Objectives To investigate the relationship between musculoskeletal pain, sociodemographic, individual and work place factors, and self-perceived work ability among hospital based nurses in Melbourne, Australia, as part of the international CUPID study.

Methods Nurses working across three major hospitals who were first surveyed 12 months ago were included. Information on demographic, lifestyle, mental and physical health and wellbeing (SF-12), workplace physical and psychosocial factors, musculoskeletal pain at several body sites for 7 days or more in the previous 12 months, and a modified Work Ability Index were collected. The association between risk factors and work ability was investigated using an ordinal logistic regression model.

Results Of 1111 baseline participants, 839 nurses (response rate 75.5%) completed the questionnaire. 768 nurses who had continued working in the study hospitals in the previous 12 months were included in the analysis. The majority (66.4%) reported excellent work ability, while 25.1% reported good, 7.3% moderate and 1.2% poor work ability. The independent factors associated with decreasing work ability were pain at any site (OR 1.86; 95% CI 1.08 to 3.19), pain at three or more sites (OR 1.99; 1.01 to 3.93), high job strain (OR 2.22; 1.27 to 3.88), female gender (OR 2.45; 1.04 to 5.75), part-time/casual employment (OR 1.69; 1.08 to 2.64), better mental (OR 0.86; 0.84 to 0.88) and physical (OR 0.86; 0.83 to 0.88) health and wellbeing.

Conclusions While the majority of nurses reported excellent or good work ability, efforts to retain the current workforce and improve their work ability should include the management of musculoskeletal pain, reducing job strain and improvement in general mental and physical health and wellbeing.