Thanks to numerous scandals, the FDA is under mounting pressure to step up its oversight of drug development and manufacturing. But what do the metrics look like? Well, the agency has now released numbers showing its efforts to inspect and audit facilities that are involved in drug development specifically, such as clinical trial investigators and institutional review boards.
So here is a summary: in fiscal year 2010, the agency conducted 792 inspections as part of its Bioresearch Monitoring Program, compared with 739 in fiscal year 2003. In each case, look-sees into clinical trial investigators accounted for half of that activity, underscoring the importance placed on ensuring that clinical trial work is handled correctly.
However, in fiscal year 2010, there were 400 inspections of trial investigators, down from 474 in the fiscal year 2009, although this figure was the largest annual tally since 2003. Of the 400 conducted in fiscal year 2010, 77 percent were the result of audits, with the rest due to complaints or referrals. By the way, 116 international investigators were inspected.
Not surprisingly, foreign inspections now occupy more energy. In fiscal year 2010, 70 percent of the investigators inspected were in the US, compared with 91 percent in fiscal year 2003. And 10 percent of the inspections were in Asia and the Pacific region, compared with 4 percent back in 2003. Inspections in Western Europe tripled, to 9 percent of the total.
However, when we eliminate the US from the data, Eastern Europe and Latin America are getting more attention which, of course, reflects the growing number of clinical trials taking place in those regions. Of 116 foreign investigator inspections, 34 percent in fiscal year 2010 were conducted in Asia, compared with 34 percent in 2003. The number of inspections of Eastern European investigators grew to 9 percent, from nothing in 2003. Latin American went to 14 percent from 9 percent.
Overall, there was a big increase in what the FDA calls an Official Action Indicated, rising to 10 percent in fiscal year 2010, from just 1 percent in fiscal year 2003. There were usually multiple deficiencies, but the vast majority involved protocols, recordkeeping, informed consent and accounting for drugs. Twelve percent involved fraudulent data, ahead of 7 percent for adverse events and 2 percent pertained to patient rights or welfare (you can see all the slides here).
6 Comments
any data on the % of clinical trial sites looked at?
Bill, an old rule of thumb used to be for FDA to do a 20% audit of the total patient enrollment, and then to get to that 20% they would, for the sake of efficiency audit the highest enrolling site. This could be problematic if the patients are not evenly distributed across sites.
Let say if you have have a 500 patient trial and you have 40 investigators enrolling 10 patients/site, and two investigators enrolling 50 patients/site. FDA would audit the two high enrolling sites to get their 20% (100 patients), and the remaining 40 sites would go untouched. Thus, in reality, only 4.8% of the total number of sites would be audited.
Usding this algorithm, you can see how it is easy for FDA to miss problems.
It woudl be interesting to see the % of clinical trial sites looked at.
OII, thank you!
Clinicaltrials.gov lists just over 95,000 trilas being conducted, other sites mention just under 12,000 principal investigators in the US (13,000 rest of the world). Clearly only a small percentage are being investigated.
@Rod Thorell - the most disturbing thing about that is that you can see that many sites that have been issued 483s were found to have the same issues on previous audits. So we're seeing a lack of response or change in how PIs are running trials when they've been through one of the few audits that occur so just imagine what's being missed at those sites that aren't audited (or know there are so few audits they don't feel a need to clean up their act)!