Methodological challenges in economic evaluations of fertility treatments

Consider This
Methodological challenges in economic evaluations of fertility treatments


Willings Botha, Ph.D.

The National Perinatal Epidemiology and Statistics Unit, Centre for Big Data Research in Health, School of Women’s and Children’s Health, University of New South Wales

Consider This:

Economic evaluations are increasingly being undertaken in healthcare for policy formulation, practices, and resource allocation decisions. An economic evaluation compares two or more alternative treatments to assess which one produces more benefits or outcomes with optimal costs (1). The assessment of costs is similar for all economic evaluation methods. However, outcomes are assessed differently and determine the method of evaluation to be used. Below is a table that presents different economic evaluation methods together with some advantages and disadvantages:  

Method of evaluation

Outcome measurement



Cost-utility analysis (CUA)

Quality-adjusted life years (QALYs) which combines health-related quality of life (HRQoL) utilities with duration in a health state

A generic measure of health which allows comparisons across different treatments in healthcare sector based on acceptable willingness to pay for a health gain

Utilities vary depending on the method and instrument used for measuring HRQoL. It becomes difficult to make comparisons of treatments across sectors of the economy other than health

Cost-effectiveness analysis (CEA)

Natural units (live birth, healthy baby, pregnancy, lives saved, or clinical event avoided or achieved)

Outcomes are measured in their natural units without restrictions

Narrow in focus for allocation of resources across different treatments because comparisons can only be made with evaluations with similar outcome measure

Cost-minimization analysis (CMA)

Outcomes of alternative treatments considered equivalent with the lowest-cost intervention being preferred

Easy to undertake as it only looks at costs

Provides partial economic evaluation by considering only costs and assuming that outcomes are the same for treatments, which is rare in practice


Cost-consequence analysis (CCA)

Varied outcome measures presented in a disaggregated form.

Very transparent and allows the presentation of broad outcomes beyond health

Falls short of aggregating the total value of outcomes in a single measure, hence comparison with costs becomes difficult. Furthermore, the onus is on decision makers to choose the relevant information from the disaggregated outcomes which can be a burden to some

Cost-benefit analysis (CBA)

Outcomes expressed in monetary terms usually obtained through willingness to pay (WTP) approaches

Allows judgements on worthiness of treatment to be made within healthcare sector and or across other sectors of the economy, hence considered broader in focus and suitable for policy, practice and resource allocation decisions

Measuring health outcomes in monetary terms can be challenging, imply ability to pay, and has ethical concerns on valuing health or life


Using the above methods in fertility treatments raises considerable methodological challenges. Fertility treatments generate broad benefits or outcomes that go beyond health which include non-health and process outcomes. The objective of fertility treatments is to create new life rather than extend or improve health-related quality of life (HRQoL). Furthermore, fertility treatments usually involve multiple stakeholders such as mother, father, donor, and society which makes it difficult to establish the perspective of economic evaluation. Usually, infertile individuals would want to conceive now rather than wait and may want another child in future which adds to the problem of determining the time horizon for economic evaluations. It also becomes problematic to determine whether to consider the costs and consequences of the treatments before or after the child is born.

Currently, it is difficult to conduct economic evaluations of fertility treatments that incorporate and consider the complexity described above. This paper discusses the specific methodological challenges in the economic evaluations of fertility treatments using the available methods and recommends a way-forward.

When a cost-utility analysis (CUA) method is used, alternative treatments are compared in terms of costs and outcomes which are measured as quality-adjusted life years (QALYs) - a combination of HRQoL and quantity of life. The HRQoL utilities are often indirectly elicited using the EQ-5D off-the-shelf questionnaire by the EuroQol group from five dimensions: mobility, self-care, usual activities, pain/discomfort, and depression/anxiety (1).  The response scores to the dimensions result in an index which in turn is used to derive utilities from  the predefined tariffs obtained from the general population (2). Subsequently, the utilities are combined with the duration in a given health state to calculate QALYs.  The ratio of the differential between the cost and QALYs of the new and existing treatment known as the incremental cost-effectiveness ratio (ICER) allows comparisons across treatments to be made as cost per QALY gained. This ratio is used to inform policy, practices or aid decision-making as to whether a given treatment is worthwhile on the basis of an acceptable willingness to pay for each QALY gained.

When applied to fertility treatments, the QALY framework becomes problematic as it is only restricted to measuring and maximizing health outcomes. It totally disregards other non-health and process outcomes which may matter to individuals who seek fertility treatments such as the opportunity of motherhood despite small chances of having a live birth, and the conduct of clinical staff during treatment, for example. Another important question that arises is the determination of whose QALYs to assess among the different stakeholders involved in fertility treatments: mother, father, donor, and society. Furthermore, the decision-rule on the worthiness of treatment is based on an established willingness to pay threshold for a QALY gained which becomes a limitation in fertility treatments when this amount is non-existent and where the outcome measure is not a QALY but a live birth or achieved pregnancy.

Cost-effectiveness analysis (CEA) is another method which can be used in fertility treatments. Outcomes can be expressed in their natural units such cost per live birth, healthy baby or pregnancy. The objective can be to identify alternative treatment which produces more outcomes at the same cost or where the same outcomes can be achieved at a lower cost. As can be seen, the CEA offers a restrictive evaluative space to a single outcome measure and excludes other aspects of outcomes of fertility treatments. The complex nature of fertility treatments with multiple and varied outcomes beyond health clearly makes this method of analysis inadequate or unsuitable. It also becomes difficult to compare relative effectiveness of treatments with outcomes measured in different metrics such as live births or number of pregnancies.

Another option is to use cost-minimization analysis (CMA) method which only measures and compares costs related to a particular fertility treatment with those of alternative treatment(s) with the assumption that the outcomes are similar. This essentially, culminates to only measuring costs and ignoring the outcomes which results in a decision-rule solely based on costs, with the lowest-cost option being preferred. In practice, it is rare or very difficult to demonstrate fertility treatments with the similar outcomes. Generally, the assumption of similar outcomes for treatments in healthcare is very unrealistic and ineffective in informing policy, practice or healthcare resource allocation decision-making. Given that this method does not consider the outcomes of treatments on the basis of being equivalent, it does not qualify to be considered a full economic evaluation method in the strictest sense (1). For these reasons, this method has not gained much traction in the health arena and cannot be considered fit for purpose in economic evaluations of fertility treatments (3, 4).

Cost-consequences analysis (CCA) is also a useful method of economic evaluation in the context of fertility treatments. It allows listing of multiple and different elements of treatment under the cost and outcomes side in a disaggregated  outline format of a balance sheet (1). For example, different elements making up the cost of fertility treatment can be listed under costs while multiple outcomes such as live births, health babies, pregnancy rates, QALYs (where successful treatment may improve the psychological well-being), safety of treatment, obstetric, perinatal and neonatal complications reduced can be listed under the outcomes side. A single economic evaluation can incorporate various perspectives in terms of value judgement from varied interest groups of fertility treatments such as individuals, health providers, insurance companies, government and society. The drawback of a CCA method is that it does not compare costs with outcomes of treatments because it falls short of aggregating the total value of varied outcomes in a single measure. Furthermore, the onus is on decision-makers to choose the relevant information from the CCA balance sheet to make decisions which can be a burden to some.

Lastly, cost-benefit analysis (CBA) method can be used in economic evaluation of fertility treatments. It is distinct from the other methods in that it compares the costs and outcomes of treatment on a monetary scale. As such, it is possible to make judgements of whether a treatment is worthwhile within the healthcare sector and or across other sectors of the economy. For this reason, it is considered to have a broader focus, hence suitable to inform policy, practices and deal with resource allocation decisions. To assess the value of outcomes of treatment, the CBA method uses willingness to pay (WTP) estimates which are elicited using revealed preference (RP) and stated preference (SP) methods. The RP methods elicit individuals’ WTP by examining the actual real-life behaviour, which is rare in a healthcare situation because of lack of readily available healthcare data. For this reason, the SP methods, which elicit WTP values based on hypothetical scenarios (what individuals would do) as opposed to what they are observed to real situation, are the most preferred.

The SP methods include contingent valuation (CV) approaches and stated preference discrete choice experiments (SPDCE). CV approaches value non-market priced goods as whole using direct elicitation methods such as asking someone directly how much they are willing to pay for fertility treatment programs, for example while SPDCE approaches indirectly elicit individuals’ WTP values for hypothetical configuration of characteristics (attributes) of a fertility treatment program. The CV approach has been criticised for its implication on the ability to pay which is considered to discriminate those who cannot afford to pay; and its attempt to directly monetize health outcomes or life. These concerns directly relate to fertility treatments; hence CBA is rarely used for their economic evaluations. The good news is that the SPDCE method addresses the limitations of the CV method through indirect elicitation of WTP values through the marginal rate of substitution (MRS)- a ratio between any preferred characteristic of a treatment program and a cost of treatment expressed by the negative sign to indicate the marginal utility of income.

Of all the methods presented above, the QALY framework using cost-utility analysis (CUA) is the conventional economic evaluation approach in healthcare. However, as discussed earlier, CUA together with CEA, CMA and CCA methods are inadequate or unsuitable in the economic evaluation of fertility treatments. Compared to these methods, cost-benefit analysis (CBA) method using the SPDCE approach appears appealing and a suitable way-forward for evaluating the complex and broad outcomes of fertility treatments in monetary terms. For example, in a SPDCE for fertility treatments, individuals can be presented with choice-sets consisting of specially designed hypothetical scenarios of fertility treatment alternatives where at least one characteristic is varied systematically in terms of its levels. Individuals can then be asked to choose between the two fertility treatment programs which can also include opt-out. The assumption is that individuals make choices which give them the highest level of satisfaction by considering and trading-off all characteristics and their value judgement is based on a bundle of the characteristics. The SPDCE approach uses regression models and by including cost as part of the characteristics of a fertility treatment, trade-offs would reveal how much an individual is willing to pay for marginal changes in characteristics using the marginal rate of substitution (MRS) of the estimated parameters. This willingness to pay for a characteristic of a fertility treatment would represent the benefit to an individual. It would be possible, therefore, to assess whether offering this type of characteristic as part of a fertility treatment program provides a positive net benefit through comparison of the marginal WTP for the characteristic with the cost of its delivery in a cost-benefit analysis (CBA) framework. Furthermore, it would also be possible to aggregate the marginal WTP estimates for all characteristics of the most preferred configuration of fertility treatment program and compare the total WTP of this configuration with the total cost of delivering the fertility treatment program. The decision-rule would be to adopt a fertility treatment program if its net benefit is positive.

In addition, the SPDCE approach can also produce other measures for different policy questions in fertility treatments. For example, if  “Success rate” or “Travel time” form part of the characteristics of a preferred configuration of a fertility treatment program, it is possible to calculate the WTP for a given probability of having a baby; or the willingness to sacrifice travel time to access a fertility treatment program with preferred characteristics such as a fertility treatment program with a success rate of 25%; that offers the same doctor and nurse during the treatment as part of continuity of care, for example. The regression results can also reveal the extent to which characteristics of fertility treatments influence choice decisions through the statistical significance, sign, and size of the estimated parameters. The strength of the SPDCE approach lies in the use of characteristics of treatment without being restricted to any dimension of outcome and can go beyond health to include non-health and process-related outcome characteristics.



1.         Drummond MF, Sculpher MJ, Claxton K, Stoddart GL, Torrance GW. Methods for the Economic Evaluation of Health Care Programmes: Oxford University Press, 2015.

2.         Dolan P. Modeling valuations for EuroQol health states. Medical care 1997;35:1095-108.

3.         Briggs AH, O'Brien BJ. The death of cost-minimization analysis? Health Economics 2001;10:179-84.

4.         Dakin H, Wordsworth S. Cost-minimisation analysis versus cost-effectiveness analysis, revisited. Health Economics 2011:n/a-n/a.