Effectiveness and cost effectiveness of hysterectomy, microwave endometrial ablation and thermal balloon ablation for the treatment of menorrhagia

PRIMARY RESEARCH DESIGN: The clinical effectiveness element of this project will be conducted using analysis of existing data from two sources: (1) Anonymised pooled data record linkage of relevant national datasets and randomised cohorts in Scotland, (2) Individual patient data (IPD) meta-analysis of existing randomised controlled trials. The output from these two analyses will be used to inform the development of cost effectiveness models incorporating UK based cost data. SETTING: Analysis of existing data from the Scottish national clinical database and existing randomised controlled trials both from UK and elsewhere. TARGET POPULATION: Women with heavy menstrual bleeding who have not responded to or do not wish medical treatment. HEALTH TECHNOLOGIES: Second-generation ablative techniques (microwave endometrial ablation and thermal balloon ablation) and hysterectomy. MEASUREMENT OF COSTS AND OUTCOMES: (1) Using pooled anonymised data from the Information and Statistics Division -( ISD) Scotland, the analyses will determine the characteristics of women which predict the likelihood of long term treatment failure with endometrial ablation. (2) Using individual patient data meta-analysis of relevant RCTs the likelihood of treatment success will be determined. This second analysis has the advantage of detailed information relating to predictive characteristics of the women and the opportunity to perform subgroup analyses. Thus the PRIMARY CLINICAL OUTCOMES of interest are: 1) treatment failure as demonstrated by the need to undergo a subsequent ablation or a hysterectomy, 2) complications - early and late. THE COSTS AND ECONOMIC OUTCOMES : The economic evaluation will take the form of a cost effectiveness analysis within a decision analytic framework based on an outcome of cost per woman successfully treated and hysterectomy avoided. If quality of life data are available then the economic evaluation will present results in terms of cost per QALY. The model will be populated using data derived from the analysis of national registers and existing randomised controlled trials (as above). Resource use will be estimated from the existing published evidence and additional costs data will be sought from other sources such as the national schedule for reference costs. ALGORITHM GENERATION: These results will be used to develop a parsimonious algorithm to guide clinical decision making regarding the appropriate choice of first line surgical treatment for women with menorhagia. SAMPLE SIZE: From the Scottish Information and Statistics Division (1990-2004) alone, we envisage assembling a cohort of at least 13,000 women post-ablation and 40,000 post-hysterectomy. With a dataset of 13,000 ablations, the two-sided 95% confidence interval around an estimated prevalence of re-treatment of 25% would be (24.3%, 25.7%). In addition, we anticipate that we can access IPD on 90% of 1,600 women randomised to alternative ablation techniques and 400 randomised to hysterectomy versus ablation. PROJECT TIMETABLE: An 18 month project is envisaged: (1) Aggregation and cleaning of national cohort dataset and IPD data aggregation; identification of cost data (months 1-6); (2) Analysis of cohort data and IPD meta-analysis (months 7-12); (3) Cost-effectiveness modelling (months 12-16); (4) Final algorithm development and report writing (months 17-18).