Eur J Nucl. Med. 5, 271-276 (1980) EuropeanN u c l e a r Journal of Medicine © by Sprmger-Verlag 1980 Cost Effectiveness Calculations for the Diagnosis and Treatment of Tuberculous Meningitis Barbara J. McNeil, Mark Thompson, and S. James Adelsteln Department of Radiology, Harvard Medical School, the Joint Program m Nuclear Me&cane, Peter Bent Brigham Hospttal, and Center for the Analysis of Health Practices, Harvard School of Public Health, Boston, Massachusetts, USA Abstract. This study illustrates how cost-effectiveness calculations provide help in decisions involving a choice between introduction of a new diagnostic procedure or a new therapy for a particular clinical problem. This type of problem is critical for areas where financial resources are limiting. Our analysis is centered on the value of diagnosis and treatment in tuberculous meningitis (Tbm) and, because of its importance to developing countries, our epidemio- logic data were derived from India. When financial costs are ignored, the introduction of second line ther- apy (e.g., Rifampin) leads to more cures than does the introduction of even a perfect diagnostic test. However, diagnostic tests (e.g., the Bromide partition test or possibly radioassays) markedly improve case finding and to some extent increase cure rates. All affects vary markedly with the prevalence of tubercu- lous meningitis in the population under study. For example, net financial savings would occur were a perfect nuclear test available and conventional ther- apy used at a prevalence of Tbm of 30% whereas there would be a net loss if the prevalence rose to 80%. This study underscores the need for detailed studies on the radiobromide partition test and for the development of new diagnostic tests, perhaps a radioimmunoassay of either the acid fast bacillus or of an antibody to it. This work was supported in part by a contract from the Interna- tional Atomic Energy Agency. Dr. McNml is a recipient of Re- search Career Development Award 5-K04-GM00194 from the Na- tional InsUtute of General Medical Sciences Offprint requests to. Barbara J. McNeil, M.D., Department of Radiology, Harvard Medical School, 25 Shattuck Street, Boston, Massachusetts 02115, USA Introduction Over the past few decades, many new diagnostic tests and new therapeutic regimens have been introduced for a variety of diseases. Most cost-effectiveness calcu- lations of these tests and therapies have been restricted to an analysis of the impact of the test or of the therapy. They have not discussed a common problem existing in many countries where either total financial resources or the proportion of resources devoted to health care are limiting. In such situations, if a new diagnostic procedure and a new therapy exist for the same problem, a difficult choice may have to be made between them or among various combinations of them in competing for available funds. In this study we present a prototypical example of an approach to this problem of tradeoffs at the hands of a problem common to many developing countries the diagnosis and treatment of tuberculous meningitis (Tbm). This issue is particularly challeng- ing for two reasons. (1) The early clinical diagnosis of this disease is exceedingly difficult. As a supple- ment to the clinical diagnosis, the radiobromide parti- tion test already exists and many individuals hope that radioimmunoassay studies of the cerebrospinal fluid (CSF) will be even more helpful. (2) Current treatment is, in general, long, costly, and frequently ineffective. Although improved treatment regimens have recently been available in developed countries, their use in developing countries has been limited because of high financial costs. The setting of this analysis will involve the diag- nosis and treatment of Tbm in India. The first part involves determination of the benefits and costs asso- ciated with the use of an existing test (the radiobro- mide partition test) and of a theoretically perfect test 0340-6997/80/0005/0271/$01.20