Study design: Retrospective analysis of administrative claims data in a single workers' compensation jurisdiction.
Objectives: To evaluate the effects of alternative definitions and follow-up parameters on rates of low back pain recurrence based on detailed administrative data.
Summary of background data: Previous studies reported low back pain recurrence rates ranging from 14% to 45%, without consistency in definitions of recurrence or specifications of follow-up.
Methods: Patients with new claims for low back pain reported in New Hampshire to a large workers' compensation provider in 1996 and 1997 were selected (N = 2944). Definitions of recurrence included: new workers' compensation claim, new episode of care, and new episode of lost work time (work disability). For the latter two definitions, various minimum between-episode gaps were applied and related to recurrence rates. Two follow-up structures (constant length of follow-up post end of the first episode and fixed-period length of follow-up since the onset of low back pain) were examined for sensitivity of recognizing low back pain recurrence, with a maximum of 3-year follow-up.
Results: Recurrence rate using a claims-based definition was 7.9% and 7.1% for the entire cohort and the subset with work disability days, respectively, for the 3 years of follow-up. Care-based recurrence rates ranged between 12% and 49%, whereas disability-based recurrence rates ranged between 6% and 17% over the 3 years, inversely related to the length of the minimum between-episode gap (R = -0.86 for disability and care, P < 0.001). Two-year follow-up was sufficient to identify 85% to 100% of recurrences regardless of the follow-up structure.
Conclusions: Recurrence rates are highly sensitive to variations in definitions. Consistency of definitions and application across studies is required to enable valid comparisons.