Background/Purpose: Disease activity has an important impact on work productivity in patients with Ankylosing Spondylitis (AS). However, if and to what extent this is the case in early axial Spondyloarthritis (axSpA)  and if it is similar in subgroups of patients e.g. male vs. female patients. The aim is to investigate if the impact of disease activity, assessed by ASDAS (AS Disease Activity Score, CRP based), on work productivity is similar according to gender, age, medication use, profession, HLA-B27 status and duration of back pain at baseline in early axSpA patients.

Methods: The SPACE-cohort includes patients (chronic back pain ≥3 months, ≤2 years, onset <45 years) from 5 European centers. Patients who fulfilled the ASAS criteria for axSpA were included in the analysis. Work Productivity and Activity Impairment questionnaire (WPAI) was completed by patients at baseline to assess Work Productivity Loss (WPL; i.e. total work impairment due to disease), presenteeism (i.e. decreased work functionality due to disease) and absenteeism (i.e. absence at work due to disease) in the past 7 days. Higher scores indicate greater impairment (range 0-100). ASDAS was used to assess disease activity. Gender, age, medication use, profession and duration of back pain were tested for effect modification (p<0.20) one by one in a linear regression model. Continuous variables were split by the median.

Results: 124 axSpA patients working at baseline were included; 73 fulfilled the clinical arm, 51 patients the imaging arm of the ASAS axSpA criteria. Patients were on average 31.1 years old (SD 7.7), 50.8% were male and had a mean duration of back pain of 13.7 months (SD 7.8). They worked on average 28.0 (SD 15.7) hours and missed 3.3 (SD 8.8) hours per week at work due to axSpA. Mean WPL, presenteeism, and absenteeism (SD) were 35.7% (29.4), 33.2% (27.0) and 9.3% (22.6) respectively. Patients had a mean ASDAS of 2.4 (SD 0.9). In the univariable model (Table 1),  1 point increase in ASDAS resulted in an increase of 18.5%, 16.9%, 9.6% in WPL, presenteeism and absenteeism, respectively. Fulfilment of the clinical or imaging arm was not an effect modifier (WPL p=0.69; presenteeism p=0.66; absenteeism p=0.58). Gender was an effect modifier in the associations between ASDAS and WPL (p=0.22, borderline significant) and presenteeism (p=0.15). The use of NSAIDs (p=0.16), age, and gender (two-way interaction p=0.06) were effect modifiers of absenteeism. The associations remained statistically significant in all stratified models, except in several models of absenteeism.

Conclusion: In early axSpA, higher disease activity is associated with increased work productivity loss including  presenteeism and absenteeism. The same level of disease activity appears to have more adverse impact on work productivity in women than in men. Disease activity was associated with higher absenteeism in patients using NSAIDs, younger men and older women.