A National Survey of Orthopaedic Residents Identifies Deficiencies in the Understanding of Medical Statistics

Abstract

Background: Biomedical research is essential for optimizing patient care. Research has suggested inadequacies in nonorthopaedic trainees' understanding of study design and biostatistics. This study assesses orthopaedic residents' knowledge of common biostatistical and study design concepts, as well as their confidence in utilizing the medical literature. Methods: A validated survey assessing knowledge and the application of study design concepts was administered to residents at 10 U.S. institutions. The survey tested knowledge as well as confidence and attitudes regarding common biostatistics principles. The association of demographic characteristics, work activities, and confidence and attitude ratings with test performance were examined using t tests and analysis of variance. Results: The survey response rate was 64% (178 of 279). The largest group of participants were men (83%, 137 of 165), were between the ages of 26 and 30 years (59%, 105 of 177), and had graduated medical school within the past 4 to 10 years (43%, 76 of 175). Fifty-three percent (93 of 176) had prior biostatistics training, while 44% (77 of 176) had prior epidemiology training. Less than 5% of biostatistics or epidemiology training had taken place after medical school. Forty-seven percent (83 of 176) were unable to determine a study's design. Thirty-eight percent (67 of 178) could not apply the concept of specificity and sensitivity. Eighty-three percent (147 of 178) could not assess the strength of a relationship using odds ratios. Sixty-nine percent (123 of 178) understood the implications of p values. Previous biostatistics training, but not epidemiology or evidence-based medicine training; inclusion of reading research, attending conferences, and data analysis; as well as a self-reported finding of statistics as important for the analysis of one's own research data were significantly associated with better test performance (p < 0.05). Conclusions: Notable deficits exist in orthopaedic residents' biostatistical knowledge. Greater emphasis is needed to improve biostatistics and research design training. The impact of biostatistics knowledge and/or aptitude on clinical decision-making is an area of suggested research.Research and statistics are essential for optimizing patient care through the production and consumption of sound research. The current literature suggests inadequacies in nonorthopaedic trainees' understanding of the science of research and statistics because of poor development, maintenance, and monitoring of these skills [1][2][3][4][5][6][7][8][9][10][11] . It has been reported that 30% to 90% of the medical literature contains erroneous application and interpretation of statistical methods 6,7,12 . As the reliance on and requirement for evidence-based medicine (EBM) continues to grow, it is crucial that physicians are confident and competent in their *Additional members of the Collaborative Orthopaedic Educational Research Group are listed in a Note at the end of the article. Disclosure:The authors indicated that no exter...

Statistical Assumptions in Orthopaedic Literature: Are Study Findings at Risk?

“…Our data demonstrate the lack of reported assumptions in the published orthopaedic literature, which is consistent with the lack of understanding of medical statistics amongst orthopaedic residents [ 20 ]. It remains possible that the underlying assumptions were investigated, but not reported.…”

A Guide to Basic Statistics for Educational Research

“…19 Statistical content was determined based on the low literacy regarding these concepts seen in prior studies of residents and educators. [8][9][10][11][12][13] A video of the PowerPoint presentation contained in this seminar is available in Appendix A. All figures in the presentation were created by the author using Stata statistical software version 14.2 (StataCorp) from fabricated data for illustrative purposes only.…”

“…3 A recent Association of American Medical Colleges survey showed that a majority of medical schools lack specific biostatistics training. 7 This lack of training can contribute to low statistical knowledge among resident trainees [8][9][10][11] and subsequently to low statistical literacy among faculty. [11][12][13] A recent scoping review of clinician-educator faculty development programs found that few programs focus on research or scholarship skills.…”

Analysis of Statistical Knowledge of Peruvian Medical Students: A Cross-Sectional Analytical Study Based on a Survey

“…Knowledge specifically regarding the p-value has been studied in isolation, showing a low amount of positive response [22][23][24]. Nevertheless, in the study of Araoye et al [25], 69% made an adequate interpretation of said value. On the other hand, advanced methods, such as survival analyses including Cox proportional hazards regression and Kaplan-Meier analyses, require more advanced knowledge, which would explain the reason why it also had a percentage of correct answers [26].…”