Association between anxiety, health-related quality of life and functional impairment in primary care patients with chronic pain

Abstract Objective Anxiety and chronic pain are prevalent and frequently co-occur. Our purpose was to examine the association between anxiety, health-related quality of life (HRQL) and functional impairment in primary care patients with chronic musculoskeletal pain. Methods Data were drawn from baseline interviews of the 250 primary care patients enrolled in the Stepped Care to Optimize Pain care Effectiveness trial. Validated measures were used to determine the proportion of patients screening positive for five common anxiety disorders: generalized anxiety, panic, social anxiety, posttraumatic stress and obsessive–compulsive disorder. Bivariate analyses examined associations between the type and number of anxiety disorders for which patients screened positive and representative pain, psychological and other HRQL outcomes. Multivariable models controlling for major depression and other covariates examined the association between the number of screen-positive anxiety conditions and functional impairment in psychological [SF-12 mental component summary (MCS) score], pain [Brief Pain Inventory (BPI) interference score] and work (disability days) domains. Results One hundred fourteen (45%) patients screened positive for at least one anxiety disorder and, compared to the 136 screen-negative patients, had significantly worse scores across multiple pain, psychological and other HRQL domains. Substantial impairment was seen for each of the five screen-positive anxiety conditions and progressively worsened as the number of conditions increased from one ( n = 54) to two ( n = 26) to ≥ 3 ( n = 34). The number of screen-positive anxiety conditions was strongly associated ( P < .0001) with worse BPI interference and MCS scores and more disability days in models adjusting for age, sex and medical comorbidity. After further adjusting for major depression, associations were attenuated but remained significant for BPI interference ( P < .0001) and MCS ( P = .018) and marginally significant for disability days ( P = .062). Conclusion Nearly half of primary care patients with chronic pain screen positive for one or more anxiety disorders, which in turn are adversely associated with impairment across multiple domains of HRQL. Detecting and treating anxiety may be an important component of pain management.