implicit is that one knows the actual value of these parameters during process validation, at setup for a production run, and during the run. 820.75(b) requires monitoring and recording of the data.

Consequently, these parameters must be measured, and so IM&TE is involved. Often, in injection molding, the IM&TE is built into the molding machine. For example there may be a pressure sensor with a display on the operating screen for the machine. The pressure sensor and display are IM&TE and subject to 820.70(a). The section equally applies to IM&TE built into production equipment and to stand-alone IM&TE. It is the existence of the measuring function that invokes 820.75(a).

FDA would not accept process validation or product verification in lieu of calibration, since they address very different issues. In the conventional process validation model, the IQ step would ensure that any IM&TE is in the calibration program. In addition, the OQ and PQ portions typically require that the IM&TE is within its calibration interval at the start and end. If not, you won't get valid results. In addition, the IM&TE must be in the calibration interval for any production runs.

Based on your description, the 483 citation is valid. You response should be a project to identify all IM&TE in the plant; establish the procedures to ensure that equipment is routinely calibrated, inspected, checked, and maintained; and bring all of the IM&TE under the aegis of the procedure. 

In my opinion, you should also expect a Warning Letter.
-------------------------------------------
Dan O'Leary
Swanzey NH
United States
-------------------------------------------


Original Message:
Sent: 06-25-2015 07:40
From: Bethany Hills
Subject: Calibration Procedures - Needed?

I am working with a client that uses an injection mold process and a sonic weld in manufacturing medical device.  They have an FDA 483 inspection report that is requiring them to calibration the instruments.  Since they haven't done calibration before (only validation and verification), they are worried that calibration would impact their validation.  They are looking for some help understanding what others are doing - verifying time and temp and pressure?  Would FDA accept verification plan in place of calibration?

I told them I'd reach out to my contacts that would have injection molding and get some thoughts.  The FDA is citing 820.72(a) and saying they need to establish procedures to ensure equipment is routinely calibrated, inspected, checked and maintained...Does this mean they must calibrate a piece of equipment or only aspects that need calibration (like gauges, etc).

-------------------------------------------
Bethany Hills JD
Attorney
Epstein Becker & Green PC
New York NY
United States
-------------------------------------------

Validation resources for both pharma and device industries!

http://regulatorydoctor.com/process-validation-drugs-medical-devices/



Original Message:
Sent: 06-29-2015 08:28
From: Victor Mencarelli
Subject: Calibration Procedures - Needed?

Hi folks.

I completely agree with Dan's assessment.  The most likely cause of the 483 observation is the fact that somewhere in some document the company references a master validation plan.  Mind you, it could be in something as far away from your equipment as a product validation SOP referencing a MVP.  Once the term is mentioned, if the company cannot supply the documentation to the auditor, it becomes a potential 483 issue.

That said, we have likely ALL had that auditor who comes in "expecting" certain things and finds fault when you do not have the ability to produce them - regardless of what your procedures actually state you will do.  I would probably have less of an issue if he had made the comment the no MVP existed but was necessary in order to provide the records needed to perform a full functional audit.  I also wonder (I am not sure anymore!) whether or not the FDA has established guidance on the issue where the "recommend" a MVP be created and maintained.  While we all know from guidance documents that the guidance does not establish any right or responsibility on the part of either the industry or the FDA, there does seem to be a hold in training of inspectors sometimes because they site guidance as the reason for the observation. 

I would suggest if you really believe your system is in compliance that you discuss the issues with someone at FDA.  The problem, as Dan notes, is that the document is a legal document that can be used as evidence against you later and more importantly can damage your corporate credibility with investors, physicians, and patients.  Additionally, in conjunction with the EIR the 483 is going to be reviewed to determine if the FDA issues a Warning Letter or takes further regulatory corrective actions to "ensure compliance" to something that is not necessarily a compliance fault - especially if you have all the necessary documentation to prove that you are in control.

Obviously, these are just my personal thoughts on the issue.

-------------------------------------------
Victor Mencarelli
Sr. Manager - Regulatory Affairs
United States
-------------------------------------------


Original Message:
Sent: 06-26-2015 10:22
From: Dan O'Leary
Subject: Calibration Procedures - Needed?

A hot button for me is auditors and Investigators who impose their own sense of best practice rather than the requirement. This is especially troubling in 483s and Warning Letters, since they are legal documents.

Having said that, many companies get themselves into trouble by writing procedures that they don't follow. Often these procedures call out the best practices. The key word here is "establish", which means to define, document, and implement. If a company's procedure calls for a Master Validation Plan and the company doesn't have one, it is a legitimate 483 observation.

820.75 doesn't require IQ, OQ, & PQ; a master validation plan; or any other specific method or tool. Short of a company's procedure calling for these approaches, a 483 that cites failure to use specific method is, in my opinion, incorrect.

-------------------------------------------
Dan O'Leary
Swanzey NH
United States
-------------------------------------------


Original Message:
Sent: 06-26-2015 10:02
From: A Arthur Rankis
Subject: Calibration Procedures - Needed?

While we are on this subject ... How about the need for a Master Validation Plan?
I recognize that an MVP is useful and broadly used, but is a 483 that cites the failure to have a MVP legit ?
Art
PS the auditor could not identify any missing validations, but still required a MVP
-------------------------------------------
A Arthur Rankis
President
A. A. Rankis & Associates, Inc.
Acton MA
United States
-------------------------------------------


Original Message:
Sent: 06-25-2015 12:25
From: Dan O'Leary
Subject: Calibration Procedures - Needed?

820.70(a) applies to inspection, measuring, and test equipment (IM&TE). When you say they did validation, I infer you mean process validation under 820.75. In this case, one controls the process input parameters to provide a high degree of assurance that the process output, the product, is conforming. The reason for this approach is that the product cannot (or is not) fully verified by subsequent inspection or test.

Process validation set the acceptable values of the input parameters: time, temperature, and pressure in your example. Implicit is that one knows the actual value of these parameters during process validation, at setup for a production run, and during the run. 820.75(b) requires monitoring and recording of the data.

Consequently, these parameters must be measured, and so IM&TE is involved. Often, in injection molding, the IM&TE is built into the molding machine. For example there may be a pressure sensor with a display on the operating screen for the machine. The pressure sensor and display are IM&TE and subject to 820.70(a). The section equally applies to IM&TE built into production equipment and to stand-alone IM&TE. It is the existence of the measuring function that invokes 820.75(a).

FDA would not accept process validation or product verification in lieu of calibration, since they address very different issues. In the conventional process validation model, the IQ step would ensure that any IM&TE is in the calibration program. In addition, the OQ and PQ portions typically require that the IM&TE is within its calibration interval at the start and end. If not, you won't get valid results. In addition, the IM&TE must be in the calibration interval for any production runs.

Based on your description, the 483 citation is valid. You response should be a project to identify all IM&TE in the plant; establish the procedures to ensure that equipment is routinely calibrated, inspected, checked, and maintained; and bring all of the IM&TE under the aegis of the procedure. 

In my opinion, you should also expect a Warning Letter.
-------------------------------------------
Dan O'Leary
Swanzey NH
United States
-------------------------------------------


Original Message:
Sent: 06-25-2015 07:40
From: Bethany Hills
Subject: Calibration Procedures - Needed?

I am working with a client that uses an injection mold process and a sonic weld in manufacturing medical device.  They have an FDA 483 inspection report that is requiring them to calibration the instruments.  Since they haven't done calibration before (only validation and verification), they are worried that calibration would impact their validation.  They are looking for some help understanding what others are doing - verifying time and temp and pressure?  Would FDA accept verification plan in place of calibration?

I told them I'd reach out to my contacts that would have injection molding and get some thoughts.  The FDA is citing 820.72(a) and saying they need to establish procedures to ensure equipment is routinely calibrated, inspected, checked and maintained...Does this mean they must calibrate a piece of equipment or only aspects that need calibration (like gauges, etc).

-------------------------------------------
Bethany Hills JD
Attorney
Epstein Becker & Green PC
New York NY
United States
-------------------------------------------

You asked, "Does this mean they must calibrate a piece of equipment or only aspects that need calibration (like gauges, etc.)."

If the equipment performs three measurements (time, temperature, and pressure), there are three calibration activities. One would expect three calibration certificates, each one of which is traceable to NIST. Depending on the stability of the equipment, there could be three different calibration intervals.

More commonly, however, the calibration house (or manufacturer) will calibrate the unit as a whole and provide one calibration report (with the appropriate sections) and one calibration sticker.

-------------------------------------------
Dan O'Leary
Swanzey NH
United States
-------------------------------------------


Original Message:
Sent: 06-25-2015 07:40
From: Bethany Hills
Subject: Calibration Procedures - Needed?

I am working with a client that uses an injection mold process and a sonic weld in manufacturing medical device.  They have an FDA 483 inspection report that is requiring them to calibration the instruments.  Since they haven't done calibration before (only validation and verification), they are worried that calibration would impact their validation.  They are looking for some help understanding what others are doing - verifying time and temp and pressure?  Would FDA accept verification plan in place of calibration?

I told them I'd reach out to my contacts that would have injection molding and get some thoughts.  The FDA is citing 820.72(a) and saying they need to establish procedures to ensure equipment is routinely calibrated, inspected, checked and maintained...Does this mean they must calibrate a piece of equipment or only aspects that need calibration (like gauges, etc).

-------------------------------------------
Bethany Hills JD
Attorney
Epstein Becker & Green PC
New York NY
United States
-------------------------------------------

Beth,

What a practical question!

5% or around of FDA warning letters issued to device firms may include citations of 21 CFR 820.72. 

As an analogy, when we buy a brand new car, no maintenance check-up is needed - let's say that is because it has been "fully verified and validated" before its release for sale.    

As we drive for a year or longer, we have to go through a maintenance check up, meeting some specifications - let's say this is part of "calibration, measuring, and test equipment."  

Now under the umbrella of FDA QSRs (per 21 CFR Part 820, Section 820.72(a)-(d), a firm shall:

• establish and maintain SOPs for calibration (e.g., 21 CFR 820.72 (b)) including pre-established specifications (e.g., temp, time, pressure based on your description) ("SPEC #2").
• documentation of calibration records should be maintained.  
• etc.

Under 21 CFR 820.75, your client may have validated the process (e.g., IQ/OP/PQ) based on the prior-established process parameters ("SPEC #1").

What should be considered as best approaches for your client:

1. establish procedures for calibration based on pre-established specifications (SPEC #2).
2. revalidate the process using any equipment (not calibrated) UNLESS SPEC #1 and SPEC #2 match within the error margins/ranges potentially affecting the safety and performance of your devices.

Based on the results, your client should consider what to do further including a recall, if warranted under the circumstances, as part of a retrospective corrective action. .  

In responding to 483s, a response should be complete and in full including retrospective corrective action unless it is justified. 

____________________________________________________________
Dr. David Lim, Ph.D., RAC, ASQ-CQA 

Original Message:
Sent: 06-25-2015 07:40
From: Bethany Hills
Subject: Calibration Procedures - Needed?

I am working with a client that uses an injection mold process and a sonic weld in manufacturing medical device.  They have an FDA 483 inspection report that is requiring them to calibration the instruments.  Since they haven't done calibration before (only validation and verification), they are worried that calibration would impact their validation.  They are looking for some help understanding what others are doing - verifying time and temp and pressure?  Would FDA accept verification plan in place of calibration?

I told them I'd reach out to my contacts that would have injection molding and get some thoughts.  The FDA is citing 820.72(a) and saying they need to establish procedures to ensure equipment is routinely calibrated, inspected, checked and maintained...Does this mean they must calibrate a piece of equipment or only aspects that need calibration (like gauges, etc).

-------------------------------------------
Bethany Hills JD
Attorney
Epstein Becker & Green PC
New York NY
United States
-------------------------------------------