Economic analysis of mental health services
Martin Knapp
Dan Chisholm
Introduction
Economics is concerned with the use and distribution of resources within a society, and how different ways of allocating resources impact on the well-being of individuals. Economics enters the health sphere because resources available to meet societal needs or demands are finite, meaning that choices have to be made regarding how best to allocate them (typically to generate the greatest possible level of population health). Economics provides an explicit framework for thinking through ways of allocating resources.
Resource allocation decisions in mental health are complicated by the fact that disorders are common, debilitating, and often long-lasting. Epidemiological research has demonstrated the considerable burden that mental disorders impose because of their prevalence, chronicity, and severity: globally, more than 10 per cent of lost years of healthy life and over 30 per cent of all years lived with disability are attributable to mental disorders.(1) Low rates of recognition and effective treatment compound the problem, particularly in poor countries.
However, disease burden is not in itself sufficient as a justification or mechanism for resource allocation or priority-setting. A disorder can place considerable burden on a population but if appropriate strategies to reduce this burden are absent or extremely expensive in relation to the health gains achieved, large-scale investment would be considered misplaced. The reason is that scarce resources could be more efficiently channelled to other burdensome conditions for which cost-effective responses were available. For priority-setting and resource allocation, it is necessary to ask what amount of burden from a disorder can be avoided by using evidence-based interventions, and at what relative cost of implementation in the target population.
Cost and cost-effectiveness considerations enter into health care reform processes, priority-setting exercises within and across health programmes, and regulatory decisions concerning drug approval or pricing. Two broad levels of economic analysis can be distinguished: macro and micro.
Economic analyses at macro level: the mental health system
Macro-level economic analyses are concerned with how health systems function and what they achieve. What, for example, are the motivations and behaviour of key ‘stakeholders’, with what implications for access, quality, and costs? What roles do economic forces play and can they be shaped to improve health outcomes and cost-effectiveness? Do different organizational or financial arrangements produce different resource configurations? For example, do markets achieve fairer or more efficient allocations than state bureaucracies? Through their macro analyses, economists can contribute to a better understanding of how health systems can improve utilization of resources (for example, see Box 7.7.1).
While improved psychological well-being in the population is likely to represent the primary goal of the mental health system, there are other (social) goals that could also feature prominently, including quality improvements in service provision, and financial (as well as human rights) protection for people with mental health problems.(2) Meeting these goals is achieved via a number of key health system functions, including resource generation, their allocation via appropriate modes of financing, actual provision of services, and overall stewardship and evaluation of these various functions.(3) Economic analysis contributes to policy formation relating to each of these functions.(4,5,6)
There are barriers to implementation of evidence-based mental health care in even generously resourced health systems. An insidious barrier is resource insufficiency: mental health services are often grossly under-funded in comparison to both needs and the
cost-effectiveness of interventions to meet them. This is a major issue for countries where the proportion of national income devoted to health care is low, or where the proportion of the health budget allocated to mental health is minimal. With limited funds it is difficult to build any kind of service system, because it is difficult to recruit, train, and retain skilled staff.
cost-effectiveness of interventions to meet them. This is a major issue for countries where the proportion of national income devoted to health care is low, or where the proportion of the health budget allocated to mental health is minimal. With limited funds it is difficult to build any kind of service system, because it is difficult to recruit, train, and retain skilled staff.
Box 7.7.1 The relevance of a health systems perspective
The need for a systems approach to mental health policy and planning is made apparent from a simple illustration: cheap, effective drugs exist for key neuropsychiatric disorders, including tricyclic antidepressants, conventional neuroleptics, and antiepileptic drugs, which are affordable even to resource-poor countries. The availability and prescription of these drugs to those in need, however, are determined by the extent to which such drugs have been distributed and by the ability of health care providers to detect and appropriately treat the underlying condition. Access to and use of such medications may further be hampered by the private cost of seeking and receiving health care, particularly if it is out-of pocket. User fees, provider incentives, and clinical practice are in turn influenced by the availability of national legislation, regulation, and treatment guidelines.
Even when resources are committed, available services might be poorly distributed, available at the wrong place or time relative to the distribution of needs. They may only be delivered by specialist clinics or concentrated in big cities, or affordable only by wealthier individuals. Improvements to practice take time to work through to improved health outcomes, cost-effectiveness gains or fairer access, even when suitable professionals can be recruited or new facilities opened. Decision makers must think long-term, for the immediate consequences of many interventions could be modest but longer-term benefits immense.(7)
A more general difficulty is that available services do not match what is needed or preferred. Indeed, there may be scant information on population or individual needs, and patients may have few opportunities to participate in treatment decisions. Another problem could be poor coordination of services because of professional rivalry, stultifying bureaucracy or ‘silo budgeting’ (resources held in one agency’s ‘silo’ cannot be allocated to other uses).
Economic analyses at micro level: cost-effectiveness
The most frequently posed micro questions relate to the cost-effectiveness of interventions, such as an emphasis on community-based care, the use of new drugs, or the development of secure accommodation. For example, consider what happens following development of a new treatment (say a new medication for schizophrenia). Decision makers want answers to two questions when considering whether to use or recommend this drug. The first is the clinical question: is it effective in alleviating psychotic symptoms and generally improving health-related quality of life? If the answer to the clinical question is ‘yes’, then there is a second question: is it cost-effective? That is, does the drug achieve the improved outcomes at a cost that is worth paying? The meaning of ‘worth’ is far from straightforward to establish and laden with controversy.
These two questions sit at the heart of economic evaluation: outcomes must be assessed (and compared between different treatments) and the (relative) costs of achieving them must be examined. Looking only at costs is not an economic evaluation.
There are different variants of economic evaluation. They share a common approach to the conceptualization, definition, and measurement of costs, but adopt different approaches when considering and measuring outcomes, primarily because they seek to answer slightly different questions. We set out these differences by discussing the questions a study might address, measuring costs and outcomes, making trade-offs between them, and utility and benefit measurement.
Question and perspective
The choice of evaluative approach depends on the question to be addressed. If the question is essentially clinical—what is the most appropriate treatment for someone with particular needs in particular circumstances—information is needed on the comparative costs of alternative treatments and comparative outcomes measured in terms of symptom alleviation, improved functioning, and so on. A cost-effectiveness analysis would be appropriate (see ‘Effectiveness measurement’ below).
If, to take a broader stance, the question is whether to treat depression rather than spending the funds elsewhere in the health system, then decision makers need to know the costs, but now need an outcome measure that uses a common metric across different health domains. The most common such metric is ‘utility’ and a costutility analysis would be undertaken (see ‘Utility measurement’ below).
To widen the perspective further, if the question is whether to increase expenditure in the health system or in (say) improving transport or launching a new environmental policy, then an evaluation needs to ask about the comparative costs and impacts of the different options, where ‘impact’ will need to be measured in a common unit across all public policy areas. The usual choice of broad measure is monetary, leading to cost–benefit analysis (see ‘Benefit measurement’ below).
The question to be addressed thus influences the type of evaluation needed, but the choices are not mutually exclusive: a single study can support more than one approach if the right measures are used. The broader the question, the lower the likelihood that the outcome measure will be sensitive to the particular circumstances of a specific disorder such as depression, but the greater the usefulness in terms of resource allocation decisions.
Linked to specification of the question to be addressed is the perspective of a study. Is the evaluation needed to help resource allocation within a particular agency (such as primary care clinic), or a particular system (such as the health care system), or the whole society? The perspective will determine the breadth of both cost and outcome measurement.
Cost measurement
Some costs are directly associated with a disorder or its treatment, such as the money spent on medications and services used by patients, and some are more indirect, measuring lost productivity because ill-health can disrupt someone’s employment pattern or the social cost of unpaid care provided by families. How broadly the costs are measured will depend upon the purpose of the study.
In carrying out evaluations in practice, economists need data on service use patterns by patients. This information might come from organizational ‘billing’ systems (recording amounts transferred between purchasers and providers for services used), or from routine information systems that record service contacts, or from research instruments that specifically collect data on service use patterns through interviews with patients, caregivers, or service professionals. One widely used instrument is the Client Service Receipt Inventory.(8)
The next task is to attach unit cost estimates to these service use data. In England, there is an excellent annual compendium of health and social care unit costs, which provides just such figures.(9) In other countries, it might be necessary to estimate unit costs anew. A range of data sources could be used, including government statistics, health system expenditure figures, and specific facility or organization accounts. The main cost categories to be quantified would be:
Effectiveness measurement
The most intuitive mode of economic evaluation is cost-effectiveness analysis (CEA): it measures costs as set out above, and outcomes along the dimensions that would be recognized by clinicians and used in clinical studies (changes in symptoms, behaviour, functioning, and so on). A CEA can help decision makers choose between interventions aimed at specific health needs. A cost-effectiveness analysis looks at a single outcome dimension—such as change in symptoms—and computes and compares the difference in costs between two treatments and the difference in this (primary) outcome. If one treatment is both more effective and less costly than another, then it would clearly be the more cost-effective of the two. But if it is more effective and more costly then a trade-off is needed (see below).
Often the economist will compute cost differences and a range of effectiveness differences (one for each outcome dimension)—an approach sometimes called cost-consequences analysis—which has the advantage of breadth but poses a challenge if one outcome is better and another worse for a particular treatment. It is then not always obvious which treatment is to be preferred, and the decision maker must weigh up the strength of evidence.
Making trade-offs
If an evaluation finds a new intervention to be both more effective but simultaneously more expensive than an older intervention, which is the more cost-effective of the two? A trade-off must be made between the better outcomes and the higher costs necessary to achieve them.
The classical way of determining this trade-off has been via the derivation of a incremental cost-effectiveness ratio (CER), which divides the extra cost associated with a new intervention by its additional effect (see Box 7.7.2).
More recently, health economists have developed the ‘net benefit approach’ to explicate the nature of the trade-off (see Box 7.7.3 for an example). It is commonly seen today in the construction of cost-effectiveness acceptability curves (CEACs). These curves show the probability that an intervention will be cost-effective for each of a number of pre-specified or implicit valuations of an outcome improvement by the decision maker.
Utility measurement
One way to overcome the potential problem of different outcome dimensions pointing in different directions is to employ a single, over-arching measure. A preference-weighted, health-related quality of life measure could be used. The value of the quality of life improvement is gauged in units of ‘utility’, usually expressed by a combined index of the mortality and quality of life effects of an intervention. The best known such index is the Quality Adjusted Life Year (QALY).
Related posts:
Stay updated, free articles. Join our Telegram channel
Full access? Get Clinical Tree
