CAPA Best Practices:

How to ensure proper implementation of this

integral component of a Research Quality System
Mary-Tara Roth, RN, MSN, MPH
Director, Clinical Research Resources Office
June 18, 2019
Objectives
• Define CAPA and state the purpose of CAPAs.
• List Components of a strong CAPA.
• Summarize process for a Root Cause Analysis (RCA).
• Explain how to best respond to audit findings.
• Describe what makes a good CAPA vs a poor/inadequate CAPA.
What is a CAPA?
• Corrective and Preventive Action Plan
• Corrective Actions/Preventive Actions
• Documenting and communicating the plan to address the
problem
• Correct: REACTIVE steps to correct the immediate problem
• Understand: IDENTIFY underlying cause(s) and extent of the problem(s)
• Prevent: PROACTIVE steps to prevent future recurrence of the problem(s)
• Communication of the actions (assessment, approval)
• Study team, IRB, FDA, Sponsor, Funder, etc.
Corrective and Preventive Action Plan (CAPA)
• A system for resolving quality issues
• Resolve/correct problem and keep it from happening again
• Term originated in manufacturing field
• Required in FDA device/device manufacturing regulations
• “Quality System Regulation” (21 CFR 820.100 and 21 CFR 211)
 A Research study as a quality system
(paraphrased from 21 CFR 820.100)

• Analyze processes to identify potential causes of “nonconforming product”

• Investigate causes of nonconformities
• Identify the actions needed to correct/prevent problem
• Verify corrective and preventive actions to ensure they are effective
• Implement and record needed changes in methods and procedures
• Ensure information related to quality problems is disseminated to those
responsible for assuring quality of product
• Submit relevant information on identified quality problems and corrective
and preventive actions for management review
• Document activities and results
A Research study as a quality system
• 4 main components of a quality system…
• Say what you do
• Do what you say
• Prove it
• Improve it

From CTTI Summary of Expert Meeting, 2010, Developing Effective Quality Systems
in Clinical Trials: An Enlightened Approach
Regulatory perspectives…. CAPAs in Clinical Research
FDA Guidance:
• The PI should ensure a procedure for the timely correction and documentation
of problems identified by study personnel, outside monitors or auditors, or
other parties involved in the conduct of a study.
FDA Guidance - Investigator Responsibilities, Oct. 2009
ICH GCP 2.13 (Principles)
• Systems with procedures that assure the quality of every aspect of the trial
should be implemented.
ICH GCP 5.1.1
• The sponsor is responsible for implementing and maintaining quality assurance
and quality control systems with written SOPs to ensure that trials are
conducted and data are generated, documented (recorded), and reported in
compliance with the protocol, GCP, and the applicable regulatory requirements
When do we need a CAPA?
• Depth of CAPA investigation and implementation
should match the risk.
• Is the quality of the data and/or the safety of
subjects potentially adversely impacted?
7-step Plan for Successful CAPA
• Evaluate the extent of the problem:
• Assess for potential harm to subjects

DOCUMENT EACH STEP!

• Determine the cause(s) of the problem
• Report to IRB, sponsor, other entities as applicable
• Multiple updates may be necessary
• Correct the problem as it relates to current subjects
• Develop processes to ensure the problem is prevented in the future
• Train on new processes
• Follow up to ensure that all steps of the CAPA are successful
Elements of your CAPA
• Description of the problem
• Narrative of events
• Number of subjects affected/harmed
• Number potentially affected/harmed
• The root and contributing causes for each finding
• Include how this was determined
• Corrective actions and preventive actions taken or to be taken
• Include description of new or changed processes and/or SOPs
• Describe plan for training
• Describe plan for evaluating the effectiveness
• Reporting
 How did this happen?
The Swiss Cheese model
Two reasons for holes in the system
• Active failure: Unsafe acts by people in
direct contact with the subject or system
• Latent conditions: Arising top level
mngt./procedure developers,
institutional culture, etc.
- “Error-provoking conditions” within the
workplace: inadequate staffing,
insufficient training
- Can create long-lasting holes or
weaknesses in defenses
J. Reason, Human Error: models and management, BMC 2000;320:768
Root Cause Analysis (RCA)
• Understand that clinical research studies are
complex systems
• Often will have to assess multiple
levels/processes/individuals
• Perform as soon as possible after the problems
identified
• Include all individuals involved in the error
Root Cause Analysis
• Identify the problem….
• Review processes, interview those involved …..
• What happened?
• How did it happen?
• Why did it happen? When? Where?
• What were contributing factors?
• Who was involved? Who was affected?
• How often did it happen? How many were affected? How serious is
the problem? How extensive is the problem?
• Usually these questions will lead you to the underlying cause(s)
“5 Whys” technique (aka the “Toddler technique!”)
• State the problem: Urine dipstick result not recorded.
• Why 1: Staff did not write down baseline urine dipstick result.
• Why 2: Staff did not know they needed to record the results at
baseline.
• Why 3: Staff had not been trained on what data elements had
to be captured at baseline visit.
• Why 4: PI relied on the sponsor provided baseline CRF to
specify what data points need to be captured at the baseline
visit.
• Why 5: PI believed sponsor would ensure all needed data
points would be placed on the baseline CRF.

More on 5 Whys: P. Williams, BUMC PROCEEDINGS 2001;14:154–157

Corrections/Preventions
• Different levels within the research infrastructure may have
contributed to the problem (remember the Swiss cheese!)
thus
• Changes may be needed at multiple levels CHANGE
• Institutional policies
• Department/clinic policies CHANGE

• Processes for study team working on multiple studies CHANGE

• Internal study processes specific to a particular study CHANGE

• Example: study team, department, pharmacy,

GCRU/clinical center, etc.
Who gets the blame?
20 inspections … 23 parties blamed
Highest, at 39%? …… Study Coordinator
Scientific Misconduct: The F Word, Stan Woollen, 2001

“When an investigator blames

colleagues, it’s meaningless to the FDA…”
- Adil Shamoo, PhD, Bioethicist
“FDA cites investigators for poor
recordkeeping…. blaming staff” Clinical Trials
Compliance, J. Harzbecker, Dec. 2005
 Corrections/Preventions: Own the Problem(s)
“Prior to the detailed replies, I wish to “Unfortunately, our study coordinator was not
clearly state that I understand I am sufficiently trained on the process of obtaining
responsible for oversight of all activities informed consent. She also did not
that occurred at our site regarding this understand the electronic IRB system and the
protocol. We do not dispute any of the importance of ensuring that when there are
findings of the inspection and we amendments to the study there needs to be a
acknowledge the serious nature of the new consent form printed from the system.
observations and have taken action This contributed to the issues raised in the
both in the past and now to directly audit. We have provided training to the study
address these specific issues and help coordinator.”
ensure similar research-related issues
do not occur at our site on any current
or future study.”
Don’t be this investigator!
Clues to deficient CAPAs: FDA Warning Letters
Not enough detail “… you identified the problem and
have established certain corrective
• Not addressing why the problem
occurred
actions…. You identified and
addressed the integrity of the
• Not describing the investigational procedures and data
extent/pervasiveness in other clinical investigations in
• # times/#subjects/#studies, etc. which the terminated research
• Not detailing timeframe of the nurse was involved. However, we
corrective actions find that you have not adequately
• Note detailing how you assess addressed how you will improve
effectiveness of the corrective your supervision of study staff in
actions future….”
Clues to deficient CAPAs: FDA Warning Letters
“Your corrective actions to ensure
Describing reporting of deaths to FDA and IRB
include: reconfiguring your team,
corrective actions holding an IRB training for staff,
informing staff that you must be
notified immediately of any subject’s
without developing death… We are unable to undertake an
informed evaluation of your response
because you did not provide
SOPs documentation further explaining your
corrective action plan, for example, an
SOP that shows your staff is to notify
you immediately upon becoming
aware of any death.”
Clues to deficient CAPAs: FDA Warning Letters
“You indicated that you have added a
Insufficient detail ‘clinical trials link’ to your site’s EMR
to provide access to study
to determine if information for study staff…. Your
response is inadequate because you
CAPA will correct did not provide sufficient information
to enable us to evaluate the
adequacy of your corrective action
the problem…. plan….. it is unclear how adding a
“clinical trials link” to your site’s EMR
• Poor root cause analysis! will ensure that protocol
requirements will be met for studies
conducted at your site.”
Clues to deficient CAPAs: FDA Warning Letters
“We are concerned that the majority
Not providing detail of the corrective actions appear to
on corrective actions represent actions taken by the xxx
Medical Center and do not reflect
the investigator corrective actions that you personally
have taken.”
him/herself is taking
 Clues to deficient CAPAs: FDA Warning Letters
“You noted you were unaware at the
Indicating that the PI time of the study that the xxx
doesn’t understand assessments for these subjects were
not completed properly, and this
his/her responsibilities violation was not brought to your
attention by either your staff or
as clinical investigator monitors… We wish to emphasize
that as the clinical investigator, it was
your ultimate responsibility to ensure
that these studies were conducted
properly…”
Clues to deficient CAPAs: FDA Warning Letters
Not providing “Although you stated that your SOP
has been put into effect and that your
documentation that research coordinators are well aware
of this requirement, you failed to
corrective measures provide documentation that your
have been done or when research staff have been adequately
trained in this SOP …”
they will be done
Also keep in mind….
• Don’t promise corrective/preventive actions that could never realistically be
carried out
• Try to get the “just right” (think “Baby bear” )
• What is feasible, with the expected positive effects
• Example: “Site staff will review the EMR for all subjects for AEs on a weekly basis.” (There are
~100 subjects.)

• Make sure to DO WHAT YOU SAY YOU WILL DO

• If you find that you cannot, amend the CAPA (and get it approved)
• Ex: “All protocol deviations have been entered on a deviations log.”
• A deviation log was never created and instead deviations were entered in the EMR
• What is the potential problem with this practice?

• Recognize that problems in one study can likely mean problems in other studies
• Your CAPA might have to involve other studies under the PI and/or using the same study
staff
Activity 1
Instructions: Review the examples below, some from real (but modified) draft or
final CAPAs. Provide constructive advice on needed modifications.

Example 1: Reporting AEs to Sponsor

• Finding: Multiple SAEs not reported to the Sponsor within the protocol-
specified timeframe (48 hours).

• Response: All AEs were reported according to the [institution name] IRB
policies and procedures. As all care delivered through this trial has been
“standard of care” there are no safety concerns for research subjects
regarding these observations….We acknowledge that the timing of
reporting was delayed during the early phase of the study in 2013, and
have already worked diligently in 2014 to comply with the timing required
by the protocol. There are multiple examples of our compliance with the
required reporting timeframe [list of subjects].
Activity 1
Instructions: Review the examples below, some from real draft or final CAPAs.
Provide constructive advice on needed modifications.

Example 1a: Reporting AEs to Sponsor

• Finding: Multiple SAEs not reported to the Sponsor within the protocol-specified
timeframe (though they were reported to the IRB).

• Response: In order to improve communications internally between study team

members, a review of enrollment logs and occurrence of hospital admissions
and/or assessment of medically significant events will be conducted by the PI at
weekly/biweekly research meetings to cross check with study investigators and
further improve timing of SAE reporting.
• Study coordinator will perform bi-weekly follow-up phone calls to study
participants to monitor admissions or medically significant events outside of BMC
and will inform study investigators immediately regarding such events … study
investigators will report AEs to the Sponsor according to the protocol.
Activity 1
Instructions: Review the examples below, some from real draft or final CAPAs.
Provide constructive advice on needed modifications.
Example 2:
• Finding: Subject 011 enrolled in xxx, the Source Document Worksheet for Visit 17 contains
discrepancies regarding the fundoscopy exam. The document originally noted in that the
fundoscopy exam was performed on Jan. 11, 2011 but the entry of “yes” for the
performance of the exam was crossed out, and a notation of “not performed” was added.
On June 2, 2011, the words “not performed” were crossed out, with the word “error”
entered above that crossout. The document also contains another late entry, dated June 2,
2011, with fundoscopy examp findings. In addition, the document contains the undated
entry, “visual acuity and fundoscopy were done but not documented – missed
documentation.” We were unable to determine whether the fundoscopy exam was
performed on Jan. 11, 2011 or June 2, 2011.
• Response: All discrepancies in the study records noted in the finding were directly related to
our poor documentation practices. We failed to document the fundoscopy exam properly
because the physical exam worksheets did not contain a section to capture fundoscopy
exams. In addition, missed assessments were due to a confusion of the protocol; the
fundoscopy exams were not required for every visit and some were inadvertently missed,
while others were captured when they were not required. We revised source documents to
include the physical and fundoscopy exams on the required visit dates, as required by the
protocol. The staff have been re-educated on proper source documentation and on GCP. In
addition, moving forward, we will document all required assessments at the time of
subjects’ visits to avoid late entries.
Activity 1
Instructions: Review the examples below, some from real draft or final CAPAs.
Provide constructive advice on needed modifications.

Example 3:
• Finding: Audiometry reports that were represented as reports for subject 3
at visit 2 and visit 3 are obscured audiometry reports that were originally
for other subjects…..

• Response: The CRO site monitor instructed our study coordinator to

obscure identifying subject information. In addition, our study coordinator
made errors in transcribing the subject information. The errors in
transcribing the subject information were not intentional and involved only
5% of all audiograms that our site generated. Where applicable, we have
instituted additional measures and procedures to address the inspection
findings.
CAPA - DOs
A great plan may not pass IRB or FDA muster if you don’t:
• TAKE RESPONSIBILITY
• APPROPRIATELY ASSESS CAUSES OF THE PROBLEM
• DEVELOP/MODIFY WRITTEN PROCESSES (SOPs)/PROTOCOL
• TRAIN STAFF ON NEW PROCESSES
• EVALUATE TO ENSURE YOUR CAPA WORKS
• DOCUMENT EVERY STEP OF YOUR CAPA
• Training
• SOPs
• Assessment of whether CAPA initiatives are effective
CAPA - DOs
If you develop SOP(s) or modify the protocol as
part of your CAPA
• Ensure the SOP addresses the root cause(s)
• Ensure SOP details procedures to fix and prevent the
problem
• Train staff on this new SOP
• Document training (when, what, who, who)
• Perform self-assessment to ensure SOP worked as
part of the CAPA
Food for thought…… “PA-CAPA”
• Is the idea of a CAPA too REACTIVE?
• Think of this idea of research as a quality system…
• Think of YOUR study…. what can you ensure is in place to
emphasize PREVENTION rather than waiting for problems
to arise?
• Well-written protocol • Detailed written SOPs
• Well-trained staff • Appropriate delegation of
• PI Oversight responsibilities
• A “culture” that promotes questions • Adequate monitoring
from staff when something doesn’t • Of subjects
“feel right.” • Of study conduct and processes
Activity 2
Instructions: Review the findings below and break into small groups to discuss the
next steps, including specifics of the CAPA. You may have to make some
assumptions on causes to come up with corrective/preventive actions, that is OK.

Please refer to Activity 2 handout.

Need assistance?
• BMC/BU Medical Campus
• Clinical Research Resources Office
• www.bumc.bu.edu/crro
• 617-358-7679
• BU/BMC Institutional Review Board (IRB)
• www.bumc.bu.edu/irb
• 617-358-5372

• UVM/ UVM Medical Center

• Office of Clinical Trials Research
• clinicaltrials@med.uvm.edu
• 802-656-8990
• UVM Research Protections Office
• IRB@uvm.edu
• 802-656-4050
Templates to Assist you in Documentation
On CRRO website
http://www.bumc.bu.edu/crro/
• Under Resources in the top menu
• Select Study Documentation Tools
• Scroll to Regulatory Files
• See Customizable templates to the right

On IRB website
http://www.bumc.bu.edu/irb
• CAPA template
• Go to INSPIR II in the right menu
• Click on IRB Templates under INSPIR II
Templates to Assist you in Documentation
On UMV website:
https://commons.med.uvm.edu/dean/c
omclntril/SitePages/Regulatory%20Guid
ance%20Resources.aspx
Helpful guidance
• Investigator Responsibilities - Protecting the Rights, Safety, and Welfare of
Study Subjects https://www.fda.gov/media/77765/download
• IRB Continuing Review after Clinical Investigation Approval
https://www.fda.gov/media/83121/download
• Oversight of Clinical Investigations - A Risk-Based Approach to Monitoring
guidance https://www.fda.gov/media/116754/download
• FDA Inspections of Clinical Investigators
https://www.fda.gov/media/75185/download
• FDA Inspectional Objectives for CAPAs
• https://www.fda.gov/ICECI/Inspections/InspectionGuides/ucm170612.htm
Thank you!