A cost-benefit analysis should be seen not as a mechanism for deciding mechanically on the allocation of funds and resources among programs but as a structure for weighing advantages and disadvantages (that is, for organizing knowledge). Considering all the forms of benefits and costs that we were able to derive in monetary terms, the experimental program provided both additional benefits and additional costs as compared with the conventional treatment. However, the added benefits, some +1,200 per patient per year, are nearly +400 more per patient per year than the added costs. A number of the forms of benefits and costs that we have measured in quantitative but nonmonetary terms show additional advantages of the community-based experimental program. The generalizability of a single experiment is limited, but the methodologies developed may be useful if their proper role is appreciated.