In order to adequately care for patients with chronic disorders and to properly allocate resources, the epidemiology of the underlying disease must be know. Proper population based studies involve substantial planning and educational programs, however. To prepare for such a study of pediatric rheumatic disorders, we performed a referral-based pilot study. During an eleven-month period pediatricians all over Costa Rica were asked to refer to us all new cases of possible rheumatic disorders among children less than 16 years of age, using the EULAR criteria for juvenile chronic arthritis. The children were evaluated at the National Children's Hospital. An annual incidence of 5.9 cases of all types of pediatric rheumatic diseases per 100,000 was found. Incidences of 5.4 per 100,000 children below 16 years of age and 6.1 for children below 12 years of age were observed for juvenile chronic arthritis (JCA). 77% of the JCA cases were of pauciarticular onset, and 23% were of polyarticular onset. No cases of systemic JCA were diagnosed. The female to male ratio was 3:2. Antinuclear antibodies were positive in 13% of the JCA cases, and IgM rheumatoid factor was found in 15% of the children. Chronic iritis was noted in only 2 cases; both were girls and both were ANA positive. The incidence found was low compared to population-based studies, but in the same range as hospital-based investigations.