Illness perceptions are the main predictors of depression and anxiety symptoms in patients with chronic pain

Abstract

Depression and anxiety symptoms in chronic pain are associated with adverse clinical outcomes, and appear highly related to patient's illness perceptions as well as with marital adjustment. This study aimed to investigate the predictive value of pain variables, marital adjustment and illness perceptions on depression and anxiety in patients with chronic pain. Two hundred patients were recruited from a pain unit in a public hospital in the north of Portugal. Patients completed a questionnaire that assessed illness perceptions (IPQ-Brief), marital adjustment (revised dyadic adjustment scale), depression and anxiety symptoms (hospital anxiety depression scale) and pain variables (pain intensity and pain disability index). Depression and anxiety symptoms were associated with pain intensity, pain-related disability, marital adjustment and illness perceptions. Results from hierarchical regression showed that illness perceptions contributed significantly to depression and anxiety symptoms over and above the effects of pain intensity, pain-related disability and marital adjustment, after controlling for gender. In multivariate analyses, pain intensity, pain-related disability and marital adjustment were uniquely related to depression and anxiety symptoms, whereas specific illness perceptions were uniquely related to depression symptoms (identity, treatment control, emotional response and coherence) and to anxiety symptoms (identity, emotional response and concern). Perceptions of greater symptomatology (identity) and of emotional impact, and lesser perceptions of treatment control and understanding of chronic pain (illness comprehensibility) were significantly associated with increased depression symptoms. Perceptions of greater symptomatology (identity), emotional impact and greater concern were associated with anxiety symptoms. These findings indicate that the contribution of illness perceptions was greater than that made by traditional covariates, and may therefore be a useful basis for future psychological interventi