National health expenditures represented nearly 18 percent of the U.S. gross domestic product (GDP) in 2015, while efforts to limit cost growth continue to be central objectives for public and private payers alike. Dissemination of research on the evidence base for clinical practice is an important strategy for persuading policymakers, providers, and patients that some things in medicine are not worth doing: they create too much risk, have unacceptable side effects, lack adequate effectiveness, and even, occasionally, that they are too costly for the benefits they confer. Evidence on the latter is derived from cost-effectiveness studies that allow comparison of the value for money of an intervention or program when compared to one or more alternative health-promoting interventions.

The use of cost effectiveness analysis (CEA) in U.S. policy-setting environments has a checkered history given perceptions that the methodology lacks transparency, that it is applied to some interventions and not others, and that it would be used to support health care rationing. Nonetheless, the annual publication of CEAs has increased more than 10 fold since the 1990s, the lion’s share of which have been studies of the cost-effectiveness of clinically delivered prevention and treatment services. Studies that focus on non-clinically directed health promoting interventions such as tobacco or sugar taxes, or placement of urban greenways are in short supply in spite of the fact that a robust literature has documented that health care’s impact on population health is dwarfed by social and environmental factors.

Expert Panels Seek to Improve and Standardize Economic Evaluations

Responding to lack of uniformity and accompanying difficulties in interpreting CEAs, the U.S. Department of Health and Human Services appointed in 1993 an expert panel to review methods and develop recommendations to improve the quality and standardization of CEAs intended to inform decisions about the efficient allocation of clinical and population health-directed resources. The report of the first U.S. Panel on Cost-effectiveness in Health and Medicine issued in 1996 has guided the field over the past two decades. Last fall, a Second Panel on Cost Effectiveness in Health and Medicine provided an update that reviews, and in certain instances revises, the first Panel’s recommendations, covering innovations in methods and expanding the discussion of areas such as evidence synthesis and ethical issues. Although the current report authoritatively provides a framework for cost effectiveness evaluation that should guide the field for many years, its detailed discussions focus largely on analyses of health care directed at individuals rather than at analyses that examine health care systems or broader determinants of health.

What are the costs, where are the savings? It depends on the perspective

The first panel recommended the “societal” perspective as the default perspective to take for a “reference case” analysis, a standard set of methods to serve as a point of comparison across CEAs that cross populations and/or illnesses. The societal perspective is the most comprehensive perspective, incorporating all costs and all effects regardless of who incurs them, and regardless of whether the costs and effects are within or outside of the health sector. The new report recommends the addition of a “health care” perspective for reference case analyses. On the cost side, the health care perspective limits the accounting of costs (and savings) to those reimbursed by third party payers or paid out-of-pocket by patients. It does not require charting costs or savings that occur in non-health care sectors. The current panel recommends use of an “impact inventory” that lists all consequences of an intervention, within and outside of the formal health care system. It is intended to capture the costs/savings and effects of a health intervention on sectors including social services, education, housing, and environment, but is silent on use of a health impact inventory for non-health sector interventions whose impact includes health.

Research needs in health and well-being

Recent estimates suggest that health care contributes a modest 15-20 percent to health outcomes, and the literature is full of accounts of communities living next door to one another with disparities in life expectancy ranging from 8-15 years, reinforcing the understanding that it is the social and physical environment people are exposed to that most influences health.. The new panel’s report is not unaware of health care’s opportunity costs and how it crowds out non-health sector investments. Investments, in housing, education, wages, safe environments, and communities have been shown to be cost efficient in addition to supporting the strength and vitality of individuals, communities, and the nation overall.

Chokshi and Farley compared the cost-effectiveness of environmental health interventions with individually oriented approaches to health. In their review of 400 prevention-directed CEAs contained in the Tufts Medical Center Cost-Effectiveness Analysis (CEA) Registry nearly half of the 31 environmentally focused, non-person directed preventive interventions (i.e. those whose target is an element of the environment rather than of the individual) were cost-saving, and the remainder had relatively modest costs per quality-adjusted life year when compared with clinically oriented or person-directed approaches. Economic analyses that look to non-health sector interventions are, however, few and far between. However, Neumann et al in reviewing CEAs directed at Healthy People 2020 objectives report a paucity of CEAs that look at broadly directed social and environmental determinants, or compare them to health care technologies.

Why are these types of CEAs in short supply? Given the large spend in the health sector, health services researchers and economists understandably gravitate to clinical interventions to investigate where greater efficiency and effectiveness can be found. And because so much money is invested in medical care and the research that underpins it, such studies are comparatively easy to undertake. Meanwhile, the evidence base is much less straightforward when it comes to analyzing the cost-effectiveness of interventions that target more broadly health-beneficial policies, those that through tax policies promote and inhibit health-affecting behaviors, those directed at built and natural environments, education, and other strategies that directly or indirectly affect people’s health. .

The methodologic challenges of conducting more broadly conceptualized prevention interventions are real; they include the complexity of interacting processes, multiple outcomes, the often extended timeframe to observe change, and the absence of funding, and therefore research, that provides a detailed evidence base. Investment in building the methods and evidence base for non-clinical approaches to health was a key recommendation of the Committee on Public Health Strategies to Improve Health. To date, we know of few additional resources that have been directed at supporting this work. Absent serious investment in evaluating broader-based interventions, we will continue to look under the lamppost for our keys.

Whither cost-effectiveness methods and research

Going forward, we should look for cost-effectiveness analyses that evaluate interventions whose benefits confer salutary effects that not only include, but go beyond what can be obtained by medical care alone. These studies will often originate from teams working outside the traditional health and health care sector, requiring trans-disciplinary efforts that health economists, methodologists and practitioners develop and engage in.

The field needs ongoing work in creating further guidance for implementing a societal perspective including a systematic way to incorporate the non-health related costs and benefits. While the current Panel’s impact inventory will allow a better understanding of what benefits and costs health-related interventions confer, researchers need additional direction on methods that incorporate these costs and benefits more consistently and systematically into the analysis.

Moreover, CEAs that take a societal perspective need to conceptualize and measure outcomes that represent a deeper understanding of what people actually seek in life. The quality-adjusted life year measure of outcome, endorsed by both Panels, defines health in a manner that is necessary, but insufficient for true well-being. More germane to the societal perspective is a summary measure that captures health the way the World Health Organization envisioned it; “as a state of physical, mental and social well-being.” And consideration then would need to be given as to how to value such a measure so that it appropriately frames the impact of interventions whose effects are on populations, rather than on individuals. Creating a measure of this nature, with a different approach to valuation will be challenging. But meaningful measures have uses outside of the scientific community; they can serve to pinpoint what a society thinks is important and assist the public and its policymakers in understanding what’s really going on so we can better allocate our resources.

As a nation we have yet to take serious stock of non-medical approaches to health. The dominant view of lay people and of policymakers alike is that health is determined primarily by individually directed medical services. What is less well understood by the health delivery system and its decision makers is that resources, when directed thoughtfully, can confer health gains that accompany other goods and services that people value, quite independent of their health impact.

The research community needs to evaluate and document the costs and effects of such multi-sectoral initiatives and raise the visibility of the different paths to a healthier nation. Unless this is accomplished, Americans’ understanding of value for money in health will continue to focus in the clinical arena, a sector of the economy whose powerful business interests are too frequently out of line with those of the public. With a fuller understanding of the forces that shape health and well-being, the nation will be better placed to make wiser use of public and private dollars.