King Drew Medical Center is a County hospital in South Central Los Angeles, serving primarily inner city residents. After discharge from our NICU all infants are offered services through the High Risk Clinic (HRV), including well child care, developmental follow-up, referral to state agencies, and interventions by subspecialty clinics (ophthalmology, ENT, neurology, cardiology, orthopedics, surgery). An affiliated clinic (Hub) provides specific services directed to foster families. Coverage for the costs is provided by the State through Medi-Cal, CHDP, California Children Services, and other agencies. Return appointments and reminder phone calls are provided. Nonetheless, the rate of return to the clinic for follow-up is less than 50%. For very low birth weight infants (VLBW, BW ≤ 1500 grams) only 34/74 infants born in 1994 were regularly seen in 1997. When discharge diagnosis is considered, rate and modality of health care access vary. For chronic lung disease, loss of follow-up in HRC corresponded to increased use of emergency services, while patients with severe lung disease and high attendance in the HRC and Allergy clinic had lower rates of access to emergency care. For infants with neurological impairments, severity correlated with close follow up in HRC. Comparing NICU graduates of 1988 vs 1994, we also noticed that the number of admissions for VLBW babies with maternal substance abuse increased from 27% to 36%. Upon discharge, many of these babies are routed through the foster care system of the Greater Los Angeles area, as far as over 50 miles from the hospital of birth. Phone interviews showed that distance was the primary reason for not returning to the HRC. A significant number of Hispanic patients were lost to follow up because of family relocation. No further data were available for most of these patients. Conclusions: we investigated what factors appear to affect return to follow up in a specialized, comprehensive, no or low cost clinical setting. Presence of complications of prematurity was not sufficient to warrant consistent follow up under the described conditions. Population variables appear to affect return to a considerable degree. We believe that this situationm may be common to many County institutions, and carries significant implications for the welfare of a high risk infant population.