The most award winning
healthcare information source.
TRUSTED FOR FOUR DECADES.
Self-audits will help prevent major problems
PIs need to take full responsibility for site activity
Clinical trial sites and investigators should not rely on clinical research organization (CRO) monitors to find systemic problems. Instead, they should be proactive with their own in-depth audits, an expert suggests.
When investigators rely on the reports of superficial problems reported by monitors, they risk missing the systemic problems that cause the symptoms often reported by the monitors, says Tamera Norton Smith, PhD, MT (ASCP), president and senior consultant of Norton Audits Inc. of Lexington, SC. Smith has more than 17 years of clinical trial experience and, formerly, was a Food and Drug Administration (FDA) investigator. Smith was a scheduled speaker at the Association of Clinical Research Professionals (ACRP) Global Conference and Exhibition, held April 20-24, 2007, in Seattle, WA.
"To me, as an auditor, when I find things on the surface and observe the symptomology, the problem is really three times worse," Smith says.
Site monitors will send out a monitoring report, listing items identified, and investigators often do not take corrective actions based on the monitor's findings, she says. (See chart on sample critical findings.)
"This is a huge risk because monitoring letters are available to the FDA to review," Smith explains. "If monitoring reports are sitting in files with all of these issues that haven't been corrected, then it's a major concern."
Even better, a principal investigator should be proactive and conduct thorough self-audits before problems draw the attention of regulatory officials and negative findings prompt a random audit occur.
For example, a physician investigator in Atlanta, GA, had worked with Norton Audits for several years to improve research compliance, Smith recalls.
"He'd had some bad experiences with sponsor organizations and knew enough to want his own external audit, so we did this audit three years ago," Smith says. "He has turned around his business and knows the regulations better than any physicians I know, and he knows enough to protect himself from making the same mistakes he made in the past."
Recently, the investigator was inspected for more than two months by FDA investigators, who looked at three different studies, and he came out of that intense scrutiny with only two findings for which he had already put corrective actions in place and the problems had been resolved, Smith explains.
"It was amazing that he came through this audit so well," Smith adds. "If he had not been conducting self-audits and made all the corrections and procedures, then he might have had problems."
While the fact that the audit occurred at all was a major inconvenience to the investigator, the good news is that a thorough look at his records turned up nothing of significance, she says.
In another case, Smith worked with an investigator in Florida who had had 27 monitoring visits over a four-year period, and not a single one identified the bigger issues, Smith says.
"The monitors did have some issues in their monitoring reports, but the PI didn't even sign off on his monitoring report that he'd seen these issues," Smith says.
For these reasons, self-auditing is necessary.
Smith offers these key guidelines for how to conduct and improve auditing and compliance at a research site:
1. Use good clinical practice (GCP) and the Corrective And Preventive Action (CAPA) system.
Staff need to be taught both GCP and CAPA, Smith says.
"Sponsors have investigator meetings where they train them on the protocol and, most of the time, the GCP training is maybe 15 minutes long," Smith says.
Instead, investigators and clinical trial staff need thorough training and education in GCP, all research regulations, and how to conduct informed consent, as well as document properly, she says.
Under the CAPA system, a site will first identify what the problem is, analyze the root cause of why it happened, come up with a corrective and preventive plan, and monitor/audit the process.
Also, it's important to have a buddy system put in place where one study coordinator is looking over the shoulders of another one, she adds.
"They need to learn how to audit themselves and not rely on monitors," Smith explains. "If you are depending on monitors to tell you how you're doing, then they're only finding surface issues and not looking at the root causes that will help them improve their organization."
For example, during the informed consent process, there should be another person, who was not involved with the informed consent, checking all of the signatures on the informed consent form and making sure everything is dated properly and that all of the pages are there, she says.
This could be accomplished with a checklist, Smith adds.
"Be proactive rather than reactive," Smith says
"Most principal investigators will have more than one study coordinator," she explains. "If one is assigned to a study, then a different one will look behind her and do quality control checks, looking at the other coordinator's informed consent or regulatory binder."
2. Create standard operating procedures (SOPs) individualized for the organization.
"Many investigators think they can buy SOPs from any company and be compliant," Smith says. "But they need to know how to follow them, organize the site, and conduct a self-audit."
SOP development and training should be thorough so that investigators have greater skills in this area than do monitors, she says.
"We make SOPs for investigators, but we won't give them to them without the online training," Smith says. "They need the online training to know how to do the procedures."
3. Emphasize investigator duties and responsibilities.
Another goal is to make certain the investigator clearly understands his or her responsibilities, Smith says.
"Most investigators turn over their responsibilities to the clinical research coordinator, but they can't turn over all of the responsibility because they're responsible for the trial," Smith says.
Those who delegate too much of their duties and responsibilities run the risk of having negative findings during an FDA audit, she notes.
For example, the FDA had these comments for an investigator whose site was investigated in 2004, according to a letter written, Oct. 21, 2005, by the director of the FDA's Office of Compliance, Center for Devices and Radiological Health:
• "You failed to adequately supervise the conduct of the study. [21 CFR 812.100, 812.110].
"When you signed the Investigator's Agreements for the above-referenced clinical investigations, you agreed to take on the responsibilities of a clinical investigator at your site. Your general responsibilities include ensuring that an investigation is conducted according to the signed agreement, the investigational plan and applicable FDA regulations, for protecting the rights, safety, and welfare of subjects under the investigator's care, and for the control of devices under investigation [21 CFR 812.100]. The Investigator's Agreements that you signed required that you or your sub-investigators personally supervise all testing of the device involving human subjects. In addition, regulations provide that an investigator will permit an investigational device to be used only with subjects under the investigator's supervision. An investigator shall not supply an investigational device to any person not authorized to receive it, in accordance with 21 CFR Part 812 [21 CFR 812.110(c)]. Although you may delegate study tasks to individuals qualified to perform them, you may not delegate your general responsibilities as a clinical investigator.
"You failed to supervise the study, so as to ensure that your general responsibilities were fulfilled. As detailed in charges 2-6 below, there were numerous violations such as ones involving informed consent, including the falsification of your signature on informed consent documents; protocol violations, including enrollment of patients not meeting eligibility criteria and problems with follow-up visits; and record keeping violations, including the failure to maintain device accountability records. Despite the widespread nature of these problems, in most cases, you made no effort at correction until action was requested by study monitors or your IRB.
"In your responses to the two Form FDA 483s issued to you, you admitted generally that the noted deficiencies did occur, but in those responses, as well as in your statements to the FDA investigator and in your correspondence with your IRB, you repeatedly attributed them to poor performance and lack of experience by research staff, including your former study coordinator, and in some cases, to poor oversight by the study monitor. While your 483 responses also indicate that those staff have been replaced, that you yourself are no longer the principal investigator for these studies, and that new procedures have been implemented, these responses do not excuse your failure to adequately supervise the conduct of the studies during the period in which you were the principal investigator......"
When the FDA has findings, such as in the above case, it means the site will be suspended from conducting that type of research, and the principal investigator could lose his or her medical license, Smith notes.
"The PI blames the study coordinator for most of the noncompliance, but the law says he's the most responsible person, so you can't delegate your responsibility for the trial," Smith says.
Smith was teaching a compliance course when a number of study coordinators told her that it was okay that they handled many of the details of a clinical trial.
"I said, 'It's okay for you to do them, but not without being supervised,'" Smith recalls. "If you have a coordinator who does bad things, then the PI has no one to blame but himself."
4. Look at critical observations on audit report.
When sites conduct self-audits, or when monitors write reports, the most efficient way to document problems is to use the FDA's deficiency codes, Smith says.
For instance, a monitor might say that the informed consent is missing, but what they should report is that there is a violation of code 01, and there are no records of informed consent available, Smith explains.
"It sounds like an administrative issue, but it means the patient wasn't properly provided informed consent, which should be documented, also, as a violation of deficiency code 02 — 'Failure to obtain and/or document subject consent,'" Smith says.
It would also be very helpful if these audits and monitoring reports referred to CFR codes when problems are noted, she adds.
"Monitors right now are so far away from being able to do that, but hopefully, in the next three to five years, we can get most of the industry to do that," Smith says.
Here are some examples of deficiency codes and the federal regulations reference numbers:
"Most of the time, I learn from investigators that there's not a lot of intent to do research the wrong way," Smith says. "It's about education, because very few of them are deliberately or repeatedly trying to be noncompliant."