Two controlled studies were undertaken to evaluate the effectiveness of comprehensive and continuous pediatric care. In the first study, 220 infants of primiparous adolescents were randomly allocated to either a comprehensive care (CC) or traditional care (TC) group. CC infants received all medical care, preventive and therapeutic, in a hospital-based program staffed by a pediatrician, public health nurse, and social worker. Mothers of TC infants were left to obtain care from emergency rooms, well-baby and outpatient clinics. One year after delivery, each mother was interviewed and her infant's medical records abstracted. No differences were found between CC and TC infants in completeness of immunization, utilization of medical resources, morbidity, or mortality. In a second study, it was hypothesized that patient compliance with physicians' recommendations would be favorably influenced by continuous care. Seventy-seven children on daily oral penicillin prophylaxis for history of rheumatic fever (RF) were studied. For one year, compliance was determined by periodic urine tests for penicillin. Patients were then stratified for age, sex, and compliance and randomly allocated to continuous care (CC) or traditional care (TC) groups. CC patients received all medical care, even for problems unrelated to RF, from the same two physicians for 15 months. TC patients continued to receive specialty clinic care, were seen at the RF clinic by different physicians and referred elsewhere for all problems unrelated to RF. Urine specimens were tested periodically for penicillin. After 15 months, no differences were observed between CC and TC groups in proportion of noncompliers or in internal shifts in compliance which had occurred during the study. Thus, neither of these studies was able to demonstrate that comprehensive and continuous care was more effective than conventional ambulatory care for children.