Background: Heavy menstrual bleeding (HMB) and post-menopausal bleeding (PMB) together constitute the commonest gynaecological presentation in secondary care and impose substantial demands on health service resources. Accurate diagnosis is of key importance to realising effective treatment, reducing morbidity and, in the case of PMB, reducing mortality. There are many tests available, including transvaginal scan (TVS), endometrial biopsy (EBx), saline infusion sonography and outpatient hysteroscopy (OPH); however, optimal diagnostic work-up is unclear. Objectives: To determine the most cost-effective diagnostic testing strategy for the diagnosis and treatment of (i) HMB and (ii) PMB. Data sources: Parameter inputs were derived from systematic quantitative reviews, individual patient data (IPD) from existing data sets and focused searches for specific data. In the absence of data estimates, the consensus view of an expert clinical panel was obtained. Methods: Two clinically informed decision-analytic models were constructed to reflect current service provision for the diagnostic work-up of women presenting with HMB and PMB. The model-based economic evaluation took the form of a cost-effectiveness analysis from the perspective of the NHS in a contemporary, 'one-stop' secondary care clinical setting, where all indicated testing modalities would be available during a single visit. Results: Two potentially cost-effective testing strategies for the initial investigation of women with HMB were identified: OPH alone or in combination with EBx. Although a combination testing strategy of OPH + EBx was marginally more effective, the incremental cost-effectiveness ratio (ICER) was approximately £21,000 to gain one more satisfied patient, whereas for OPH it was just £360 when compared with treatment with the levonorgestrel intrauterine system (LNG-IUS) without investigation. Initial testing with OPH was the most cost-effective testing approach for women wishing to preserve fertility and for women with symptoms refractory to empirical treatment with a LNG-IUS. For the investigation of PMB, selective use of TVS based on historical risk prediction for the diagnostic work-up of women presenting with PMB generated an ICER compared with our reference strategy of 'no initial work-up' of £129,000 per extra woman surviving 5 years. The ICERs for the two other non-dominated testing strategies, combining history and TVS or combining OPH and TVS, were over £2M each. Limitations: In the absence of IPD, estimates of accuracy for test combinations presented some uncertainty where test results were modelled as being discordant. Conclusions: For initial investigation of women presenting to secondary care with HMB who do not require preservation of their fertility, our research suggests a choice between OPH alone or a combination of OPH and EBx. From our investigation, OPH appears to be the optimal first-line diagnostic test used for the investigation of women presenting to secondary care with HMB wishing to preserve their fertility or refractory to previous medical treatment with the LNG-IUS. We would suggest that the current recommendation of basing the initial investigation of women with PMB on the universal TVS measurement of endometrial thickness at a 5-mm threshold may need to be replaced by a strategy of restricting TVS to women with risk factors (e.g. increasing age-raised body mass index, diabetes or nulliparity), obtained from the preceding clinical assessment. Funding: The National Institute for Health Research Health Technology Assessment programme.