Contributions of prognostic factors for poor outcome in primary care low back pain patients

Background: Back pain is common and some sufferers consult GPs, yet many sufferers develop persistent problems. Combining information on risk of persistence and prognostic indicator prevalence provides more information on potential intervention targets than risk estimates alone.

Aims: To determine the proportion of primary care back pain patients with persistent problems whose outcome is related to measurable prognostic factors.

Methods: Prospective cohort study of back pain patients (30-59 years) at five general practices in Staffordshire, UK (n=389). Baseline factors (demographic; episode duration; symptom severity; pain widespreadness; anxiety; depression; catastrophising; fear-avoidance; self-rated health) were assessed for their association with disabling and limiting pain after 12-months. The proportion of those with persistent problems whose outcome was related to each factor was calculated.

Results: Prevalence of prognostic factors ranged from 23% to 87%. Strongest predictors were unemployment (adjusted relative risk (RR) 4.2; 95% CI 2.0, 8.5) and high pain intensity (4.1; 1.7, 9.9). The largest proportions of persistent problems were related to high pain intensity (68%; 95% CI 27, 87%) and unemployment (64%; 33, 82%). Combining these indicated that 85% of poor back pain outcome is related to these two factors. Poor self-rated health, functional disability, upper body pain and pain bothersomeness were related with outcome for over 40% of those with persistent problems.

Conclusions: Several factors increased risk of poor outcome in back pain patients, notably high pain and unemployment. These risks in combination with high prevalence of risk factors in this population distinguish factors that can help identify targets or sub-groups for intervention.