THU0703 Identifying clinical, psychological and work related factors associated with presenteeism: the international eular-pro at-work productivity study

Background Worker productivity loss, including presenteeism, is an important outcome for patients with inflammatory (IA) diseases and osteoarthritis (OA) and is frequently seen as a health outcome in clinical studies. It is important to understand which factors are related with this patient reported outcome in order to inform future work related interventions. Objectives To assess the association between disease related, psychological and work related factors with presenteeism. Methods In a large international study evaluating measures of presenteeism we recruited patients with IA and OA from UK, Fr, NL, Es, Se, Ro, It, Pt, and Ca. Absence rates and presenteeism levels (range 0–10=worst score) were measured using the Work Productivity and Activity Impairment (WPAI) questionnaire. Other job related questions were about demands and satisfaction, help from colleagues and opportunities to postpone or organise ones work. Disease related variables included HAQ, EQ-5D, VAS health status. The Hospital Anxiety and Depression Scale (HADS) was also completed. Cross-sectional univariable and multivariable Zinb regression models were applied to assess the association between these disease related, psychological, and work related factors and presenteeism, adjusting for age, gender and country. Due to high co-linearity only HAQ and HAD-anxiety were included in the multivariable model. Results A total of 544 patients (AS=138, OA=43, PsA=97, RA=266) were recruited with a mean (SD) age of 4710 yrs and a median symptom duration [IQR] of 105–18 yrs; 62% were women. 17% had a manual occupation. 111/544 (20.4%) reported being absent during the previous 7 days. Mean (SD) presenteeism score was 2.9 (2.7). In univariable analyses, worse self-reported disease activity and high levels of depression and anxiety were significantly associated with presenteeism in both the count and inflate part of the model (table 1). Less consistent results were observed for work related factors. In the multivariable model (s, 95% CI), worse functional disability was associated with presenteeism in both the count and excess zero part of the model (0.63; 0.49, 0.78: −2.70,–3.73, −1.66, resp). In addition, higher HAD-anxiety score (−0.09; −0.18; −0.00) was associated with a decreased likelihood of excessive zeros whilst not receiving help from colleagues (1.47; 0.64, 2.30) was associated with an increased likelihood of excessive zeros. There was a trend towards an association between very demanding jobs (−0.79; −1.63, 0.05) and presenteeism. VAS=visual analogue scale (0–100 worst score); EQ-5D=EuroQol-5D (−0.11–1=perfect health); HAD=Hospital Anxiety Scale (0–21=abnormal); PASS=do you feel your current condition is satisfactory, when you take your general functioning and your current pain into account (satisfactory/unsatisfactory). Count part=score in those with a score >0; inflate part=estimate of excess zero scores. Conclusions In this large international study of patients with IA and OA having a median symptom duration of 10 years, worse functional ability was the main factor associated with presenteeism. Disclosure of Interest None declared