Session Type: Abstract Submissions (ACR)
Background/Purpose: Almost all studies on the impact of Ankylosing Spondylitis (AS) on the employment status or work disability (WD), the far majority of these studies are cross-sectional analyses and none compared work outcome over time with the general population. We aimed at quantifying the long-term impact of AS on participation in labour force over 12 years in a prospective cohort study, at identifying predictors of an adverse work outcome (AWO) and at comparing the incidence of WD over time with the general population.
Methods: Work related information and clinical characteristics were obtained from patients followed in the Outcome in AS International Study (OASIS) during 12 years. Patients who were at risk of having an AWO at any time were included in the analysis (i.e. those working, either at baseline or starting/resuming work later). AWO was defined as WD or a reduction in the number of working hours. Survival analysis was used to investigate AWO. With Cox regression, first baseline and secondly time-varying predictors (with 2-year time-lag) were identified; the latter adjusted for 2 time dependent variables (change in social security system and market availability of anti-TNF). Incidence rate for WD over 12 years among Dutch patients was compared with the general population using indirect standardization.
Results: Of the 215 (of 216) patients in OASIS cohort with baseline data on work status, 139 (65%) were at risk for AWO (mean age 38(SD 10) years, 75 % male, 81% HLA-B27 positive, mean disease duration 9 (SD 7) years), of whom 11% (n=15) were not working at baseline but resumed work over time. Thirty one of 139 patients (22%) experienced AWO; 12 (39%) became full-work disabled (full-WD), 5 (16%) partial work disabled (partial-WD) and 14 (45%) reduced working hours. For a total 1404, person years, Kaplan Meier analysis revealed a 2.2% of annual rate of AWO. Multivariable Cox analysis showed that higher baseline BASFI (HR 1.3; 95% CI 1.01-1.4) and Dutch origin (HR 3.6; 95% CI 1.3-10.0) predicted long-term AOW. In the time-lagged prediction analysis, higher BASFI (HR 1.3; 95%CI 1.1-1.5) and Dutch origin (HR 3.0; 95% CI 1.2-8.2) significantly predicted AWO over time. Of 130 Dutch patients in working age, 9% (n=12) became WD over 12 years. Among Dutch patients the incidence rate of WD over 12 years was 2.9 (95% CI 1.2; 4.6) and 1.2 (95% CI -0.4; 2.9) times higher compared with the general population, for males and females, respectively
Conclusion: Even in AS patients with already long disease duration, the disease continues to have a negative impact on work outcome.. Self-reported physical function (BASFI) and system characteristics, but not job-type, disease activity or structural damage consistently influence AWO.
J. D. Castillo-Ortiz,
Pfizer, Janssen, Merck, Abbott,
Pfizer, Janssen, Merck, Abbott, Amgen, Astra, BMS, Centocor, GSK, UCS, Vertex,
D. van der Heijde,
F. van Den Bosch,
Abbott, MSD, Pfizer, UCB,
Abbott, Bristol-Myers Squibb, MSD, UCB,
« Back to 2013 ACR/ARHP Annual Meeting
ACR Meeting Abstracts - https://acrabstracts.org/abstract/work-outcome-in-patients-with-ankylosing-spondylitis-12-year-results-from-oasis/