Measuring self-reported recall of unintentional musculoskeletal injuries in an Army Airborne Division

Background/Purpose: Self-reported data are often used in epidemiology, but self-reported recall of comprehensive injury data has not been measured among soldiers. The aim of this analysis was to assess self-reported recall of unintentional musculoskeletal injuries among soldiers in an Army Airborne Division.

Methods: Self-reported and medical chart-reviewed injuries among 115 soldiers (age 26.6 ± 5.8 years (mean ± SD), 87.0% male) were matched by anatomic location, side (for extremity injuries), year, and type. The injuries included in the analysis were those that had occurred during the year of survey (recent injuries), and during the preceding calendar year (old injuries). Recall was expressed as the percent of medical chart-reviewed injuries correctly recalled in the self-report. Proportions were compared using the Fisher's exact test.

Results/Outcomes: Eighty-seven injuries were recorded in the medical charts. Common injury types were pain/spasm/ache (29/87, 33.3% of the injuries), sprain (18/87, 20.7%), and strain (15/87, 17.2%). Overall, recall was low (9/87 = 10.3%). Recall was higher for severe injuries (traumatic/stress fractures, 1/4 = 25.0%) as compared to less severe injuries (non-fracture injuries, 8/83 = 9.6%), but the difference was not statistically significant (p = 0.359). Recall was higher for recent injuries (3/26 = 11.5%) as compared to old injuries (6/61 = 9.8%), but the difference was not statistically significant (p = 1.000).

Conclusion: The low self-reported recall in this study underscores the need for further investigation of factors affecting recall and strategies to improve accuracy of recall of injury data in various military populations.