This thesis aimed to develop an evidence based, theory driven behaviour change intervention to increase levels of physical activity (PA) in patients with chronic kidney disease (CKD). Using mixed-methods the following studies were conducted: Cross-sectional study of self-reported PA levels and PA correlates: Survey of 1015 patients indicated a high prevalence of PA insufficiency (85.3%), but a readiness to change. Regression modelling indicated self-efficacy, physical function, older age and sex as independent predictors of PA. Observational study of walking and survival: Walking behaviours were shown through Cox proportional hazard modelling to be independently associated with mortality in a 44-month median follow-up of 437 persons and 89 deaths. Qualitative study exploring patient factors influencing exercise: Factors influencing exercise were captured via focus groups and semi-structured interviews with 36 patients. Analysed thematically findings were conceptualised within a social cognitive perspective and included: personal influences (co-morbidities, symptom burden, ageing, fear, previous experiences, and internal drive); behavioural influences (health and wellbeing, maintaining normality, and enjoyment); and environmental influences (organisation, physical, and social). Findings highlighted PA and disease information needs, and modifiable psychological targets for intervention. Person-Based Approach to the development of a PA intervention: Expert consultations (n=9), examination of theory and evidence, observations of patient education programmes; Public Patient Involvement (n=9) to establish intervention guiding principles; user testing with patients (n=14) to refine the Physical Activity Changing Together (PACT) intervention, a structured group-based PA education programme. PACT feasibility study: Recruitment, retention and engagement/acceptability were assessed using mixed methods in a 12-week study. Post intervention step counts indicated a mean increase of 2127 steps/day from baseline. PACT was feasible to implement, acceptable to patients and warrants further testing. Original contributions to knowledge include: confirming a link between walking and health; identifying factors that influence PA; and the development/evaluation of a CKD specific PA intervention; all which can be framed within the Behavioural Epidemiology Framework.