Study design: Inception cohort study.
Objective: To provide the first reliable estimate of the 1-year incidence of recurrence in subjects recently recovered from acute nonspecific low back pain (LBP) and to determine factors predictive of recurrence in 1 year.
Summary of background data: Previous studies provide potentially flawed estimates of recurrence of LBP because they do not restrict the cohort to those who have recovered and are therefore eligible for a recurrence.
Methods: We identified 1334 consecutive patients who presented to primary care with acute LBP; of these 353 subjects recovered before 6 weeks and entered the current study. The primary outcome measure was recurrence of LBP in the next year. Specifically, an episode of recurrence was defined in 2 ways: recall of recurrence at the 12-month follow-up and report of pain at the 3- or 12-month follow-up. Risk factors for recurrence were assessed at baseline. Pain intensity was assessed at 6 weeks, 3 months, and 12 months and recurrence at 12 months. Factors that could plausibly affect recurrence were chosen a priori and evaluated using a multivariable regression analysis.
Results: Recurrence of LBP was found to be much less common than previous estimates suggest, ranging from 24% (95% CI = 20%-28%) using "12-month recall" definition of recurrence, to 33% (95% CI = 28%-38%) using "pain at follow-up" definition of recurrence. However, only 1 factor, previous episode(s) of LBP, was consistently predictive of recurrence within the next 12 months (odds ratio = 1.8-2.0, P = 0.00-0.05).
Conclusion: This study challenges the assumption that the majority of subjects will have a recurrence of LBP in a 1-year period. After the resolution of an episode of acute LBP, about 25% of subjects will have a recurrence in the next year. It is difficult to predict who will have a recurrence within the next year.