Multiple somatic symptoms in employees participating in a randomized controlled trial associated with sickness absence because of nonspecific low back pain

Abstract

Background contextThe prevalence of multiple somatic symptoms is high in primary and hospital outpatient populations. Multiple somatic symptoms may be present in patients sick-listed because of low back pain (LBP) and may be associated with increased risk of not returning to work (RTW). PurposeTo explore whether multiple somatic symptoms in a subset of patients with nonspecific LBP was associated with RTW, sickness absence (SA), or other social benefits. Study designThe study was a cohort study based on a randomized clinical trial with a prospective 2-year follow-up period. Patients were referred from general practices to the Spine Center, Regional Hospital Silkeborg, Denmark. Patient samplePatients were 285 sick-listed employees (4–12 weeks), with nonspecific LBP as their prime reason for SA. Exclusion criteria were unemployment, radiculopathy, LBP surgery within the past year, previous lumbar fusion, suspected cauda equina syndrome, progressive paresis or other serious back disease, pregnancy, known substance abuse, or primary psychiatric diagnosis. Outcome measuresSelf-reported health was assessed by the LBP rating scale and questions about pain and health in general. Disabilities were measured by the Roland Morris Questionnaire, the Short Form-36, and the Fear-Avoidance Beliefs Questionnaire. Work-related questions comprised expectations about RTW and risk of losing job because of SA. The Common Mental Disorder Questionnaire (subscale SCL-SOM) was used to assess multiple somatic symptoms (12 items). We categorized multiple somatic symptoms into four groups based on the SCL-SOM sum score: <6, 6 to 12, 13 to 18 and >18. Status of SA (>2 weeks) and RTW were gathered from a national database (DREAM). MethodsThe patients (N=285) were randomized into either multidisciplinary or brief intervention at the Spine Center (2004–2008). Both interventions comprised clinical examination and advice by a physiotherapist and a rheumatologist. Data were collected from questionnaires at baseline (inclusion) and 1 year after inclusion. Data on SA benefits were gathered from the DREAM database that contains data on all social transfer payments (such as sick leave benefits and other disability benefits) registered on a weekly basis. ResultsAll health factors, female gender, and poor work ability were significantly associated with a higher level of multiple somatic symptoms. The percentage of persons with SA increased significantly with the symptom score after 1 year, and the duration of SA remained significantly longer after 2 years of follow-up between the multiple somatic symptoms groups. The percentages with RTW after 1 and 2 years were negatively associated with a higher level of multiple somatic symptoms at baseline. We found no difference between the intervention groups. ConclusionsA higher level of multiple somatic symptoms was significantly associated with poor health and work ability at baseline and with longer duration of SA and unsuccessful RTW through a 2-year follow-up period.