- Research article
- Open Access
- Open Peer Review
Menopausal hormone therapy: a systematic review of cost-effectiveness evaluations
BMC Health Services Researchvolume 17, Article number: 326 (2017)
Several evaluations of the cost-effectiveness (CE) of menopausal hormone therapy (MHT) have been reported. The aim of this study was to systematically and critically review economic evaluations of MHT since 2002, after the Women’s Health Initiative (WHI) trial results on MHT were published.
The inclusion criteria for the review were: CE analyses of MHT versus no treatment, published from 2002-2016, in healthy women, which included both symptom relief outcomes and a range of longer term health outcomes (breast cancer, coronary heart disease, stroke, fractures and colorectal cancer). Included economic models had outcomes expressed in cost per quality-adjusted life year or cost per life year saved. MEDLINE, EMBASE, Evidence-Based Medicine Reviews databases and the Cost-Effectiveness Analysis Registry were searched. CE evaluations were assessed in regard to (i) reporting standards using the CHEERS checklist and Drummond checklist; (ii) data sources for the utility of MHT with respect to menopausal symptom relief; (iii) cost derivation; (iv) outcomes considered in the models; and (v) the comprehensiveness of the models with respect to factors related to MHT use that impact long term outcomes, using breast cancer as an example outcome.
Five studies satisfying the inclusion criteria were identified which modelled cohorts of women aged 50 and older who used combination or estrogen-only MHT for 5-15 years. For women 50-60 years of age, all evaluations found MHT to be cost-effective and below the willingness-to-pay threshold of the country for which the analysis was conducted. However, 3 analyses based the quality of life (QOL) benefit for symptom relief on one small primary study. Examination of costing methods identified a need for further clarity in the methodology used to aggregate costs from sources. Using breast cancer as an example outcome, risks as measured in the WHI were used in the majority of evaluations. Apart from the type and duration of MHT use, other effect modifiers for breast cancer outcomes (for example body mass index) were not considered.
This systematic review identified issues which could impact the outcome of MHT CE analyses and the generalisability of their results. The estimated CE of MHT is driven largely by estimates of QOL improvements associated with symptom relief but data sources on these utility weights are limited. Future analyses should carefully consider data sources and the evidence on the long term risks of MHT use in terms of chronic disease. This review highlights the considerable difficulties in conducting cost-effectiveness analyses in situations where short term benefits of an intervention must be evaluated in the context of long term health outcomes.
Menopausal hormone therapy (MHT) is considered an effective treatment for menopausal symptoms . However, to assess whether MHT is cost-effective in any group, benefits for quality of life need to be considered together with the health risks, and resources (costs) associated with use. Conceptually, the cost-effectiveness (CE) of any intervention can be summarised using a single measure known as the cost per Quality-Adjusted Life Years (QALY). This measure can be estimated using modelled evaluation of the benefits, costs, and any adverse health effects (short or long term) of the intervention. However, information on the CE of MHT is limited, in part because MHT was widely available for a number of years before standardised national health technology assessment processes involving CE evaluation were established. In the 1980s and 1990s, a few CE analyses of MHT were performed [2,3,4], but these included assumptions (e.g. that MHT prevented cardiovascular disease) that have not been supported by subsequent findings . Partly as a consequence, the findings of these early CE evaluations were favourable [2,3,4]. As new evidence to better quantify the effects of MHT has emerged, updated estimates of the overall benefit and cost trade-offs for MHT continue to be of potential value since MHT is still relatively widely used in many developed countries. For example, in Australia, 13% of women in their fifties and sixties reported being current users in a national survey conducted in 2013, and of these 73% had been using MHT for 5 years or longer . Ongoing evaluation of the CE of drugs that are widely used, ensures that optimal health investments continue to be made.
Evidence of the health risks associated with MHT use has been accumulating from epidemiological studies and trials in the late 1990s and early 2000s [7,8,9,10,11,12,13,14,15]. However, in 2002, the Women’s Health Initiative (WHI) estrogen plus progestin trial was stopped early after an increased risk of breast cancer, coronary heart disease (CHD), stroke, and pulmonary embolism was reported in study participants randomised to combination MHT compared to women randomised to placebo . Since then, evidence of these health risks has been provided by additional studies including a large UK study [17, 18]. Although the interpretation of these findings has been disputed , independent reviews [5, 20] of the worldwide evidence from other trials and observational studies by regulatory authorities continue to support cautious targeted use. For example, the UK Medicines and Healthcare products Regulatory Agency (MHRA) concluded that ‘for all women the lowest effective dose should be used for the shortest possible time’ and ‘the need to continue MHT should be reviewed at least yearly taking into consideration the change in balance of risks and benefits’ . A review of randomized controlled studies of MHT versus placebo by the US Preventive Services Task Force found the risks of taking MHT, when used to prevent chronic conditions, outweigh the benefits . This resulted in the US Preventive Services issuing recommendation statements against the routine use of MHT for the prevention of chronic conditions in postmenopausal women, above 50 years of age [20, 22]. In 2015 clinical guidelines on the menopause were commissioned by the National Institute for Health and Care Excellence (NICE) to provide advice for healthcare professionals and women on the menopause and symptom relief [National Institute for Health and Care Excellence. Menopause: diagnosis and management of menopause. (NICE guideline 23.) 2015. www.nice.org.uk/guidance/ng23]. NICE recommended the adoption of an individualised approach at all stages of diagnosis, investigation and management of menopause; taking this into account, one recommendation was to offer women MHT for vasomotor symptoms after discussing with them the short-term (up to 5 years) and longer-term benefits and risks. As part of the NICE guideline, a review of economic evaluations of short term treatments for menopausal symptoms and a de novo modelled evaluation were performed. The NICE guidelines did not provide quantitative summary estimates of risks and benefits [Hickey M, Banks E. NICE guidelines on the menopause. BMJ. 2016;352:i191], although these had been earlier provided by the MHRA for chronic disease outcomes .
The overall aim of the current study was to conduct a systematic review of evaluations on the CE of MHT and to harness standardised frameworks for reporting standards and key model parameters. We confined the review to 2002 and later since the WHI 2002 report marked the beginning of a substantial change in public and clinician understanding of the overall risks associated with MHT use, and led to large scale reductions in MHT use in many developed countries [23, 24]. Evaluations conducted in the 1980s and 1990s were not included in this review because some assumptions used in their models have not been supported by subsequent findings. Using the CHEERS checklist and the Drummond framework  each identified study was assessed in terms of data sources for MHT-related utility with respect to symptom relief; the methodology for assessing costs; outcomes considered in the economic models and the comprehensiveness of the models in respect to factors that affect particular chronic disease outcomes, using breast cancer risk as an example. With respect to breast cancer, we assessed whether evaluations considered the type of MHT, the duration of use, the impact of body mass index and the timing of initiation of MHT in relation to the menopause, since the evidence supports these factors as modifiers of the relative risk (RR) of breast cancer in relation to MHT use [11, 17, 26, 27].
Search strategy, eligibility criteria and article selection
A systematic search was conducted for relevant articles published from 2002, with the date of final search on 23rd February 2016. Databases searched were Ovid MEDLINE (US National Library of Medicine, Bethesda, MD, USA), EMBASE (Reed Elsevier PLC, Amsterdam, The Netherlands), Cost-effectiveness Analysis Registry (Tufts Medical Centre, Boston, MA) and Evidence-Based Medicine Reviews (American College of Physicians, Wiley-Blackwell, New Jersey, US), which contains the NHS Economic Evaluation Database, the Health Economic Evaluations Database and other Cochrane Library databases. Search terms used (as specified for MEDLINE) were ‘hormone therapy’ , OR ‘hormone replacement therapy, OR (hormon$ or estrogen or oestrogen) adj (treatment or therp$)’ , OR ‘hormone substitution’ AND ‘cost’ , OR ‘cost-utility’ , OR ‘cost-effective$’ , OR ‘costs and cost analysis’ [explode]. Terms were searched in all fields. Searches were limited to those conducted in humans and in females, with no language or other restrictions. Reference lists of identified papers were also searched for further relevant source articles. The search strategy was based on Cochrane Review recommendations .
Inclusion criteria were a priori defined as: CE or cost-utility analyses of MHT verses no treatment, in a population of healthy women, considering a range of long term health outcomes related to MHT use (breast cancer, coronary heart disease, stroke, fractures, colorectal cancer ) and menopausal symptom relief, with outcomes expressed in cost per quality-adjusted life year (QALY) or cost per life year. Exclusion criteria were analyses conducted for women with a pre-existing condition or a higher risk for a disease than the general population, analyses without inclusion of MHT-related long-term health outcomes or exclusion of the beneficial effect of menopausal symptom relief, or articles estimating only net costs of MHT use. Two investigators (LV, US) independently conducted the searches, reviewed titles and abstracts followed by the full texts of selected publications according to eligibility criteria and extracted data from studies using a structured form. Disagreements and queries at each stage of this process were resolved by discussion with a third investigator (KC).
Eligible publications were assessed for the completeness of their reporting using the 24-item CHEERS (Consolidated Health Economic Evaluation Reporting Standards) checklist  developed by the International Society for Pharmacoeconomics (ISPOR). Full adherence to any item recommendation was noted as ‘yes’ , with partial adherence as ‘incomplete’ and non-adherence as ‘no’. Eligible papers were also assessed for methodological quality using a 36-point checklist  which is based on the Drummond checklist . The checklist considers the elements of study design, data collection, analysis and result interpretation which are expected in a sound economic evaluation.
Assessment of data sources for MHT-related utility
The quality of life benefits or ‘utility’ associated with menopausal symptom relief following MHT use is formally accounted for in a cost-effectiveness assessment via inclusion of utility scores. These scores allow the impact of the quality of life benefit to be quantitatively assessed. The source data for the utility values for menopausal relief used in each CE assessment were documented and assessed. If more than one publication was referenced by the CE analysis the methodology for combining utility values from different studies was considered.
Assessment of costing methods used in primary studies
The evaluation of the cost-effectiveness of MHT depends on the quality of the source data for all relevant health care costs in both the ‘no treatment’ scenario, and in the MHT treatment scenario, including all downstream costs relating to MHT-related outcomes. To evaluate the methodology for assessing costs, each primary CE evaluation was examined for sources of unit costs, method of cost aggregation and country of study. If multiple references were cited for costs related to a particular health outcome, then the method of identifying costs (e.g. literature review or expert opinion) as well as the process used by authors to aggregate costs, was also extracted. In addition, the CE evaluations were examined in terms of presentation of separate unit costs and resource quantities and whether ranges of costs were provided to reflect varying degrees of disease severity or staging, where applicable. Any additional costing issues identified in each evaluation were also noted.
Assessment of particular health outcomes included in economic models
Using the MHT-related health outcomes identified in the synthesis of the worldwide data on MHT risks and benefits conducted by the UK MHRA , the following chronic diseases were assessed for inclusion in the economic models: breast cancer, colorectal cancer, CHD, deep vein thrombosis, endometrial cancer, fractures, ovarian cancer, pulmonary embolism, stroke and venous thromboembolic disease. Other health outcomes associated with MHT for which some randomised control trial evidence exists  were also assessed in our review for completeness, but these did not contribute to the current assessment of model quality. These included: urinary incontinence ; gallbladder disease ; and dementia . Data sources from which the relative risk was chosen for each outcome, were assessed for each included CE assessment.
Assessment of the comprehensiveness of the models
To further evaluate the comprehensiveness of the CE evaluations with respect to the completeness of their representation of disease outcomes, breast cancer was used as an example outcome. Factors for which evidence supports a role as an effect modifier for breast cancer outcomes i.e.: duration of MHT use, MHT type, body mass index and time of initiating MHT in relation to menopause, were examined for inclusion in CE models. Relative risks for breast cancer associated with estrogen-only and estrogen plus progestin MHT were assessed and the source data documented.
This article is a systematic and critical review of economic evaluations of the cost-effectiveness of MHT and therefore Ethics Committee approval was not required.
Selection of studies
Figure 1 summarises the search process conducted. Electronic literature searches originally identified a total of 1691 citations. After accounting for duplicates, 1526 publications remained. Of these, 1514 were rejected when the title and the abstract were reviewed and found not to be relevant (e.g. referred to other hormone/endocrine therapies such as growth hormone, fertility medication or adjuvant cancer therapy). The full text of the remaining 12 publications was examined. Seven articles were excluded for the following reasons: not based on analysis for healthy women but relevant only to a subgroup of women with osteoporosis or at an increased risk of fracture [34,35,36]; not including a range of MHT-related chronic disease outcomes in the economic model [37, 38], and estimation of net costs of MHT use rather than a CE analysis . One article  compared MHT use versus no therapy, but was conducted from the perspective of osteoporosis prevention. The evaluation considered the annual costs and outcome impacts from the use of MHT, or raloxifene or alendronate (agents for primary prevention of osteoporosis), by postmenopausal women over a 7 year period. This article was excluded from the review because its economic model did not include the beneficial effect of menopausal symptom relief on the calculated QALYs and considered only three clinical outcomes: fractures, myocardial infarction and breast cancer. The article concluded that MHT use resulted in net harm, although this was not based on a full assessment of the harms and benefits of MHT use. The remaining five articles [41,42,43,44,45] met the inclusion criteria and were included in the systematic review. It should be noted that after our initial search was completed a CE evaluation of MHT and non-MHT interventions for alleviating vasomotor symptoms in menopausal women was undertaken for NICE and was published [National Institute for Health and Care Excellence. Menopause: Appendix L-Health Economics. (NICE guidelines 23). 2015. https://www.nice.org.uk/guidance/NG23/documents/menopause-appendix-l2]. However, this evaluation was not included in our review because the included outcomes (which were agreed with a Guidelines Development Group) included vasomoter symptoms, vaginal bleeding, discontinuation of treatment, breast cancer, and venous thromboembolism, but not other chronic disease outcomes. The explicit focus was on short-term MHT use (5 years or less) and on the comparison with other treatment alternatives for short term use and therefore, the scope was somewhat distinct from the included analyses in our review.
Table 1 summarises some of the characteristics, main assumptions and key findings of the included MHT CE evaluations. Two evaluations were conducted in the US and three in Europe, each modelling cohorts of women using MHT at the age of 50 years and over, for durations of use of between 5 and 15 years. All included studies used the assumption that combination MHT or estrogen-only MHT were exclusively used by women with or without a uterus, respectively. Four models considered a time horizon of 50 years or over the lifetime of the cohorts, whereas one evaluation considered a period of 9 years. All evaluations used a 3% discount rate and willingness to pay thresholds which were in line with country-specific guidelines. All studies had at least one author who declared a potential for perceived conflict of interest (i.e. were employed by or had accepted speaker or consulting fees, or funding for the evaluation or writing assistance from an organisation which could reasonably be seen to have an interest in the findings of the evaluation).
Overall, the evaluations concluded that MHT use by women 50-60 years of age was below the indicative willingness-to-pay threshold of the country for which the analysis was conducted; thus all evaluations concluded that MHT could be cost-effective. Although the incremental cost-effectiveness ratio (ICER) values from each evaluation cannot be directly compared due to difference in currencies and the year for which costs were obtained, the three most recent evaluations [41,42,43], which all used 2006 costs, found that the cost per QALY for MHT use by 50 year old women ranged from $744 to $2,803 US dollars for non-hysterectomised women and from $263 to $295 US dollars for hysterectomised women, as at April 2017 (Table 1). These estimated ICERs are well under the relevant indicative willingness to pay threshold (which in US context, for example, is $50-100,000 US dollars). Even in older women (>65 years), one evaluation found that MHT would have an ICER of less than $30,000 US, which is again less than the relevant indicative willingness to pay threshold .
The assessment of the included evaluations according to the CHEERS checklist  is presented in Table 2. Although the historical context of the included primary studies must be borne in mind, incomplete reporting was observed for all included evaluations, and in many cases reporting was incomplete across most of the reporting domains in CHEERS (title/abstract, introduction, methods, results, discussion and other). Table 3 presents the findings of the quality assessment of the included evaluations using an extended Drummond checklist . The included evaluations met most checklist criteria in the study design section, however, for all evaluations nearly half of the criteria for data collection, analysis and interpretation of results were either partially addressed or not addressed.
Data sources for MHT-related utility
Three studies [42, 43, 45] used MHT-related utility values from one study . These CE evaluations used the QOL improvement of women with mild menopausal symptoms and the QOL improvement of women with severe menopausal symptoms to produce an average utility value to inform their models. In the Finnish CE evaluation  the MHT-related utility value was based on data from a randomized intervention trial of estrogen and progestin. The trial was conducted in Finland, where 419 women were randomised to six parallel treatment groups, receiving 4 dose combinations with no placebo group [47, 48]. The economic evaluation was conducted for three of the four dose combinations and therefore data were restricted to 279 study participants. The 15D health related QOL instrument was used at year 6 and 9 of the study by which time participants had dropped to 210 and 58 at years 6 and 9, respectively. The CE evaluation did not provide any details about which QOL utility values were used from the randomised trial, nor how results were aggregated from the various treatment groups. In the remaining CE evaluation  a willingness-to-pay survey and articles with different preference classification systems were used to produce a MHT-related utility score although the method used to generate this score was not detailed.
All studies assigned health state disutilities to chronic disease conditions associated with MHT use in their model accounting for the QOL decrements associated with potential adverse effects of MHT use . However, in two studies [42, 43] assumptions were made on the disutilities of having colorectal cancer and VTE (0.9 assigned to both conditions) but these were not subsequently varied in sensitivity analyses. In another study , disutilities due to chronic diseases were not mentioned in the methods nor were they presented, although results of a sensitivity analysis varying the disutility score of all events was presented.
Summary of costing methods and other costing issues
All evaluations used treatment costs reported from other studies. Two evaluations based treatment costs on studies conducted in the same country as the economic evaluation [42, 45] whereas the remaining articles derived costs from a combination of studies from the same country and others [41, 43, 44]. No evaluation provided complete information on how data sources were selected from the available literature, nor was the methodology used to combine the data from different sources described. No evaluations provided unit costs or resource quantities. In some evaluations [42,43,44,45] costs were referenced to previously conducted CE evaluations or other secondary studies instead of primary references, making it difficult to trace the original data sources. Presentation of a range of costs according to the seriousness or stage of a disease also varied between evaluations; one  provided both an average cost and ranges for all outcome categories, two studies had ranges for some diseases such as fracture and breast cancer [42, 43] whereas another  provided a single cost estimate for each disease outcome. Three evaluations [42, 43, 45] did not conduct any sensitivity analysis for MHT-associated costs (except varying the discount rate) and one evaluation  tested one higher and one lower value for costs associated with some chronic diseases (breast cancer, colon cancer and stroke) but only one evaluation conducted one-way sensitivity analysis and probabilistic sensitivity analysis . Two evaluations were conducted from the perspective of health care [42, 44] considering direct costs and three from a societal prospective [41, 43, 45]. One evaluation  used direct medical costs associated with the first year of treatment for all chronic diseases evaluated and no long-term costs. One  assumed that there were no long-term direct costs associated with VTE and vertebral fractures. In another evaluation,  it was not clear how the authors selected direct costs for breast cancer treatment based on the single source article referenced; the source article estimated costs for women aged 65 years and older with early-stage breast cancer. Another article  did not present values for any treatment costs included in their model and therefore it was not possible to compare costs used in the evaluation against the source references provided. A total cost per age (50, 55 and 60 years old) was given for MHT-users and women not using MHT, according to whether they had a uterus or were hysterectomised, and regardless of age, the difference in health costs associated with women not using MHT versus MHT users were SEK 9,739–13,645 (as at April 2017 equivalent to ~ €1,014–1,420; ~US $1,100–1,541).
MHT health related outcomes
MHT-associated health outcomes and data sources included in the CE evaluations are presented in Table 4. All evaluations considered the following as separate outcomes: breast cancer, colorectal cancer, CHD, stroke, pulmonary embolism, deep vein thrombosis and fractures, except one evaluation  where deep vein thrombosis was not considered, and CHD and stroke were considered as a compound outcome. None of the identified studies explicitly included ovarian cancer as an outcome. None of the evaluations incorporated other effects of MHT use, such as gallbladder disease. Three out of 5 evaluations [42, 43, 45] used WHI data to inform relative risks for the included health outcomes, one evaluation  used data from trials (including the WHI) and observational studies whereas one study used its own trial data .
Comprehensiveness of models
The relative risk of breast cancer associated with MHT used in three articles [42, 43, 45] varied according to the type of MHT modelled (oestrogen-only or combination). For one evaluation  breast cancer risks for both MHT types were pooled together. Three studies considered use of MHT for 5 years [42, 43, 45], however, of the two studies [41, 44] that considered MHT use for longer than 5 years, only one  increased the relative risk with increasing time periods of MHT use. Body mass index (BMI) and time of initiating MHT in relation to the menopause were not considered in any of the CE evaluations.
To our knowledge, this is the first time economic evaluations of the CE of MHT have been critically reviewed in a systematic manner, using a standardised framework. Five evaluations, identified since 2002, met the inclusion and exclusion criteria. Although all evaluations included consideration of breast cancer, colorectal cancer, stroke, CHD, and fractures in their outcomes, none included the full range of known MHT-associated health effects which have been summarised in independent regulatory reviews . Assessment of the evaluations using the CHEERS and Drummond checklists identified incomplete reporting in various categories which hindered effective review and interpretation of study findings. For women 50 to 60 years of age, all evaluations found MHT to be cost-effective and below the indicative willingness-to-pay threshold of the country for which the analysis was conducted. Our findings must be interpreted in historical context with respect to the included studies - firstly, reporting standards for economic evaluations have improved over time and since the publication of the primary studies; and secondly, quantitative independent syntheses of the long term risks and benefits of MHT were not necessarily available at the time of some of the primary study analyses (for example, the quantitative synthesis by the independent regulatory agency, the UK MHRA, was published in 2007). Nevertheless, our findings do influence the interpretation of the results of the included primary studies, since we identified reporting and methodological issues with the evaluations which could impact the outcome of MHT CE analyses and the generalisability of their results.
The current systematic review has certain limitations. It should be noted that only the main parameters of economic models were evaluated, rather than all parameters. For example, only factors related to breast cancer risk were reviewed when assessing the comprehensiveness of the models. However, breast cancer is one of the most important disease outcomes related to MHT use and effect modifiers of this association have been well documented. Issues similar to those identified from the current methodological evaluation for breast cancer may also apply to the other MHT-related health outcomes, although further investigation would be required to confirm this. The current review was also constrained by the limitations of the included studies. There was incomplete reporting for a number of parameters as identified by the quality assessment tools used, especially in terms of costing methodology and presentation of disaggregated costs, to enable a more detailed quality review. However, despite these limitations a number of findings have been identified that are constructive and informative for future CE evaluations.
In addition to choosing the appropriate health outcomes, factors that can modify the relationship between the intervention and a health outcome, need to be carefully considered in the model’s structure. We chose to assess the comprehensiveness of the CE evaluations for MHT by specifically examining the modelling of effect modification related to MHT-associated breast cancer risk. Although MHT type and varying duration of use were considered in some evaluations, body mass index and the timing of MHT initiation in relation to the menopause [17, 26, 27] were not accounted for. A body of evidence exists to suggest that the risk of breast cancer is greater in thinner MHT users than overweight or obese users (for example one large scale analysis found for estrogen-only MHT: RR 1.65 (95% CI 1.54-1.76) for BMI <25 kg/m2; RR 1.22 (95% CI 1.15-1.30) for BMI ≥25 kg/m2; combined MHT: RR 2.20 (95%CI 2.11-2.30) for BMI <25kg/m2; RR 1.81 (95%CI 1.73-1.9) for BMI >25kg/m2) . We suggest that future evaluations should consider all effect modifiers when modelling the CE of MHT use and in sensitivity analyses.
A favourable CE outcome for MHT is driven by its effects, or potential effects, on menopausal symptom relief, fractures, and potentially colorectal cancer . Of these, the alleviation of menopausal symptoms could be considered the principal reason women would be using MHT and therefore the utility value (preference) related to MHT would be the main driver of QOL which would increase the cost-effectiveness of CE. Three evaluations [42, 43, 45] used utility values from a single study . In addition to its small sample size (n = 104) and limited assessment (2 time-trade off questions asked), this study was conducted in Sweden and use of its data may not be widely applicable, as utility values have previously been shown to vary between countries [49–51]. Quality of life improvements for menopausal symptom relief will act to increase the CE of MHT use. Conversely, the adverse health effects will act to decrease the cost-effectiveness. We suggest that future CE evaluations should consider all these effects in their models and carefully consider data sources for utility weights.
Given the significant costs associated with the treatment of chronic disease outcomes related to MHT use, it is important that costs are accounted for appropriately in a CE analysis of MHT. As treatment patterns, treatment availability and clinician preferences can differ  across health systems and countries  and methods used to collect costs can also vary (e.g. micro-costing, case-mix grouping, use of charges), aggregating the results of different economic evaluations requires a clear methodology which explains how overall findings were calculated. It is also essential that the unit costs for each resource, resource quantities and methods for aggregating costs from various sources are provided in any economic evaluation  for clarity and transparency. Details of costing methods also need to be provided so that costs included in models for CE analyses can be verified. In addition, the potential effects of MHT on costs related to outcomes such as gallbladder disease, urinary incontinence and dementia were not included in the CE evaluations performed to date.
The literature search for the current systematic review was conducted at a similar time as that for the literature review of economic evaluations for the NICE clinical guidelines on menopause [National Institute for Health and Care Excellence. Menopause: diagnosis and management of menopause. (NICE guideline 23.) 2015. www.nice.org.uk/guidance/ng23]. The review by NICE aimed at finding economic evidence relating to short term treatments for menopausal symptoms, and included tibolone in addition to estrogen-only MHT and combination therapy. The studies that were identified in the NICE review were assessed for their relevance to one of the clinical questions posed by an expert reference group (the Guidelines Development Group) which was ‘what is the most clinical and cost-effective treatment for the relief of individual menopause-related symptoms for women in menopause’. The conclusion of the review was that ‘no published health economic literature was identified that addressed the breadth of treatment alternatives included in the network meta-analysis for this guideline’. NICE then commissioned a de novo economic evaluation, which, as previously noted, had a different focus and different outcome criteria to those included in the current review, according to our pre-specified primary study inclusion criteria. This evaluation assessed the CE of 5 years of use of MHT, non-MHT drugs and other interventions for alleviating vasomotor symptoms in menopausal women aged 50 years old. It was concluded that ‘the model suggests that transdermal oestradiol and progestogen was the most cost-effective treatment in women with a uterus and that is reflected in the recommendation of this guideline. However, the Guidelines Development Group didn’t think the evidence was sufficiently strong to completely overturn clinical practice and the use of much cheaper oral oestadiol and progestogen as the principle first line treatment.’ The evaluation also concluded that ‘non-oral oestradiol was cost-effective in women without a uterus although this model relied more heavily on extrapolated data. The guideline recommendations for women without a uterus mirror the recommendations for women with a uterus with a choice given between the use of oral and transdermal preparations with the same rationale.’
As was the case for the NICE review, some of our identified primary studies focused on use of MHT for periods of not more than 5 years. In interpreting the findings at a population level, however, it should be borne in mind that substantial numbers of women continue to use MHT for durations of longer than 5 years. For example, in one recent study in Australia, three-quarters of current-users had used MHT for ≥5 years . Therefore, deriving a picture of actual cost-effectiveness at a population-level requires consideration of the actual use of MHT in that population.
In conclusion, this critical assessment of cost-effectiveness evaluations of MHT identified a range of methodological issues affecting the interpretation of their findings and incomplete reporting of parameters which hindered effective review and transparency. Our findings must be interpreted in historical context with respect to the work presented in the included studies - firstly, reporting standards for economic evaluations have improved over time; and secondly, quantitative independent syntheses of the long term risk and benefits of MHT were not necessarily available at the time of some of the primary study analyses (for example, the quantitative synthesis by the independent regulatory agency, the UK MHRA, was published in 2007). Nevertheless, our findings do influence the interpretation of the results of the included primary studies. Our review emphasises the considerable difficulties in conducting cost-effectiveness analyses in situations where short term benefits of an intervention must be evaluated in the context of long term health outcomes.We recommend that any future cost-effectiveness assessments of MHT consider the current indications for use and the current recommendations by regulatory agencies for cautious targeted use; at the same time, a population-level assessment of cost-effectiveness should optimally account for the actual proportion of long-duration users (5 years or more), and account for the consequent impact on the risks of chronic disease in this group. We also recommend that future evaluations consider the full range of known beneficial and harmful health outcomes and consider the established effect modifiers for such health outcomes. Comprehensive costing and health state utility studies should be performed to support future evaluations, and we recommend that these studies account for all health outcomes for which there is an established association with MHT use. In addition, the ISPOR CHEERS statement as elaborated in the task force guidance report should also be adhered to, so as to facilitate interpretation of findings and effective comparison of future cost-effectiveness assessments of MHT .
Body mass index
Coronary heart disease
Consolidated health economic evaluation reporting standards
Incremental cost-effectiveness ratio
Medicines and Healthcare products Regulatory Agency
Menopausal hormone therapy
National Institute for Health and Care Excellence
Quality-adjusted life year
Quality of life
Venous thromboembolic event
Women’s Health Initiative.
Board of the International Menopause Society, Pines A, Sturdee DW, Birkhäuser MH, Schneider HP, Gambacciani M, Panay N. IMS updated recommendations on postmenopausal hormone therapy. Climacteric. 2007;10:181–94.
Daly E, Roche M, Barlow D, Gray A, McPherson K, Vessey M. MHT: an analysis of benefits, risks and costs. Br Med Bull. 1992;48:368–400.
Cheung AP, Wren BG. A cost-effectiveness analysis of hormone replacement therapy in the menopause. MJA. 1992;156:312–6.
Weinstein MC. Estrogen Use in Postmenopausal Women — Costs, Risks, and Benefits. N Engl J Med. 1980;303:308–16.
UK Medicines and Healthcare products Regulatory Agency (UKMHRA). Drug Safety Update September 2007, vol 1 issue 2: 2. http://www.mhra.gov.uk/home/groups/s-par/documents/websiteresources/con2032228.pdf. Accessed 30 May 2016.
Velentzis LS, Banks E, Sitas F, Salagame U, Tan EH, Canfell K. Use of menopausal hormone therapy and bioidentical hormone therapy in Australian women 50 to 69 years of age: results from a national, cross-sectional study. PLOS ONE. PLoS One. 2016;11:e0146494.
Hulley S, Furberg C, Barrett-Connor E, Cauley J, Grady D, Haskell W, et al. Non-cardiovascular disease outcomes during 6.8 years of hormone therapy: Heart and Estrogen/progestin Replacement Study Follow-up (HERS II). JAMA. 2002;288:58–66.
Schairer C, Lubin J, Troisi R, Sturgeon S, Brinton L, Hoover R. Menopausal estrogen and estrogen-progestin replacement therapy and breast cancer risk. JAMA. 2000;283:485–91.
Ross RK, Paganini-Hill A, Wan PC, Pike MC. Effect of hormone replacement therapy on breast cancer risk: estrogen versus estrogen plus progestin. J Natl Cancer Inst. 2000;92:328–32.
Magnusson C, Baron JA, Correia N, Bergstrom R, Adami H-O, Persson I. Breast-cancer risk following long-term oestrogen and oestrogen-progestin-replacement therapy. Int J Cancer. 1999;81:339–44.
Collaborative Group on Hormonal Factors in Breast Cancer. Breast cancer and hormone replacement therapy: collaborative reanalysis of data from 51 epidemiological studies of 52,705 women with breast cancer and 108,411 women without breast cancer. Lancet. 1997;350:1047–59.
Schulman SP, Thiemann DR, Ouyang P, Chandra NC, Schulman DS, Reis SE, et al. Effects of acute hormone therapy on recurrent ischemia in postmenopausal women with unstable angina. J Am Coll Cardiol. 2002;39:231–7.
Grodstein F, Manson JE, Colditz GA, Willit WC, Speizer FE, Stampfer MJ. A prospective, observational study of postmenopausal hormone therapy and primary prevention of cardiovascular disease. Ann Intern Med. 2000;133:933–41.
Herrington DM, Reboussin DM, Brosnihan KB, Sharp PC, Shumaker SA, Snyder TE, et al. Effects of estrogen replacement on the progression of coronary artery atherosclerosis. N Engl J Med. 2000;343:522–9.
Hulley S, Grady D, Bush T, Furberg C, Herrington D, Riggs B, et al. Randomized trial of estrogen plus progestin for secondary prevention of coronary heart disease in postmenopausal women. JAMA. 1998;280:605–13.
Rossouw JE, Anderson GL, Prentice RL, LaCroix AZ, Kooperberg C, Stefanick ML, et al. Risks and benefits of estrogen plus progestin in healthy postmenopausal women: principal results From the Women's Health Initiative randomized controlled trial. JAMA. 2002;288:321–33.
Beral V. Million Women Study Collaborators. Breast cancer and hormone-replacement therapy in the Million Women Study. Lancet. 2003;362:419–27.
Reeves GK, Beral V, Green J, Gathani T, Bull D, Collaborators MWS. Hormonal therapy for menopause and breast-cancer risk by histological type: a cohort study and meta-analysis. Lancet Oncol. 2006;7:910–8.
Fenton A, Panay N. The Women's Health Initiative - a decade of progress. Climacteric. 2012;15:205.
US Preventive Services Task Force. Menopausal Hormone Therapy: Preventive Medication. http://www.uspreventiveservicestaskforce.org/uspstf/uspspmho.htm. Accessed 30 May 2016.
Nelson HD, Walker M, Zakher B, Mitchell J. Menopausal Hormone Therapy for the Primary Prevention of Chronic Conditions: A Systematic Review to Update the U.S. Preventive Services Task Force Recommendations. Ann Intern Med. 2012;157:104–13.
Moyer VA. U.S. Preventive Services Task Force. Menopausal hormone therapy for the primary prevention of chronic conditions: U.S. Preventive Services Task Force recommendation statement. Ann Intern Med. 2013;158:47–54.
Lokkegaard E, Lidegaard O, Moller LN, Agger C, Andreasen AH, Jorgensen T. Hormone replacement therapy in Denmark, 1995–2004. Acta Obstet Gynecol Scand. 2007;86:1342–51.
Hersh AL, Stefanick ML, Stafford RS. National use of postmenopausal hormone therapy: annual trends and response to recent evidence. JAMA. 2004;291:47–53.
Drummond MF, Sculpher MJ, Torrance GW, O’Brien BJ, Stoddart GL. Methods for the economic evaluation of health care programmes. New York: Oxford University Press; 2005. p. 55–95.
Prentice RL, Manson JE, Langer RD, Anderson GL, Pettinger M, Jackson RD, et al. Benefits and risks of postmenopausal hormone therapy when it is initiated soon after menopause. Am J Epidemiol. 2009;170:12–23.
Beral V, Reeves G, Bull D, Green J. for the Million Women Study Collaborators. Breast Cancer Risk in Relation to the Interval Between Menopause and Starting Hormone Therapy. J Natl Cancer Inst. 2011;103:1–10.
Higgins AM, Harris AH. Health economic methods: cost-minimization, cost-effectiveness, cost-utility, and cost-benefit evaluations. Crit Care Clin. 2012;28:11–24.
Husereau D, Drummond M, Petrou S, Carswell C, Moher D, Greenberg D, et al. Consolidated health economic evaluation reporting standards (CHEERS)—Explanation and elaboration: A report of the ISPOR health economic evaluations publication guidelines good reporting practices task force. Value Health. 2013;16:231–50.
Centre for Review and Dissemination. Systematic reviews of economic evaluations. In: Systematic Reviews, CRD’s guidance for undertaking reviews in health care. CRD, York Publishing Service Ltd. University of York; 2009. p. 199-218. https://www.york.ac.uk/media/crd/Systematic_Reviews.pdf.
Hendrix SL, Cochrane BB, Nygaard IE, Handa VL, Barnabei VM, Iglesia C, et al. Effects of estrogen with and without progestin on urinary incontinence. JAMA. 2005;293:935–48.
Cirillo DJ, Wallace RB, Rodabough RJ, Greenland P, LaCroix AZ, Limacher MC, et al. Effect of estrogen therapy on gallbladder disease. JAMA. 2005;293:330–9.
Shumaker SA, Legault C, Rapp SR, Thal L, Wallace RB, Ockene JK, et al. Estrogen plus progestin and the incidence of dementia and mild cognitive impairment in postmenopausal women: the Women's Health Initiative Memory Study: a randomized controlled trial. JAMA. 2003;289:2651–62.
Lekander I, Borgström F, Ström O, Zethraeus N, Kanis JA. Cost effectiveness of hormone therapy in women at high risks of fracture in Sweden, the US and the UK--results based on the Women's Health Initiative randomised controlled trial. Bone. 2008;42:294–306.
Mobley LR, Hoerger TJ, Wittenborn JS, Galuska DA, Rao JK. Cost-effectiveness of osteoporosis screening and treatment with hormone replacement therapy, raloxifene, or alendronate. Med Decis Making. 2006;26:194–206.
Fleurence R, Torgerson DJ, Reid DM. Cost-effectiveness of hormone replacement therapy for fracture prevention in young postmenopausal women: an economic analysis based on a prospective cohort study. Osteoporos Int. 2002;13:637–43.
Botteman MF, Shah NP, Lian J, Pashos CL, Simon JA. A cost-effectiveness evaluation of two continuous-combined hormone therapies for the management of moderate-to-severe vasomotor symptoms. Menopause. 2004;11:343–55.
Coyle D, Cranney A, Tugwell P. Economic evaluation of norethisterone acetate/ethinylestradiol (FemHRT) for women with menopausal symptoms. Pharmacoeconomics. 2003;21:661–9.
Ohsfeldt RL. Estimating the cost effectiveness of alternative drug treatments for postmenopausal osteoporosis. Expert Rev Pharmacoecon Outcomes Res. 2004;4:637–44.
Mullins CD, Ohsfeldt RL. Modeling the annual costs of postmenopausal prevention therapy: raloxifene, alendronate, or estrogen-progestin therapy. J Manag Care Pharm. 2003;9:150–8.
Salpeter SR, Buckley NS, Liu H, Salpeter EE. The cost-effectiveness of hormone therapy in younger and older postmenopausal women. Am J Med. 2009;122:42–52.
Lekander I, Borgström F, Ström O, Zethraeus N, Kanis JA. Cost-effectiveness of hormone replacement therapy for menopausal symptoms in the UK. Menopause Int. 2009;15:19–25.
Lekander I, Borgström F, Ström O, Zethraeus N, Kanis JA. Cost-effectiveness of hormone therapy in the United States. J Womens Health (Larchmt). 2009;18:1669–77.
Ylikangas S, Bäckström T, Heikkinen J. Cost-effectiveness of continuous combined hormone replacement therapy in long-term use: economic evaluation based on a 9-year study in Finland. Curr Med Res Opin. 2007;23:57–64.
Zethraeus N, Borgström F, Jönsson B, Kanis J. Reassessment of the cost-effectiveness of hormone replacement therapy in Sweden: results based on the Women's Health Initiative randomized controlled trial. Int J Technol Assess Health Care. 2005;21:433–41.
Zethraeus N, Johannesson M, Henriksson P, Strand RT. The impact of hormone replacement therapy on quality of life and willingness to pay. Br J Obstet Gynaecol. 1997;104:1191–5.
Heikkinen J, Vaheri R, Timonen U. Long-term safety and tolerability of continuous-combined hormone therapy in postmenopausal women: results from a seven-year randomised comparison of low and standard doses. J Br Menopause Soc. 2004;10:95–102.
Heikkinen JE, Vaheri RT, Ahomäki SM, Kainulainen PM, Viitanen AT, Timonen UM. Optimizing continuous-combined hormone replacement therapy for postmenopausal women: a comparison of six different treatment regimens. Am J Obstet Gynecol. 2000;182:560–7.
Galante J, Augustovski F, Colantonio L, Bardach A, Caporale J, Marti SG, et al. Estimation and comparison of EQ-5D health states' utility weights for pneumococcal and human papillomavirus diseases in Argentina, Chile, and the United Kingdom. Value Health. 2011;14:S60–4.
Brunner RL, Gass M, Aragaki A, Hays J, Granek I, Woods N, et al. Effects of conjugated equine estrogen on health-related quality of life in postmenopausal women with hysterectomy: results from the Women's Health Initiative Randomized Clinical Trial. Arch Intern Med. 2005;165:1976–86.
Hays J, Ockene JK, Brunner RL, Kotchen JM, Manson JE, Patterson RE, et al. Effects of estrogen plus progestin on health-related quality of life. N Engl J Med. 2003;348:1839–54.
Higgins JPT, Green S. Cochrane Handbook for Systematic Reviews of Interventions Version 5.1.0 [updated March 2011]. The Cochrane Collaboration, 2011. Available from http://handbook.cochrane.org.
Cox HL, Laupland KB, Manns BJ. Economic evaluation in critical care medicine. J Crit Care. 2006;21:117–24.
We thank Cancer Council NSW for funding the study. KC receives salary support from the National Health and Medical Research Council Australia. We also thank Yaping Liu and Ying Zhao for help with obtaining relevant references.
This study was funded by Cancer Council NSW. The funding sources had no involvement in study design, data collection, analysis, interpretation of data, or writing of this article.
Availability of data and materials
This study is a systematic review of peer-reviewed articles. No datasets were required to support the conclusions of this article.
KC and LV conceived the idea for the review. LV and US performed the database search and extracted the data. LV tabulated and interpreted the data. LV drafted the manuscript with input from KC. All authors read and approved the final manuscript.
LSV and US have no competing interests to declare. KC is co-PI of an investigator-initiated trial of cytology and primary HPV cervical screening in Australia (‘Compass’) (ACTRN12613001207707 and NCT02328872) which is conducted and funded by the Victorian Cytology Service (VCS) Inc Ltd., a government-funded health promotion charity. The VCS Inc Ltd. have received a funding contribution for the Compass trial from Roche Molecular Systems and Ventana Inc., USA. KC is also a PI on Compass in New Zealand, (‘Compass NZ’) (ACTRN12614000714684) which is conducted and funded by Diagnostic Medlab (DML), now Auckland District Health Board. DML received an equipment and a funding contribution for the Compass trial from Roche Molecular Systems. However neither KC nor her institution on her behalf (Cancer Council NSW) received direct or indirect funding from industry for Compass Australia or NZ or any other project.
Consent for publication
Ethics approval and consent to participate
Springer Nature remains neutral with regard to jurisdictional claims in published maps and institutional affiliations.