Background Low back pain is a prevalent injury among army soldiers during military training and is also a significant cause of non-combat casualties in the army during peacetime.

Objective To explore the factors that contribute to low back pain among military personnel and develop a model for predicting the associated risks.

Methods From February to April in 2024, stratified random cluster sampling was used to enroll officers and soldiers of the Army as research subjects, and univariate analysis, Lasso regression analysis, and multivariate logistic regression analysis were used to determine the independent influencing factors of low back pain and construct a prediction model. The diagnostic efficacy and clinical application value of the prediction model were evaluated by receiver operating characteristic curve, calibration curve and decision curve analysis.

Results The incidence of low back pain among 1 945 army officers and soldiers was 38.66% (752/1 945). Multivariate logistic regression analysis results showed that military age, sleep duration, smoking, anxiety, prolonged sitting, long-term standing, long-term walking, long-term bending during training, carrying or carrying objects weighing more than 5 kg, cold wind or significant temperature changes, frequent short-term but maximum-strength movements, and frequent stretching and relaxation activities were independent influencing factors for low back pain (P<0.05). A low back pain risk prediction model was constructed. ROC curve analysis showed that the area under the curve (AUC) was 0.774 (95% CI: 0.743-0.816). The prediction curve was basically consistent with the standard curve. The results of the Hosmer-Lemeshow goodness of fit test showed that the model had good consistency (P>0.05). The decision curve suggested that when the model prediction probability threshold was in the range of 13% to 83%, the prediction model showed good clinical applicability.

Conclusion The prevalence of low back pain among army soldiers is high, and the onset is affected by military age, sleep duration, anxiety and long periods of sitting. The nomogram prediction model constructed based on this has good discrimination and clinical applicability.