Predictive Model for Flexion Pattern of Chronic Low Back Pain Among Office Workers

Abstract Background The purpose of this study was to establish a prediction model for classifying the presence of flexion pattern chronic low back pain (CLBP) in office workers, including physical, individual, psychological, and occupational factors. Methods Ninety-five office workers (48 workers with CLBP and 47 workers without CLBP) participated in this study. Hip flexion ROM (HFR), knee extension ROM, knee extension ROM with ankle dorsiflexion, hip internal rotation ROM (HIR), hip flexor strength, pelvic posterior tilt angle (PPTA) in habitual sitting, and trunk extensor endurance were measured. Twelve variables (9 physical variables and age, sex, body mass index, Beck Depression Index, and working duration) were used to develop the prediction model for the presence of flexion pattern CLBP using logistic regression. Results The probability of the presence of flexion pattern CLBP significantly decreased with an odds ratio of HFR 0.884 (95% confidence interval CI = 0.817―0.957) and HIR 0.860 (95% CI = 0.799―0.926), but it increased with an odds ratio of PPTA in habitual sitting 1.190 (95% CI = 1.058―1.339). Our model showed acceptable accuracy of 82.1% and an area under the receiver operating characteristics curve of 0.898 Conclusions HFR, HIR, and PPTA in habitual sitting could serve as guidelines for preventing and managing flexion pattern CLBP in office workers.