Research committee approval for use of the Behavioral Risk Factor Surveillance System data: a cross-sectional study

Background: Research involving medical diagnoses, comorbidities, drugs and alcohol use, and physical activity often requires ethical approval from institutional review boards (IRBs) prior to initiation, with good reason. Historically, there have been many experimental studies that have unethically treated human research participants—several that have performed extreme atrocities. However, the US government national survey data, such as the Behavioral Risk Factors Surveillance System (BRFSS); which is often used for public health research. Funded by the Center for Disease Control (CDC), data from BRFSS has no individually identifiable information, and does not meet the requirements of human subjects research as defined by the Department of Health and Human Services. However; there is no standardization of IRB guidelines regarding ethical review of non-HSR, and universities and journals vary on what is perceived to warrant ethics review. As no study has investigated the use of BFSS and rates of IRB submission, the primary aim was to evaluate the frequency of ethics review among cross-sectional studies of BRFSS and their associated study characteristics.

Methods: A systematic search of PubMed was conducted for publications that included ‘BRFSS’ or ‘Behavior and Risk Factor Surveillance System’ in the title or abstract published after 2011. Studies were eligible for inclusion if they include a cross-sectional examination of the specified dataset. Editorials, comments, or methodological examination of the survey design, sampling methods, missing data, etc., including data simulation studies, will be excluded. Data extraction was conducted by two authors in masked, duplicate fashion, in accordance with best practices as noted in the Cochrane Handbook.

Results: Of 154 included studies, 34 stated it was submitted to an IRB, 38 reported not having submitted the study for ethical review, and 54 studies included no mention of ethics review. Only 2 of the 34 studies submitted for review were published in journals that require an ethics determination, and none of the universities or institutions that approved the studies had such a requirement in their research guidelines. We found a statistically significant association between study funding and IRB submission (X2 = 16.5443, P = 0.002). Among the 60 funded studies, 22 (36.7%) reported being submitted to an IRB, while only 2 (5.0%) of the 40 studies reporting no funding and 10 (18.5%) of the 54 that did not include a funding statement were submitted to an IRB. We found no significant differences among studies and ethical approval from before and after 2019 (X2 = 0.0264, P = 0.871), with approximately 30% of the studies having declared they were submitted to an IRB.

Conclusions: BRFSS is not categorized as human subjects research. Yet, despite this information, 32 of 34 (94.1%) cross-sectional studies of secondary BRFSS data were submitted for IRB approval, with little discernible reason as to why. Submission of non-HSR can be burdensome to universities, and varied guidelines may cause confusion among ethics in research among junior researchers. Public availability of large datasets provides many advantages including cost-effective, preliminary studies, that can lead to clinical research investigations that require substantially more resources.