Breast cancer affects one in eight women and approximately 40% will require a mastectomy. The loss of a breast may dramatically impact upon quality of life. Breast reconstruction (BR) is offered to improve outcomes. Making decisions regarding BR surgery, however, is challenging and women and healthcare professionals (HCPs) need to weigh-up the likely benefits of surgery against the risks of adverse outcomes. Decisions are informed by published outcomes and surgeon and patient preferences. Well-designed studies, such as multi-centre randomised controlled trials (RCTs) provide the optimal evidence, but few RCTs have been undertaken in BR. The aim of this thesis was therefore to explore the need for and the feasibility of RCTs in BR. Three systematic literature reviews (SRs) critically appraised and summarised the quality of outcome reporting in BR and semi-structured qualitative interviews were conducted with women and HCPs to explore decision-making and attitudes to randomisation in BR. The clinical outcomes SR included 123 observational studies and 11 RCTs. The majority were at high-risk of bias and outcome reporting was heterogeneous. The cosmetic outcome SR included 122 studies. Cosmesis was assessed by patients and HCPs, but the methodology was inconsistent. The patient-reported outcome SR included 62 studies, less than 60% of which were considered methodologically robust. Sixty-two interviews with women and HCPs demonstrated decision-making in BR to be complex. A third of women reported decisional regret and insufficient time and information were identified as barriers to decision-making. Inequalities in access to care, however, emerged as the most significant determinant of women's reconstructive experience. Both women and HCPs accepted RCTs in BR in particular circumstances. There is an urgent need for well-designed studies with standardised outcome assessment in BR. Selected randomised trials may be feasible, but they are unlikely to address the key questions in BR. Service re-organisation and interventions to improve decision-making are needed to improve outcome for women considering BR in the UK.

This thesis aimed to determine the causes of the better survival of women treated for breast cancer by specialist surgeons. Two separate studies were performed. Firstly, a geographically defined database of pathology and mortality data from 1980-88 was updated to 1996. Secondly, a retrospective cohort study was performed of 2762 (94% of total) case-records of women diagnosed during 1986 to 1991. Women treated by specialists, even in the screening time-period, had a long-term survival advantage. Inadequate compared to adequate treatment was associated with double the risk of recurrence independent of other factors. The case-mix for non-specialists was more often unknown. Specialists were more exact in the treatment given: a preoperative diagnosis and mammography were obtained twice as frequently; staging was more selective; breast conserving surgery was performed with equal frequency, but when performed by non-specialists was inadequate twice as frequently and was four times more likely to be due to more than one error. Non-specialists omitted axillary staging three times more often and inadequately treated the axilla ten times more often. Overall, inadequate loco-regional treatment was twice as likely if treated by a non-specialist. Specialists prescribed chemotherapy twice as often, but the prescription of systemic therapy to those at highest risk was the same. The loco-regional recurrence for non-specialists was twice that of specialists after accounting for pathology and treatment. Specialists had a 26% lower risk of death after allowing for demographic, prognostic and treatment variables. This was negated if adequacy of treatment was included as an independent variable. There was poor correlation between caseload and adequacy of treatment. Adequacy of loco-regional treatment correlated with death from breast cancer.

The portfolio is compromised of three parts: a systematic literature review; an empirical paper and appendices. Part one, the systematic literature review, looks at the pre-surgical psychological predictors of outcomes following breast cancer surgery. The review begins with a synthesis of background information relevant to the area, which leads to specific research questions. The review process is then outlined, results presented and synthesised in the discussion. Limitations of the review and ideas for further research are discussed. Part two, the empirical paper, describes a study that researches pre-surgical predictors of satisfaction following breast reconstructive surgery. The current literature in the field is introduced and reviewed, and hypotheses established. The methodology is then described, results presented and findings discussed in relation to theoretical models. Finally, limitations of the study and ideas for further research are described. Part three, the appendices, present relevant further information for parts one and two.