FDA BIMO 483 findings: More than a decade with no change
Insanity is doing the same thing over and over and expecting a different result. As we look at the data on 483 findings from FDA clinical investigator (site) inspections from over a decade (shown above), we can see the same errors at the top of the leaderboard each year. In 2020 and 2021, the FDA broke down these large topic areas into their components. You can see that this breakdown enhances our understanding and show the value of a controlled vocabulary.
Let’s start with Informed Consent.
The FDA listed the following themes for Informed Consent:
· ICF Not Obtained
· Revised consent not obtained/timely
· ICF not obtained prior to screening/reconsent/IP admin
· ICF not obtained at rescreening
· ICF not obtained for Sub-Study
· ICF copy not provided
· ICF Changes not approved by IRB
· ICF Not in Understandable Language
· ICF Short Form Not Witnessed
· ICF Circumstances—not sufficient opportunity/not enough time
· ICF coercion
Note that these findings are categorized in two places: ICF (50.20) and Protocol compliance (312.60/812.100, 812.110(b). Ideally, from an analysis standpoint they would be classified in one place. There was no way to know that ICF findings were in two places without the breakdown provided recently. Another learning—classify items consistently and in one place for best insights.
Many of the errors are not at the beginning of the trial but are more likely to happen when there are changes or amendments. Everyone knows that you need to have an IRB approved consent obtained at the start of the study. When there are amendments, are your site processes set up to identify when a new amendment and consent have been IRB approved so that subjects can be reconsented promptly? Debra Fontana recently posted about the importance of a eTMF index to assure that all documents FOR EACH STUDY CHANGE are included in the eTMF.
From a risk standpoint, what are your oversight methods that will identify whether the changes from an amendment are implemented correctly? Note, this requires study specific analytic methods, which are more effective than manual checks. They can be implemented faster and don't require a monitoring visit to identify the error.
Stay tuned—my next posts will cover the rest of the breakdown and some innovative approaches to prevent these errors from derailing your development.
Feel free to contact me if you would like to implement ongoing quality oversight for your trials to prevent these types of errors.
President, MOAB Clinical Research Consulting, Inc.
2w100%