Background Lower socioeconomic status (SES) is associated with worse outcomes in several rheumatic diseases. In Ankylosing Spondylitis (AS), SES has been associated with poorer functional outcomes and certain physical activities with more radiographic disease severity in long-standing AS. The relationship of SES in radiographic progression has never been assessed.
Objectives The purpose of this study was to determine whether SES predicted radiographic progression in AS.
Methods This is a prospective longitudinal cohort study from 3 countries and 6 academic medical centers in patients with AS, meeting the modified New York criteria. There were 368 patients included with 2 sets of radiographs at least 1.5 years apart. Patients with advanced radiographic disease were excluded. The modified Stoke Ankylosing spondylitis Spine Score(mSASSS) was used to measure radiographic disease. Progression was defined as more than 2 mSASSS units in 2 years. SES was defined as education (in number of years). Other SES covariates, including occupational status and occupational physical activity demand were also assessed. In a subgroup of 240 patients, exposure to physical therapy and exercise history were also assessed. The univariate analysis was performed using Wilcoxon ranksum and Chi square tests. Multivariate analysis using logistic regression adjusted for disease duration.
Results Baseline inflammation and radiographic damage predicted radiographic progression. Use of TNF inhibitors decreased the risk of progression, as we have recently described. In this study, male gender also predicted radiographic progression (OR 2.37, 95% CI 1.02 – 5.48 p = 0.04). Radiographic progression was inversely related to education, with the lowest risk of progression in those with university degrees or higher (OR 0.24, 95% CI 0.11 - 0.53, p < 0.0001). Occupational activity (derived from the United States Bureau of Census Occupation Classification) and employment status were not significant in the multivariate model. Although smoking and age of onset were significant in the univariate analysis, they did not predict progression in the multivariate model. In the subgroup of patients with exercise data, there was no association with exposure to physical therapy or history of exercise patterns. There was no interaction between SES and country of origin.
Multivariate Logistic regression adjusted for disease duration
Conclusions SES, defined as level of education, independently predicts radiographic progression in AS. This is not accounted for by occupational activities or employment status at baseline. Male gender predicts radiographic progression supporting previous radiographic severity data.
References Ward MM et al., Arthritis Rheum. 2009 Jul 15;61(7):859-66
Disclosure of Interest L. Gensler Consultant for: Abbott, UCB, N. Haroon Consultant for: Abbott, Janssen, Amgen/Pfizer/Wyeth, J. Reveille Consultant for: Abbott, UCB, T. Learch: None Declared, M. Brown: None Declared, M. Weisman: None Declared, R. Inman Consultant for: Abbott, Janssen, Amgen/Pfizer/Wyeth, M. Ward: None Declared