Background: Chronic disease is associated with aging and the average age of many Western populations is steadily increasing. The International Classification of Functioning, Disability, and Health (ICF) defines disability as behaviour, allowing for the application of behavioural models to the study of disability as well as interventions to affect behaviour change. Results of behavioural inerventions in chronic disease often produce heterogenous results and therefore, N-of-1 designs may be a useful way of testing such interventions in patients with chronic illness. Existing behavioural models must be tested on an individual level to allow for their application to intervention development on an individual level. The present thesis aimed to address these issues. Methods: Two studies were conducted. First, a systematic review and meta-analysis of self-management interventions for rheumatoid and osteoarthritis was carried out. Based on the findings that results from the reviewed group studies were heterogenous and that goal-setting and action planning were the most commonly applied self-management intervention techniques, the first study was a randomised, controlled series of N-of-1s testing the effect of a simple goal-setting and action planning intervention on physical activity (PA) within and between healthy individuals and those with chronic pain. The second study was a series of four N-of-1 trials testing a combined ICF/cognitions and a combined ICF/emotions model of disability in participants with arthritis. Results: Interventions employing self-management techniques positively affect outcomes in rheumatoid and osteoarthritis, but guidelines regarding core outcomes and measures are required along with clearer reporting of intervention content, potentially with the aid of taxonomies. A randomised, controlled series of N-of-1s was found to be a feasible method of testing a simple behavioural intervention. Goal-setting and action planning did not affect PA in healthy or chronic pain participants overall, but results varied between individuals. Longitudinal N-of-1s identified cognitions as the most effective predictors of PA behaviour in people with arthritis. Both combined models were feasibly and affectively applied at the individual level and were better at predicting behaviour than ICF constructs alone. Discussion: The findings of the present thesis have important implications for reporting methods and intervention design in health psychology and personalised behavioural medicine.
|Date of Award||23 Apr 2019|
- University Of Strathclyde
|Supervisor||Diane Dixon (Supervisor) & Susan Rasmussen (Supervisor)|