CRS-USA LASIK 8/18/99 Clinical Trial Software Validation
There was no verification that the data submitted by the clinical investigators matched source data at the site. The only data verification was performed by the computer data service, xxxx and involved scrutinizing the data for such errors as out-of-range data, field entry omissions, exam dates out-of-range, and missing exams. This process could not verify that the data entered into the computer system was an accurate record of the source data at the site. According to 21 CFR 860.7, it is the responsibility of the applicant to assure that FDA receives valid scientific evidence of the safety and effectiveness of a device. 21 CFR Part 814, Premarket Approval of Medical Devices, references section 860.7 as criteria that will be used in deciding whether to approve or deny approval to a PMA.
It is noted that the protocol for both the initial study and the study expansion states, under Data Reporting Procedures, Access, and Limitations, that “The sponsors have do not assume responsibility for monitoring individual results.” (It is assumed that the typographical error is the addition of “have.”) Such a statement cannot serve to release a sponsor from their responsibility. Data validation is part of monitoring and, as noted above, Part 812 gives the sponsor the responsibility for assuring that a clinical study is properly monitored.
Lack of documentation for software validation. Dr. Guy Kezirian, study coordinator, stated to Mr. Gonzalez during the inspection that he has never seen any validation of the software package used by4~. Enclosed is a copy of the FDA guidance document on “Computerized Systems Used in Clinical Trials.”
SoftwareCPR keywords: software validation,
