Predictors of early improvement in low back pain amongst consulters to general practice: the influence of pre-morbid and episode-related factors

Low back pain symptoms are extremely common, affecting as many as 80% of th e population at some time in their lives. However, the majority of the medi cal costs arise from the minority of patients whose symptoms become chronic . The authors propose a model in which chronicity is determined not only by factors related to the episode of low back pain, but also factors prior to the onset of symptoms (premorbid factors). No previous study has collected information on predictors of low back pain chronicity prior to the onset o f symptoms. Participants in the South Manchester Low Back Pain Study, recru ited by means of a cross-sectional population survey were followed prospect ively over 18 months to identify those who consulted their general practiti oner with a new episode of low back pain. At interview, 1-2 weeks post-cons ultation, it was determined whether or not subjects' symptoms had improved. In males, low levels of psychological distress, a higher than average repo rted level of physical activity, being in employment, and being satisfied w ith current work status were associated with a quick improvement in symptom s. In addition factors related to the episode, namely a short duration befo re consultation and symptoms with a sudden onset and confined to the lower back area, also strongly predicted a good early outcome. Using information on both pre-morbid and episode related factors, groups of male patients wer e identified whose probability of an early resolution of symptoms ranged be tween 0.25 and 1. Few factors, either pre-morbid or episode-related, were s trongly associated with outcome amongst females. This large population-base d study has shown, despite the known heterogeneity in the origin of low bac k pain and the pathologies associated with symptoms, an early improvement i n symptoms amongst male attenders at general practice can be predicted on t he basis of a small number of variables. (C) 1999 International Association for the Study of Pain. Published by Elsevier Science B.V.