Process error rates in general research applications to the Human Research Ethics Committee (Medical) at the University of the Witwatersrand: A secondary data analysis

Abstract

Objective. To examine process error rates in applications for ethics clearance of health research. Methods. Minutes of 586 general research applications made to a human health research ethics committee (HREC) from April 2008 to March 2009 were examined. Rates of approval were calculated and reasons for requiring revision or non-approval of the applications were grouped into eight categories. Results. Of the applications, 37% were approved at first evaluation; minor revisions were required for 56% and major revisions for 3%, while 4% were not approved. Eventually 69% of the 586 applications were approved. Surprisingly, 28% were removed from the Committee agenda because of no response from the applicants. Of the 607 instances of process error in 369 applications requiring revision or that were not approved at first evaluation, difficulty with consent documents (55%) and missing information (43%) were the most frequent; the remaining 6 types ranged in frequency from 3% to 17%. Conclusion. It is suggested that the process errors seen could be reduced in rate if applicants were to show a draft of their application to an HREC member or experienced researcher before submission.