The overall aim of the thesis is to explore the determinants of General Practitioner (GP) referrals to specialist mental health services, in particular psychological therapy. Paper 1 is intended to contribute to this literature by providing a systematic review of GP and organisational factors identified as impacting on referral in previous research. According to this literature, referral to a mental health specialist was more likely if the GP does not feel that they have the 'capacity to help' and perceives 'time constraints' on how long they can spend with the patient. Referral also appeared more likely if the GP felt they had acceptable 'access to services' and if they had a close 'consultation/liaison' relationship with specialists. The theoretical and clinical implications of these findings are discussed, and gaps in the current literature identified for further research.Paper 2 describes an empirical study aimed at exploring the determinants of GP referral for psychological interventions within Primary Care Mental Health Services (PCMHS). 132 GPs completed questionnaires, including demographic information, attitudes towards mental health and its treatment and responses to short fictional case vignettes indicating their likelihood of referral to the PCMHS. Qualitative results suggested that GPs consider a range of factors in their referral decisions, including patient preference, severity of the problem, access to services and the effectiveness of the service. Alternative options considered included signposting to other services, reviewing, medication and providing advice and support. Quantitative results suggested that younger GPs reported a higher likelihood of referral, and were more likely to refer in line with guidelines. Psychological factors were not associated with referral likelihood or referral in agreement with guidelines. In line with previous research on clinician behaviour, findings of papers 1 and 2 are considered primarily in the context of the Theory of Planned Behaviour, and the utility of this model in predicting referral behaviour is evaluated throughout. A greater understanding of predictors of referral is thought to be valuable in designing clinician and service level interventions to improve the proportion of those in need who are able to access psychological therapy.Paper 3 provides a critical evaluation of the research process as a whole, including the processes involved in the literature review and empirical study. The strengths and weaknesses of both of these elements are discussed, along with an evaluation of the overall approach taken throughout the thesis. The findings of both studies are integrated and discussed in the context of current policy and proposed changes to healthcare provision. Implications for theory, clinical practice and further research are discussed.