Reporting demographics in randomized control trials in spine surgery - we must do better

Abstract

Background contextDemographic factors contribute significantly to spine surgery outcomes. Although race and ethnicity are not proxies for disease states, the intersection between these patient characteristics and socioeconomic status significantly impact patient outcomes. PurposeThe purpose of this study is to investigate the frequency of demographic reporting and analysis in randomized controlled clinical trials (RCTs) published in the three highest impact spine journals. Study designSystematic review. Patient sampleWe analyzed 278 randomized control trials published in The Spine Journal, Spine, and Journal of Neurosurgery: Spine between January 2012 - January 2022. Outcome measuresExtracted manuscript characteristics included the frequency of demographic reporting, sample size, and demographic composition of studies. MethodsWe conducted a systematic review of RCTs published between January 2012 - January 2022 in the three highest impact factor spine journals in 2021: The Spine Journal, Spine, and Journal of Neurosurgery: Spine. We determined if age, gender, BMI, race, and ethnicity were reported and analyzed for each study. The overall frequency of demographic reporting was assessed, and the reporting trends were analyzed for each individual year and journal. Among studies that did report demographics, the populations were analyzed in comparison to the national population per United States (US) census reports. Studies were evaluated for bias using Cochrane risk-of-bias. ResultsOur search identified 278 RCTs for inclusion. 166 were published in Spine, 65 in The Spine Journal, and 47 in Journal of Neurosurgery: Spine. Only 9.35% (N=26) and 3.9% (N=11) of studies reported race and ethnicity, respectively. Demographic reporting frequency did not vary based on the publishing journal. Reporting of age and BMI increased over time, but reporting of race and ethnicity did not. Among RCTs that reported race, 88% were conducted in the US, and 85.71% of the patients in these US studies were White. White subjects were overly represented compared to the US population (85.71% vs. 61.63%, p<.001), and non-White or Black patients were most underrepresented (2.89% vs. 25.96%, p<.001). ConclusionsRCTs published in the three highest impact factor spine journals failed to frequently report patient race or ethnicity. Among studies published in the US, study populations are increasingly represented by non-Hispanic White patients. As we strive to care for an increasingly diverse population and reduce disparities to care, spine surgeons must do a better job reporting these variables to increase the external validity and generalizability of RCTs.