Corrective Action (CA) or Preventive Action (PA) need(s) to be triggered commensurate with risk. This is expected to be your primary mechanism for determining the need for CA or PA.
Specifically, if the observed (or potential) frequency of harm and severity of harm from a nonconformity exceed(s) the risk acceptance criteria set up in the subject device's Risk Management Plan, then CA or PA is/are to be initiated.
Typically there are multiple risk analysis methods that need to be employed to realize a full risk profile for the subject device. For example, PHA to help flush out consequent user-centric risks plus process FMEA for the manufacturing process and quality system processes to help flush out consequent process-centric and/or device-centric risks.
I generally recommend avoiding unqualified/undefined subjective trigger terms like "repetitive" or "systemic". Instead, such terms need to be defined in relation to the acceptance criteria set up in the Risk Management Plan.
------------------------------
Kevin Randall, ASQ CQA, RAC (U.S., Europe, Canada)
Principal Consultant
Ridgway, CO
United States
© Copyright 2022 by ComplianceAcuity, Inc. All rights reserved.
------------------------------
Original Message:
Sent: 15-Sep-2022 22:28
From: Anonymous Member
Subject: CAPA SOP
This message was posted by a user wishing to remain anonymous
Our CAPA SOP states a CAPA will be generated when repetitive or systemic errors occur. Our industry makes patient specific devices based on a doctor Rx. We are trying to associate percentages to the number of complaints or internal rework that would fall into the category of repetitive or systemic. Does anyone have guidance on how to define when a CAPA is warranted?
Thanks in advance for any help you can provide.