Testing an integrated behavioural and biomedical model of disability in N-of-1 studies with chronic pain.

Background. Previous research has supported an integrated biomedical and behavioural model explaining activity limitations. However, further tests of this model are required at the within-person level, because while it proposes that the constructs are related within individuals, it has primarily been tested between individuals in large group studies. We aimed to test the integrated model at the within-person level. Method. Six correlational N-of-1 studies in participants with arthritis, chronic pain and walking limitations. Daily measures of theoretical constructs were collected using a handheld computer (PDA), activity was assessed by self-report and accelerometer, and data analysed using time-series analysis. Results. The biomedical model was not supported as pain-impairment did not predict activity, so the integrated model was supported partially. Impairment predicted intention to move around while perceived behavioural control (PBC) and intention predicted activity. PBC did not predict activity limitation in the expected direction. Conclusions. The integrated model of disability was partially supported within individuals, especially the behavioural elements. However, results suggest that different elements of the model may drive activity (limitations) for different individuals. The integrated model provides a useful framework for understanding disability and suggests interventions, and the utility of N-of-1 methodology for testing theory is illustrated