Hypertension--high blood pressure--is a largely asymptomatic disorder which affects between 10% and 30% of the American population. It is associated with increased risk of severe illness (kidney disease, congestive heart failure, stroke) and death. Acknowledging these facts, the authors--a Harvard public health analyst and cardiologist--and several colleagues have elaborated a mathematical model measuring the cost-effectiveness of various ""interventions"" in the hypertension problem. Their approach is derived from operations research or game theory; it makes use of probabilities, expected values, estimated costs of screening programs, of referrals, treatment regimens, etc., and draws heavily upon preexisting studies. All the same they present a convincing case that funds can be allocated with greater wisdom. They demonstrate that there is greater cost-effectiveness in screening a smaller population and then following through at later stages. This means taking steps to assure that hypertension patients understand their condition and the need to take daily medication. Their data also reveal greater cost benefits for certain age groups--younger men and older women, for example. The equations make the book heavy-going for the general reader, although the authors are reasonably good at explaining their assumptions and logic. The danger is that this sort of approach tends to depersonalize. You or I with borderline hypertension may get washed out in the formula as just not as good a cost-effective bet. Nevertheless this is a pioneering effort to show how to maximize the benefits from a limited amount of health funds, and as such it deserves to be read by all concerned with improving the delivery of health care.