Medical Device Daily Washington Editor
FDA's Aug. 12 warning letter to General Electric Integrated IT Solutions (GE-ITI; Barrington, Illinois) highlighted the difficulty large firms have keeping track of all operations. In this case, computer code was at the heart of most of the GE-ITI problems, and coordination of overseas operations with those in the U.S. could have saved the company at least one of the findings on the warning letter.
The agency cited the company for failure of procedures for corrective and preventive action (CAPA) to require collection of data from "other GE Healthcare sites that share the same product components (such as shared software code)." FDA states in the warning letter that in October 2007, GE's subsidiary in France "submitted a correction and removal report for safety issues identified in their Advantage Workstation product" and that a subsequent version of the software was not coded to account for the problems identified in those reports.
FDA deemed inadequate the response in GE's June 2 letter apparently because of lack of data. FDA states that GE promised to "conduct a retrospective review of all available functional area trend analyses from all relevant data sources" from June 2006 to May 2008, and FDA requested a copy of the report on the review.
The agency cited the company for failure of procedures for finished device acceptance to include a check on the function of CD/DVD burners "to ensure that the software matches the engineer master files." FDA requested a copy of a report of a retrospective analysis of such acceptance activities, which was apparently scheduled for completion by July 31.
The warning letter cited GE for complaint procedures in connection with a file "opened on Nov. 16, 2007, because a customer ... stated that CT exam reports were being assigned to incorrect exams." According to FDA, GE provided the customer with a work-around for the problem and decided the problem required no corrective/preventive action. The warning letter stated that GE investigated the problem during the course of the inspection, which disclosed "a fix for this issue ... made in April 2005."
The agency deemed adequate GE's response to this finding, which included training of on-site personnel in CAPA procedures and FDA stated it would confirm the changes on the next inspection.
The letter also identified several instances in which GE systems allowed mixing of patient data, including one instance in which "non-identifiable patient information [was] added to an incorrect file" as a medical device report. Another finding on the letter was that GE did not report two software malfunctions as medical device reports from July and August 2007. As was the case with the first citation, FDA requested documentation of a retrospective review of the incidents.
At press time, GE had not responded to a call for comment.
IRB snared in warning letter dragnet
Also receiving a warning letter recently was the institutional review board for Valley Baptist Medical Center (Brownsville, Texas), which the agency cited on Aug. 14 for at least three instances in which the IRB did not review at least once a year ongoing studies of high-risk devices.
In one study, initially approved by Valley Baptist in 1999, "the FDA investigator found only one record of annual review and renewal of IRB approval, which is dated Jan. 11, 2008."
FDA also stated that the IRB had no documents regarding review of another study, which commenced in March 2006. Valley's offering of a tracking spreadsheet fell short because it did not indicate a preventive action, and FDA requested "a complete listing of all clinical studies reviewed by the IRB since 1999."
Another citation notes that the IRB lacked documentation showing that an unspecified number of studies had obtained informed consent from patients. Valley Baptist's April 30 response did not pass muster because "simply stamping all consent forms with 'IRB approval' will not bring the IRB into compliance with regulations."
The facility also also cited for having no procedures for reporting "any unanticipated problems involving risks to human subjects" or for reporting "any instance of serious or continuing non-compliance with regulations or the requirements or determinations of the IRB."
The warning letter indicated that on at least two occasions, Valley Baptist's board members voted on matters related to studies when less than a quorum was present, but many details were redacted from the letter.
FDA said that among the informed consent problems experienced by the IRB is that "there is no record of IRB review or approval of informed consent forms for at least [redacted] studies reviewed by the IRB since 2003." In the case of a study the name of which is not given, FDA stated, "the only version of the informed consent reform in the IRB files is missing the name of the investigator, the IRB chairman and relevant contact information. FDA indicated no direct response to this finding from Valley Baptist.
AHRQ seeks input for innovation
The search may be on for an independent entity to promote best healthcare practices, but the Agency for Healthcare Research and Quality has nonetheless issued a call for ways to accelerate the adoption of research "that has the potential for high impact."
The Innovations Portfolio is geared to generate ideas "substantially different from those already being pursued by AHRQ" and will focus on "problem solving in order to accelerate improvement in healthcare." Interested parties should respond to the request within 60 days of the posting of the announcement.