COLIN McCoRD* INTEGRATION OF HEALTH, NUTRITION, AND FAMILY PLANNING: THE COMPANIGANJ PROJECT IN BANGLADESHt

It is more than 50 years since John Grant and Andrija Stampar first proposed regionalization of health services around rural health centers, but it cannot be said that the widespread application of these concepts in developing countries has been very successful (3 ,4, 7). The causes of this failure are complex, but it has not been due to lack of trying. There have been numerous pilot projects, but many have asked whether it is possible to draw any conclusions valid for general application from demonstrations which have available more special expertise, money, logistical support, and other kinds of attention than are likely to be available ordinarily. Certainly it is unlikely that demonstration projects will produce fully developed models which can be replicated generally throughout a country or even a district. Nevertheless, there are at least three ways in which a demonstration project should be able to contribute to design of health centers: I. The identification of program failures and the forces contributing to failure. If a program or a program component does not work despite all the special inputs, this in itself is useful information. It becomes more useful if it provides an opportunity to identify the forces that caused the failure. 2. The analysis of program components to identify successful and costeffective ones which might be transplanted elsewhere. A demonstration gives