Patterns and trends in quality of response rate reporting in case-control studies of cancer

Abstract

Purpose: We assessed the quality of reporting of response rates in published case-control studies of cancer over the past four decades.

Methods: We reviewed all case-control studies of cancer published in twelve major epidemiology, public health, and general medicine journals in four publication periods (1984-86, 1995, 2005, and 2013). Information on study base ascertainment, data collection methods, population characteristics, response rates, and reasons for non-participation was extracted. Quality of response rate reporting was assessed based on the amount of pertinent information reported, and in particular, numbers of non-participants by reasons for non-participation. We calculated subject response rates by quality of response rate reporting.

Results: A total of 370 studies met the eligibility criteria, yielding a total of 370 case series and 422 control series. Overall, the quality of reporting of response rate and reasons for non-participation was poor. There was a tendency for better quality of reporting of case series, followed by population control series, and lastly by medical source control series. Quality of reporting declined from 1995 to 2013.

Conclusion: The reporting of relevant information on response rates in case-control studies of cancer has been rather poor, and it has not improved over time. This compromises our ability to assess validity of studies’ findings.