Short and long-term predictors of pain severity and interference in primary care patients with chronic musculoskeletal pain and depression

Abstract

Background: Chronic pain and depression are frequent comorbidities in primary care. Depression among other psychosocial factors play a role in the clinical course of chronic pain.

Objective: To study the short and long-term predictive factors of severity and interference of chronic pain in primary care patients with chronic musculoskeletal pain and major depression.

Methods: Longitudinal study of a cohort of 317 patients. The outcomes are severity and functional interference of pain (Brief Pain Inventory) measured at 3 and 12 months. We performed multivariate linear regression models to estimate the effects the explanatory baseline variables on the outcomes.

Results: 83% participants were women; average age was 60.3 years (SD = 10.2). In multivariate models, baseline pain severity predicted pain severity at 3 months (β = 0.53; 95% CI = 0.37-0.68) and at 12 months (β = 0.48; 95% CI = 0.29-0.67). Also, pain > 2 years of evolution predicted long term pain severity (β = 0.91; CI95%=0.11-1.71). Baseline pain interference predicted interference at 3 and 12 months (β = 0.27; 95%CI = 0.11-0.43 and β = 0.21; 95%CI = 0.03-0.40, respectively). Baseline pain severity predicted interference at 3 and 12 months (β = 0.26; 95%CI = 0.10-0.42 and β = 0.20; 95%CI = 0.02-0.39, respectively). Pain > 2 years predicted greater severity and greater interference at 12 months (β = 0.91; CI95%=0.11-1.71, and β = 1.23; CI95%=0.41-2.04). Depression severity predicted more interference at 12 months (β = 0.58; CI95%=0.04-1.11). Occupational status as active worker predicted less interference throughout the follow-up (β=-0.74; CI95%=-1.36 to -0.13 and β=-0.96; CI95%=-1.71 to -0.21, at 3 and 12 months). Currently working also predicts less pain severity at 12 months (β=-0.77; CI95%=1.52 - 0.02). With regard to the psychological variables, pain catastrophizing predicted pain severity and interference at three months (β = 0.03; 95% CI = 0.00-0.05 and β = 0.03; 95% CI = 0.00-0.05), but not at long term.

Conclusion: In a sample of adults with chronic pain and depression, this primary care study has identified prognostic factors that independently predict the severity and functional interference of pain. If confirmed in new studies, these factors should be targeted for individualized interventions.

Trial registration: ClinicalTrials.gov (NCT02605278), registered 16/11/2015.

Keywords: Catastrophizing; Chronic pain; Longitudinal studies; Major depression; Primary Health Care.