Impact of a school-based intervention on access to healthcare for underserved youth.

PURPOSE: To determine whether a multidimensional school-based intervention, which included physical and mental health services, increased adolescents' use of needed medical care and preventive care and decreased emergency room use.
METHODS: A total of 2832 seventh- through twelfth-grade students in six public urban intervention schools and 2036 students in six demographically matched comparison schools completed a previously validated survey regarding health status and healthcare utilization in spring 1998 and 1999. Bivariate analyses examined the association between intervention status and Year 1/Year 2 outcomes. The multifaceted intervention included programs such as anger management groups, substance abuse prevention, tutoring, home visits, and enhanced school health services. Stepwise multivariate logistic models tested differences between the intervention and comparison groups across years, controlling for potential confounding variables [gender, age, race/ethnicity, maternal education, grade in school, school district (city or county), health status, and chronic health problems]. The interaction term for Group x Year was used to test the effect of the intervention. Multivariable modeling was also used to determine student factors independently associated with healthcare utilization.
RESULTS: Respondents had a median age of 15 years, 56% were female, 51% were white, 42% were black, and 34% reported chronic health problems. In both years, over 45% of students in both groups reported not seeking medical care they believed they needed. The proportion with missed care in the intervention schools did not change, whereas the proportion with missed care in the comparison schools increased. Emergency room use decreased slightly in the intervention schools and increased slightly in the comparison schools between Year 1 and Year 2. There were no major changes in healthcare delivery in this area during the year, demonstrating the volatility of adolescents' perceived access to care. Among the student factors, health status, having a chronic condition, and being in a higher grade were independently associated with students' report of not seeking care they believed they needed.
CONCLUSION: These results confirm that many adolescents have unmet healthcare needs. Those with poor health status are most likely to report underutilization and unmet needs. These findings underscore the need for comparison groups when evaluating interventions and suggest the need for better understanding of community level changes in perceived healthcare access and use.