Benzodiazepines have been widely prescribed since the 1960s for the management of adverse symptoms related to anxiety, depression, and sleep problems. They were regarded as an efficacious medication when compared with their predecessor, barbiturates. Within 10 years of their introduction, concerns began to be raised regarding their potential to produce dependence and withdrawal symptoms when ceased, including symptoms not present prior to their being prescribed. Subsequent research focussed on establishing effective strategies to ameliorate the adverse symptoms experienced even when the daily intake was slowly reduced. The aim of the work undertaken for this doctorate was to establish whether there was a role for cognitive behaviour therapy (CBT) in benzodiazepine cessation. The initial step in conducting the research for this doctorate was to obtain a detailed understanding of the current state of research on benzodiazepine cessation. Study 1 therefore focussed on establishing the effectiveness of treatment approaches used to assist individuals to cease benzodiazepine use. A Meta-analysis of treatment strategies undertaken in general practice and outpatient settings established that brief intervention resulted in superior cessation rates at post-treatment than routine care. Gradual dose reduction plus CBT was slightly superior to gradual dose reduction alone. However, substitutive pharmacotherapies in combination with gradual dose reduction did not result in a superior outcome to gradual dose reduction alone, and substitutive pharmacotherapy plus abrupt benzodiazepine cessation was less effective than gradual dose reduction. While, providing CBT in conjunction with gradual dose reduction offered a superior outcome than gradual dose reduction alone, current evidence does not identify the CBT strategies that contributed to the superior outcome. The next step in the development of the CBT intervention involved obtaining a deep appreciation of the issues relating to cessation from the perspective of General Practitioners (GPs) and Benzodiazepine Users (BzUs). Accordingly, Study 2 administered semi-structured interviews about benzodiazepine use and its cessation to 28 GPs and 23 BzUs. Responses were analysed using the Consensual Qualitative Research approach, as it enabled comparisons to be made between the views of the two groups of interviewees. The study identified commonality between GPs and BzUs on reasons for commencing use, the role of dependence in continued use, and the importance of lifestyle change in its cessation. BzUs felt there was greater need for GPs to routinely advise patients about non-pharmacological management of their problems and potential adverse consequences of long-term use before prescribing benzodiazepines. Few GPs had assisted a patient to cease use reportedly due to the required time and the expectation of a poor outcome. There was a perception that patients wanted a pharmacological solution to their problems. A critical gap in assessment instruments that are needed for a comprehensive assessment of the outcomes from a treatment trial was identified. In particular, there was no measure of benzodiazepine expectancy or self-efficacy concerning maintenance of benzodiazepine dose reduction. Therefore, Study 3 adapted existing expectancy and self-efficacy measures form other substance domains to verify their applicability to benzodiazepines. Current BzUs (n = 155) were invited to complete two questionnaires either online or via hard copy. Principal component analysis (PCA) of a newly developed Benzodiazepine Expectancy Questionnaire (BEQ) resulted an 18-item, 2-factor scale, while a Benzodiazepine Refusal Self Efficacy Questionnaire (BRSEQ) formed a 16-item, 4-factor scale, Confirmatory factor analysis (CFA) in a second sample (n = 139) confirmed these internal structures, reducing the BEQ to 12 items and the BRSEQ to 14 items respectively. The qualitative study suggested that many GPs would be reluctant to engage in psychological support for benzodiazepine cessation and it was evident that specialist services would be unable to provide substantial support especially in rural and remote areas. Accordingly, it was decided to develop a treatment that was remotely delivered. The initial pilot used a correspondence-based approach, delivered via the postal service. Study 4 comprised a small pilot comparing GP managed gradual dose reduction, plus CBT via mail (M-CBT), which was either delivered immediately (IM-CBT) or after 3 months (DM-CBT). Despite substantial efforts over a 2 year period to recruit GPs and BzUs, only 6 received the allocated intervention. It was decided to trial the intervention as an internet-delivered program to enhance its accessibility to BzUs. Access to the program was promoted through the project website and links from high profile support websites. Study 5 was an uncontrolled trial of internet-based CBT (I-CBT). Access was provided to all newsletters, although, participants were given a suggested sequence for access. Despite placement on the internet and cross-listing on several key websites, the study still only recruited 35 participants (3 of which received the program by mail). Of the 32 undertaking the program via the internet, 21 completed the 3-month assessments and 14 the 6-month assessments. Eight participants reduced their weekly benzodiazepine intake by at least 50%, by 3 months, with five ceasing use at 6 months. A significant increase in self-efficacy, and a decrease in depressive symptoms and dependence were seen. Providing CBT either via mail or the internet assisted some participants to reduce or cease long-term benzodiazepine use. Recruitment to both M-CBT and I-CBT was limited, despite substantial attempts to market the intervention. The studies undertaken for this doctorate make a unique contribution to improving treatment outcomes for people wishing to cease long-term benzodiazepine use. They also provide direction for more extensive studies to definitively establish the nature of effective treatment. The current evidence clearly supports the importance of gradual dose reduction and the role of CBT in further improving treatment outcomes. However, engagement of both BzUs and GPs remains challenging. Remote delivery of CBT via mail or the internet may assist with improving access to CBT, but it does not solve the problem of GP and BzU engagement. An effective system-wide program to address long-term benzodiazepine use will require that incentives for GP involvement (a disincentive for long-term prescription) are in place.