TY - JOUR
T1 - Examining the Effects of an Adolescent Pregnancy Prevention Program by Risk Profiles
T2 - A More Nuanced Approach to Program Evaluation
AU - Vasilenko, Sara A.
AU - Glassman, Jill R.
AU - Kugler, Kari C.
AU - Peskin, Melissa F.
AU - Shegog, Ross
AU - Markham, Christine M.
AU - Emery, Susan Tortolero
AU - Coyle, Karin K.
N1 - Publisher Copyright:
© 2018 Society for Adolescent Health and Medicine
PY - 2019/6
Y1 - 2019/6
N2 - Purpose: The objective of the study was to examine whether latent class analysis (LCA) could (1) identify distinct subgroups of youth characterized by multiple risk and protective factors for early sexual initiation and (2) allow for a more nuanced assessment of the effects of a middle school program to prevent teen pregnancy/HIV/sexually transmitted infection. Methods: LCA was applied to data from the baseline (seventh grade) sample of 1,693 sexually inexperienced students participating in a randomized controlled trial of It's Your Game…Keep It Real in Harris County, Texas. Multilevel analysis was applied within subgroups defined by the latent classes to assess for potential differential program effects. Results: LCA identified 3 distinct profiles of youth: family disruption, other language household, and frequent religious attendance. Multilevel analyses found differential effects of the program across these profiles with a significant and substantial reduction (30%) in initiation of vaginal sex by ninth grade for students in the family disruption profile only. Conclusions: Application of LCA may hold promise for conducting more nuanced evaluations and refinements of behavior change interventions for youth.
AB - Purpose: The objective of the study was to examine whether latent class analysis (LCA) could (1) identify distinct subgroups of youth characterized by multiple risk and protective factors for early sexual initiation and (2) allow for a more nuanced assessment of the effects of a middle school program to prevent teen pregnancy/HIV/sexually transmitted infection. Methods: LCA was applied to data from the baseline (seventh grade) sample of 1,693 sexually inexperienced students participating in a randomized controlled trial of It's Your Game…Keep It Real in Harris County, Texas. Multilevel analysis was applied within subgroups defined by the latent classes to assess for potential differential program effects. Results: LCA identified 3 distinct profiles of youth: family disruption, other language household, and frequent religious attendance. Multilevel analyses found differential effects of the program across these profiles with a significant and substantial reduction (30%) in initiation of vaginal sex by ninth grade for students in the family disruption profile only. Conclusions: Application of LCA may hold promise for conducting more nuanced evaluations and refinements of behavior change interventions for youth.
UR - http://www.scopus.com/inward/record.url?scp=85062331740&partnerID=8YFLogxK
UR - http://www.scopus.com/inward/citedby.url?scp=85062331740&partnerID=8YFLogxK
U2 - 10.1016/j.jadohealth.2018.12.003
DO - 10.1016/j.jadohealth.2018.12.003
M3 - Article
C2 - 30850310
AN - SCOPUS:85062331740
SN - 1054-139X
VL - 64
SP - 732
EP - 736
JO - Journal of Adolescent Health
JF - Journal of Adolescent Health
IS - 6
ER -