Cognitive-behavioural and social factors do not predict recurrent secondary health care use in patients with fibromyalgia: a longitudinal study

Abstract

Background Health care use in fibromyalgia (FM) is relatively high. Besides disease-related variables, cognitive-behavioural factors have been concurrently associated with health care use. It is unknown whether cognitive-behavioural and social factors also predict future health care use. Objectives To identify cognitive-behavioural and social factors predicting recurrent secondary health care use in FM. Methods Data were drawn from a prospective cohort of recently diagnosed patients with fibromyalgia (n=199), spanning 18 months. Patients were recruited after receiving their diagnosis and protocolled treatment advice by a rheumatologist. Using self-report questionnaires, health care use, cognitive-behavioural (i.e. illness cognitions, pain coping, coping flexibility), social (i.e. invalidation by family, spousal responses to pain and well behaviour), sociodemographic and disease-related variables including comorbidities, severity of FM, and depressive and anxiety symptoms were collected. Primary outcome was recurrent secondary health care use at 18 months follow-up defined as the use or non-use for each of the following four categories of secondary health care: consultation with medical specialists, diagnostic procedures, admission to health care institutions, and multimodal treatment program. A patient was considered a recurrent secondary health care user, if secondary health care from at least one of the four categories was used in the past six months. Univariate and multivariate logistic regression models examined whether and which variables were predictors for recurrent secondary health care use. Internal validation was performed to correct for over-fit of the final multivariate model. Results Recurrent secondary health care use was lower than initial secondary health care use. Univariate analysis showed that having at least one comorbidity, depressive feelings, severe consequences of fibromyalgia, low personal control and a high severity of fibromyalgia predicted recurrent secondary health care use. In the multivariate model, having at least one comorbidity was the only remaining predictor for recurrent secondary health care use. Conclusions Our results suggest that the existence of comorbidities as communicated by the patient is the strongest warning signal for recurrent secondary health care use in FM. There seems no value in using cognitive-behavioural and social factors for early identification of patients with FM at risk for recurrent secondary health care use.