Work outcome in yet undiagnosed patients with non-radiographic axial spondyloarthritis and ankylosing spondylitis; results of a cross-sectional study among patients with chronic low back pain

BACKGROUND: To understand the impact of yet undiagnosed non-radiographic axial spondyloarthritis (nr-axSpA) and ankylosing spondylitis (AS) on work outcomes in a cohort of patients with long-lasting chronic low back pain (CLBP). METHODS: Data were used from a primary care CLBP cohort that was established to understand the prevalence of nr-axSpA and AS. Clinical characteristics comprised measures of back pain (visual analogue scale), inflammation (C-reactive protein) and physical functioning (Roland Morris Disability Questionnaire (RMDQ)). Worker outcomes comprised a question on employment and the Work Productivity and Activity Impairment (WPAI) questionnaire, distinguishing absenteeism, presenteeism, and overall work impairment in those employed and activity impairment in all patients. For each disease subgroup, employment ratio compared to the general population was assessed by indirect standardization. Factors associated with work productivity were explored by zero inflated negative binomial (ZINB) regression models. RESULTS: Patients with CLBP (n = 579) were included (41% male, mean age 36 years), of whom 71 (12%) were identified as having nr-axSpA and 24 (4%) as having AS. The standardized employment ratios were 0.89 (95% CI 0.84–0.94), 0.97 (95% CI 0.85–1.09) and 0.81 (95% CI 0.56–1.06) for patients with CLBP, nr-axSpA and AS, respectively. Scores for the WPAI subdomains were not significantly different between patients with CLBP, nr-axSpA or AS. The ZINB models showed significant associations between visual analog scale (VAS) score for pain and RMDQ and work productivity. CONCLUSION: The impact of yet undiagnosed nr-axSpA and AS on patients’ work outcomes was substantial but was not significantly different from those of patients with long-standing CLBP. Variables significantly associated with reduced work productivity were VAS for pain and RMDQ score. ELECTRONIC SUPPLEMENTARY MATERIAL: The online version of this article (doi:10.1186/s13075-017-1333-x) contains supplementary material, which is available to authorized users.