Background The adverse effects of RA on absenteeism (i.e. days absent from work) and presenteeism (at-work productivity loss) are increasingly recognised as an important issue. Research suggests the reasons are multifactorial with many factors such as disease severity, quality of life, and depression possibly contributing to poor work outcomes in patients with RA.
Objectives To identify predictors of presenteeism and absenteeism over one year in patients with RA commencing treatment with methotrexate (MTX) or biologics for the first time.
Methods Patients recruited to the Rheumatoid Arthritis Methotrexate Starters study (RAMS) or the Biologics in Rheumatoid Arthritis Genetics and Genomics Study Syndicate (BRAGGSS), and in full or part-time paid employment were included in this analysis. Demographic (e.g., age), clinical (e.g., DAS28), and psychological data (e.g., Hospital Anxiety and Depression scale (HADS)) were collected at baseline. Patients were followed up at six months and one year in both cohorts. Absenteeism in the last month (days missed), and presenteeism (0 = no interference – 10 = complete interference), were measured at all three points using the RA specific Work Productivity Survey (WPS-RA). Patients with at least one follow-up post baseline were included in this analysis. Due to excessive zeros in the WPS-RA, a repeated measure zero inflated negative binomial regression (ZINB) was used to test the association between baseline demographic, clinical and psychological variables and presenteeism and absenteeism over one year.
Results Data was available for 191 patients in BRAGGSS and 308 in RAMS. In BRAGGSS, the mean age was 51 years (SD 9.5), median symptom duration was 8 years (IQR 5–13); 78% female. In RAMS, the mean age was 52 years (SD 9.6), median symptom duration was 8.5 months (IQR 5–23); 66% female. At baseline, 26% (BRAGGSS) and 27% (RAMS) reported ≥1 days absent from work, and the median presenteeism scores were 5 (IQR 3–7) and 4 (IQR 1–7) for BRAGGSS and RAMS respectively. Presenteeism scores significantly improved over 1 year in both cohorts (BRAGGSS: median 2 [IQR 0–5], RAMS: median 2 (IQR 0–4) Kruskal-Wallis p<0.001)), as did absenteeism for both cohorts although non-significant. The ZINB revealed similar predictive patterns in both cohorts for presenteeism and absenteeism (Table 1). For example, for every 0.06 increase in EQ-5D (clinically important difference), the odds of being in the zero presenteeism group (i.e., no at-work productivity loss) over one year increased by a factor of 1.2 in BRAGGSS, and was similar in RAMS. In RAMS, a one unit increase in VAS fatigue decreased the odds of having no absenteeism over one year by a factor of 0.97.
Conclusions The results from this analysis support the notion of a multifactorial relationship between RA and work outcomes. The results suggest that greater quality of life and psychological well-being, amongst other factors, can predict improvements in presenteeism and absenteeism at one year in patients on either MTX or biologics for RA.
Acknowledgements Funding: Pfzier I-CRP
Disclosure of Interest None declared