Audit program addresses systemic problems
Find the root cause and prevent recurrence
Clinical trial audits typically lead to better research compliance, but the best strategy might be to look specifically for systemic problems.
The one-time mistake is the low-hanging fruit — easy to identify and correct. But the systemic problems are what could cause long-term compliance troubles, says Dawn Lowe-Gooden, BA, MS, CQA, CTBS, research compliance manager for the office of research compliance and regulatory affairs at the Cincinnati Children's Hospital Medical Center in Cincinnati, OH.
Systemic problems could be problems a principal investigator has with multiple studies he or she is conducting, or it could be a problem identified in different studies within a division.
"We look for systemic problems within a division because usually things happening within a division are the same," Lowe-Gooden says.
The Cincinnati Children's Hospital Medical Center has three, fulltime research compliance specialists who perform good clinical practice (GCP) audits.
"We call them 'quality reviews,' to soften the word 'audit,'" Lowe-Gooden says. "The reviews are both for-cause and random."
The for-cause reviews, which are less common, typically result when someone raises questions anonymously about a particular clinical trial, she adds.
Compliance officials could make the most use of staff and resources by limiting random audits to studies that meet one of the following criteria:
- studies that have more than minimal risk;
- studies involving drugs or devices;
- studies with large numbers of participants;
- studies with many and/or complex procedures;
- studies initiated by investigators;
- phase III studies and others that are critical for approval.1
"We attempt to do one audit per week per auditor, minus vacation time and holidays," Lowe-Gooden says.
The quality reviews typically require pre-audit preparation, one to three days at the site, including interviews with PIs and trial coordinators, a post-review report, and a response from the PI.
Here is how the GCP quality reviews work:
• Create audit log: The randomly selected study is recorded in an audit log, and then Lowe-Gooden writes a letter to the principal investigator (PI) and study coordinator to notify them of the audit.
The letter outlines what the PI will need to have prepared and suggests a date for the review. The date can be changed according to the PI's schedule, Lowe-Gooden says.
"Once we select a date that the PI agrees with, the enrollment log is sent to the research compliance specialist who determines which subject files to review," she says.
The number of files reviewed is determined by the square root plus one.
"Say you have 100 subjects in a study," Lowe-Gooden explains. "The square root of 100 is 10 and you add one, so you review 11 subject binders."
Since it would be too time-consuming to review every file, this provides a fairly representative sample, she adds.
• Use a checklist when looking at IRB files: The research compliance specialists then review the IRB files relating to the study and modify the generic checklist to make it study specific.
"They use the checklist as a guide, in conjunction with their own questions," Lowe-Gooden says.
"If you simply go through the checklist questions, you don't get the full answer," she says. "If you ask, 'Was informed consent performed,' the simple answer is, 'Yes.'"
But that "yes" answer doesn't tell the whole story, so research compliance specialists are encouraged to ask open-ended questions, such as, "Explain your process of informed consent."
With the latter question, the checklist answer would be "yes," however a clinical trial coordinator's answer could also be a poor answer if the coordinator had said, "We gave him the informed consent form; he signed it, and then we drew blood," Lowe-Gooden says.
The research compliance specialist would prefer to hear a very good answer, such as, "We gave the subject the informed consent form and gave him time to read it; then we went over the highlights and asked him questions to make sure he really understood it," she suggests.
Another checklist question is whether the participant met initial inclusion/exclusion criteria.
The open-ended question would be, "How do you determine that the participant met initial inclusion/exclusion criteria?"
The good answer would be, "We have a checklist that lists what all inclusion/exclusion criteria are, and we go through it with each participant," Lowe-Gooden says.
Then the investigator will document how the subject met each of these criteria, by listing weight, blood pressure, or whatever is required.
A poor answer would be, "Well, the person came into our clinic, and so they must be eligible," Lowe-Gooden says.
Each day during the quality review, the research compliance specialist sends out a follow-up email discussing the daily findings and giving the PI an opportunity to find any missing documentation or answer additional questions.
• Write a post-audit investigational report: Initially, the compliance office gave PIs a single or two-page report that listed only the problems found.
"The principal investigators were not happy with this," Lowe-Gooden recalls. "They wanted a history of all the things they did right."
So Lowe-Gooden changed the format to be more descriptive and to include positive findings, and reports now run up to eight pages in length.
The reports are structured to address eight criteria: eligibility, informed consent, adherence to protocol and safety measures, adverse events, documentation, laboratory, regulatory, and test article/investigational drug or device.
"We detailed what the site did, what's done well, who reviewed it, how many subjects were reviewed, and the last two pages discuss the observation or nonconformance findings," Lowe-Gooden says.
"There are two types of findings: they are either observation or nonconformance," she explains. "Observation findings are one-time events that we don't believe will lead to other issues."
But if there are three or more findings that are related, then that's considered a systemic problem and a nonconformance finding, she adds.
"Some findings are serious enough that they fall into nonconformance as a single finding," Lowe-Gooden says.
With each citation of a nonconformance finding, the report lists a reference to the regulations, protocol language, accreditation rules, institutional standard operating procedures (SOPs) or policies, or other guidelines that were violated.
"We include those so they can see what they violated so that they can understand why it's a violation," Lowe-Gooden says.
• Give PIs time to respond: PIs have 30 days to respond to the report unless there is a serious finding that requires more immediate action, she says.
"If it's so serious that a person's life is in danger, we'll contact the PI or IRB immediately," Lowe-Gooden says. "As auditors, we point things out, but we have no authority to change or make people do things differently."
By the end of the 30 days, PIs should submit a corrective action plan (CAP) that includes identification of the findings' root cause, she adds.
If they neglect to include the root cause, then the research compliance specialist will ask them to explain why it was a mistake: "Were they not trained properly? Or why don't you have an SOP to address that issue?" Lowe-Gooden says.
"We've had to go back and discuss root causes with several PIs that we've audited," she notes.
Without taking time to determine the root cause, a CT site might correct a mistake, but be unable to fully prevent it from happening again.
Once the PI sends an acceptable response, then the audit is closed.
"We will do a follow-up if there are many issues to address," Lowe-Gooden says.
Also, if systemic problems and trends are identified, the compliance office will provide additional training to the site staff or division staff, she adds.
The GCP quality reviews have generated buzz among clinical trial investigators and coordinators, and there is anecdotal evidence that they've resulted in a decreasing number of audit findings, Lowe-Gooden says.
Reference:
- Lowe-Gooden D. GCP compliance: best audit principles. GCPJ. April, 2009; www.GCPj.com.
You have reached your article limit for the month. Subscribe now to access this article plus other member-only content.
- Award-winning Medical Content
- Latest Advances & Development in Medicine
- Unbiased Content