Analysis Highlights Importance of Including Demographic Data in Human Research Papers to Reduce Health Disparities
Leslie Capo, Director of Information Services
In the current issue of The Journal of Bone and Joint Surgery, the editors cite an LSU Health New Orleans study reporting that race, ethnicity, and socioeconomic status were infrequently reported
and analyzed within articles published in top orthopaedic journals. They thanked the
LSU Health authors for “shining a light on this important issue.”
The LSU Health New Orleans study identified and examined research articles involving human subjects published in 2019 in the two highest ranked US general orthopaedics journals – Clinical Orthopaedics and Related Research and The Journal of Bone and Joint Surgery. Articles were reviewed to determine study type reporting of race, ethnicity, and any of three socioeconomic status variables such as income, education, and health insurance, and the inclusion of these demographics in multivariable analyses.
Of the 156 articles that met inclusion criteria, 56 (35.9%) reported patient race and 24 (15.4%) reported patient ethnicity. Income was reported in 13 (8.3%) of the articles, education in 23 (14.7%) and health insurance in 18 (11.5%). Of the 97 papers that reported results of multivariable analyses, 30 (30.9%) included race in the analysis and 21 (21.6%) reported significance associated with race. Income, education, and health insurance were included in multivariable analyses in 7 (7.2%), 11 (11.3%), and 10 (10.3%) of the articles, respectively.
The authors wrote, “Failure to report key demographics makes it difficult for practitioners to determine whether study results apply to their patient populations. In addition, when orthopaedic interventions are evaluated without accounting for potential disparities by demographics, clinicians may incorrectly assume that the overall benefits and risks reported in studies apply equally to all patients.
The authors concluded that “this problem is not likely to improve substantially until orthopaedic journals require the transparent reporting of demographics. Ideally, descriptors for these variables should be standardized to enable consistent comparison or aggregation of the data.”
Co-authors included Andrew G. Chapple, PhD; Amy Bronstone, PhD; Peter C. Krause, MD; and Vinod Dasa, MD, of the Department of Orthopaedics at LSU Health New Orleans School of Medicine, as well as then LSU Health New Orleans medical students, Cody Crnkovic and Robert Quiring.
“Understanding the impact of inequities and health disparities first begins with recognizing there’s a problem,” says Dr. Dasa. “Until our researchers begin including these variables in their analyses, we will not truly understand the problems, let alone solve them.”