The answers to four questions one must ask in planning to meet the health needs of any group of children define some of the most important research issues in child health. 1. What are the functionally important health problems to be found with some frequency in a group of children? Many are well defined and easy to count, and for some of these we have relatively good counts. While we know that the prevalence of many health problems is related to socioeconomic status, we know practically nothing about the mechanisms by which this relationship is mediated. There are certain health findings—for example, anemia, poor dietary history, and certain deviations of behavior and speech—that we are reluctant to label as health problems until we have much more evidence about their actual functional consequences. There are certain health problems, especially the behavior and learning problems of school-aged children, that we would like to be able to define in terms of findings at a much earlier age. 2. What techniques will efficiently identify those children who have functionally important health problems? We have a handful of effective and efficient screening tests, as well as several that are widely used but need much further definition in terms of reliability and validity. The series of tests and questionnaire items strung together in a physician's history and physical examination certainly falls into the category of tests whose reliability and validity needs vastly more study. All of the descriptive and predictive tests of behavior and learning, as well as those of nutrition and speech, need much further validation before they can be recommended for routine use. 3. What treatment or intervention techniques will be most effective in remedying these problems? Because this is the realm of traditional medical research, we know a great deal about many of the specific health problems which are to be found in children. We are, however, rarely able to critically weigh costs and benefits of one form of treatment against costs and benefits of another form of treatment or of no treatment at all. Many of the data we will need to make such logical decisions will come from studies of the natural history of illness and from double blind studies of various forms of intervention. A continuing problem is the perpetuation of ineffective intervention techniques—bed rest, tonsillectomy, much psychotherapy—because of the humanistic urge to "do something to help," even when we do not know that what we do actually helps. 4. What resources—financial, manpower, administrative, organizational—will be necessary to prevent, identify, and remedy these problems in a population of children? Given current techniques and organizations, we seem to require one children's physician for every 1,000 families with children and between $100 and $200 a year for each child. The opportunity for reallocation of tasks between the doctor and his helpers and for new organizational and financial settings is enormous. The tools to measure the effectiveness and efficiency of such changes are weak and need much greater development. We do know that use of whatever services are available can be greatly enhanced by making these services responsive to the real needs of the recipients or clients. With so many gaps in basic knowledge, it is hardly surprising that methods to best achieve better health and function for young children are criticized and debated. But, gaps in knowledge and lack of organizational models of proven usefulness do not preclude pragmatic decisions about the content and organization of programs to meet the health needs of pre-school children. Such imperfect knowledge does, however, dictate that practical decisions must be tentative, and that diversity of program content and organization is highly desirable, both in adapting to local conditions and in testing and proving new methods. It also dictates that each of the many diverse patterns and programs which develop must build into itself evaluation and monitoring systems leading both to program improvement and to more definite knowledge about effectiveness of treatment techniques and organizational plans. Perhaps the greatest research need is for tools and motivational arrangements that will assure that every practitioner of child health and every organization involved in the promotion of child health can and does fully evaluate his own results in terms which describe the real issues and modifies his programs in terms of this evaluation. John Gardner20 has described the seff-renewing individual or institution as one who is constantly aware of his actual problems and operating results and is constantly developing new resources to deal with the ever-changing situation. Perhaps the Gardner concept of self-renewal is what we need most, both in providing today's services and in defining tomorrow's research issues in child health.