Despite these issues, using administrative claims
such as Medicaid claims provides a place to start in assessing
teenage contraception utilization and compliance.
An important follow-up project would be to
compare fee-for-service and managed care Medicaid
data, looking at urban and rural differences. Examining
other health plan and private insurance data that includes
patients from higher socioeconomic strata is important
in order to validate the methodology for more
diverse groups, looking specifically at teens who become
pregnant during the study interval to see if there
is a frequency pattern and then target an intervention to
the critical times for compliance reinforcement. In addition,
it would be useful to evaluate contraceptive use
prospectively, examining a cohort of adolescents at
risk for pregnancy and performing a survival analysis.