Cost-Outcome Methods for Mental Health

If indeed politics makes strange bedfellows, so too does the co-evolution of policy environments and research disciplines generate novel interdisciplinary collaborations. A generation ago, most economists and psychologists likely would have been hard put to identify areas of common interest. The notion that one could measure the outcomes of mental health treatments with enough precision for cost-effectiveness analysis likely would have found few subscribers among economists; likewise, many psychologists would have been repelled by the idea that psychotherapy could or should be evaluated in terms of its costs and outcomes.

But growing interest in the psychometrics of outcome measurement has led to the development of well-validated and reliable tools for assessing the effects of various treatment modalities. At the same time, health economists have begun grappling with the complex issues of public and private financing of mental health care, spawning a new area of research interest and giving rise to a generation of economists who focus on mental health.

As these trends within economics and psychology were gaining momentum, an emphasis on cost-effectiveness, not altogether welcomed by many providers, began to emerge in mental health policy. What some providers considered a shotgun marriage of psychology and economics in the emerging policy environment created a new and interesting field of intellectual endeavor that will become a staple of the evaluation of mental health service systems.

From this point on, public and private insurers' demands that providers demonstrate cost-effectiveness require that clinicians, managers, and policy makers become conversant with the melded concepts and lexicons of this new evaluative perspective. Until now this familiarization process has been hampered by the lack of a literature that brings these concepts together in a cogent manner accessible to all the relevant disciplines and actors. In Cost-Outcome Methods for Mental Health, Hargreaves,Shumway,Hu, and Cuffel, a multidisciplinary group whose members have been among the pioneers in mental health cost and outcome research, make a solid contribution toward filling this gap.

The authors say that their book is designed as a graduate or postdoctoral text, but it appears that they also envisioned several other distinct audiences. One would be decision makers with no training in evaluation research, outcome measurement, or cost-effectiveness analysis. Another audience would consist of individuals with knowledge of psychology and outcome measurement but little background in economics, and vice versa. Yet another would be individuals with some background in both areas who nevertheless could benefit from examples of how they merge. A final group would be more sophisticated researchers looking for references to research studies and information on various forms of outcome measurement.

The needs of all of these groups are well met in this book. Beginning with an overview of cost-outcome research in mental health, the authors provide a synopsis of standard evaluative research designs and issues. Following are several useful chapters describing relevant economic concepts and problems in measuring service utilization and costs. Next, the "outcome" chapters offer an overview of basic outcome measurement concepts, methodological issues in using them, and appropriate measures for various outcome domains such as health status, quality of life, and public safety.

The final two chapters bring together economic and outcome measurement concepts. The first of the two discusses the statistical estimation of cost-effectiveness, while the last chapter outlines how cost-outcome data can be used to guide policy. A particularly attractive feature of the book is the presentation at the beginning of each chapter of real-world scenarios illustrating the issues to be addressed.

Cost-outcome research is a complex discipline. Reading this volume will not by itself allow one to embark on such research without additional training. It will, however, permit the uninitiated to participate in these efforts and to converse intelligently with researchers and evaluators engaged in such work. For others, it will provide a valuable reference guide to concepts, tools, and methodologies currently used in this emerging field.

Cost-Outcome Methods for Mental Health is a timely textbook that helps fill a yawning gap in mental health services research—a field that has produced so few published cost-outcomes studies that the authors resort to examples from medicine rather than mental health. Recognizing the elementary phase of development of the field, the authors wisely point out that their goal is to help readers learn a new vocabulary and become intelligent consumers of cost-outcome research findings. Although the book surely will expand the knowledge base of experienced mental health services researchers as well, the authors acknowledge that planning and carrying out cost-outcome studies is likely to need the leadership of a health economist.

Besides chapters on measuring outcomes, determining service use, and estimating per-unit costs, Hargreaves and his colleagues provide very useful chapters on aspects of cost-outcome research that are often overlooked: measuring the fidelity of the "experimental" program, determining consumer preferences, and integrating findings from multiple studies of the same program. The section on consumer preferences includes an important discussion of quality-adjusted life years. The authors conclude that this approach, which has considerable appeal and which has been successfully used in medical cost-outcome studies, poses problems for mental health services researchers. They argue that the course and outcome of many mental illnesses are difficult to map onto the 0-to-1 scale required by this approach.

Despite its many virtues, the book falls short in several spots. The outcome measurement chapter is especially disappointing because of its narrow focus on clinician-perspective instruments and lack of discussion about the value of consumer-reported outcomes. Some additional information would be required if the book were to be used as a textbook for a graduate or postdoctoral course in mental health cost-outcome research, as the authors intend. For example, it would be helpful if the authors had included at least one complete study, a "worked example," to guide the reader through the details and illustrate just how research decisions are made in a real-world context. A glossary might also have been useful, given the expected readership. Finally, the authors often describe the nature of difficult decisions between two less-than-perfect solutions to methodological challenges, but they seldom make recommendations about the preferred choice.