Objective: To determine the frequency of health care utilization for infectious illnesses in a pediatric practice and factors that affect utilization. Study Population: Children aged 6 weeks to 5 years who received health care at a large pediatric practice in Norfolk and Chesapeake, VA enrolled between 3/96 and 10/97 into an ongoing 3-year surveillance study. Methods: Attempts were made to enroll all age-eligible children in the practice. Demographic and medical history data were collected at enrollment and updated monthly. Office visit records were reviewed 2 weeks prior to enrollment for all children and when children were sick. Associations between parental education level and type of health care coverage on frequency of visits for illness were analyzed using Poisson regression analysis. Results: This analysis includes 506 (9%) of age-eligible children in the practice followed for 6618 child months (range 1-21 mos/child). 414 (82%) children had 1861 illness visits (mean 3.4/child-year) that resulted in 2526 diagnoses including 1615 (64%) that involved the upper respiratory tract. Otitis media (OM) was the most frequent primary diagnosis (21%) followed by uncomplicated upper respiratory infections (URI) (15%), sinusitis (4%), and Group A Streptococcus (GAS) pharyngitis (1%). 23% of diagnoses were combinations of the above. The mean number of visits for OM (1.0/child-year) was greater than for the other diagnoses (URI.6/child-year, sinusitis.1/child-year, GAS pharyngitis.06/child-year). 73% of parents had more than a high school education. 98% of families were covered by third party health insurance including: 82% by employer-administered health insurance(private insurance, HMO, PPO), Medicaid 9%, and CHAMPUS 7%. Children of parents with more than a high school education had more sick visits than parents with a high school education or less (Rate ratio=1.18, p=.002). Families with no insurance visited the physician the least often (Rate ratio=0.43, p=.008), followed by CHAMPUS (Rate ratio=.80, p=.03), private insurance, HMO, and PPO (reference group), and Medicaid (Rate ratio=1.13, p=.12). When level of education and type of health care coverage were adjusted, each influenced health care utilization (p<.03). Conclusions: Health care utilization in this pediatric practice was associated with type of health care coverage and parents' level of education. These findings may help target high-use and low-use groups of parents to improve appropriate health care utilization.