Effect of a School-Based Educational Intervention About the Human Papillomavirus Vaccine on Psychosocial Outcomes Among Adolescents: Analysis of Secondary Outcomes of a Cluster Randomized Trial

Abstract

Importance: Delivery of vaccination to adolescents via a school-based program provides an opportunity to promote their involvement in health decision-making, service provision, and self-efficacy (belief in one's ability to perform a certain behavior).

Objective: To examine the effect of a human papillomavirus (HPV) vaccination education and logistical intervention on adolescent psychosocial outcomes.

Design, setting, and participants: In this cluster randomized trial and process and qualitative evaluation, adolescents aged 12 to 13 years (first year of high school) were recruited at high schools in Western Australia (WA) and South Australia (SA) in 2013 and 2014. Statistical analysis was performed from January 2016 to December 2020.

Interventions: The complex intervention consisted of an adolescent intervention to promote knowledge and psychosocial outcomes, shared decisional support tool, and logistical strategies.

Main outcomes and measures: Prespecified secondary outcomes were assessed. The HPV Adolescent Vaccination Intervention Questionnaire (HAVIQ) was used to measure changes in adolescent knowledge (6-item subscale), fear and anxiety (6-item subscale), self-efficacy (5-item subscale), and decision-making (8-item subscale). The hypothesis was that the intervention would improve adolescent involvement in vaccine decision-making (measured before dose 1 only), improve vaccine-related self-efficacy, and reduce vaccine-related fear and anxiety (measured before doses 1, 2, and 3). Mean (SD) scores for each subscale were compared between intervention and control students. In the process evaluation, focus groups were conducted. Analyses of the HAVIQ data were conducted from 2016 to 2020. Qualitative analyses of the focus groups were undertaken from 2017 to 2020.

Results: The trial included 40 schools (21 intervention and 19 control) across sectors with 6967 adolescents (mean [SD] age, 13.70 [0.45] years). There were 3805 students (1689 girls and 2116 boys) in the intervention group and 3162 students (1471 girls and 1691 boys) in the control group. The overall response rate for the HAVIQ was 55%. In WA, where parental consent was required, the response rate was 35% (1676 of 4751 students); in SA, where parental consent was not required, it was 97% (2166 of 2216 students). The mean (SD) score for decision-making in the intervention group before dose 1 was 3.50 (0.42) of 5 points and 3.40 (0.40) in the control group, a small but significant difference of 0.11 point (95% CI, 0.06 to 0.16 point; P < .001). There was a small difference in favor of the intervention group in reduced vaccination-related anxiety (pre-dose 1 difference, -0.11 point [95% CI, -0.19 to -0.02 point]; pre-dose 2 difference, -0.18 point [95% CI, -0.26 to -0.10 point]; pre-dose 3 difference, -0.18 [95% CI, -0.24 to -0.11]) and increased vaccination self-efficacy (pre-dose 1 difference, 4.0 points; [95% CI, 1.0 to 7.0 points]; pre-dose 2 difference, 4.0 points [95% CI, 2.0 to 6.0 points]; pre-dose 3 difference, 3.0 points [95% CI, 1.0 to 5.0 points]). Focus group data from 111 adolescents in 6 intervention and 5 control schools revealed more confidence and less anxiety with each vaccine dose.

Conclusions and relevance: In this cluster randomized trial, there was a small difference in adolescent decisional involvement and vaccine-related confidence and reduced vaccination-related fear and anxiety that was maintained throughout the vaccine course in the intervention vs control groups. Guidelines for vaccination at school should incorporate advice regarding how this outcome can be achieved.