Effectiveness and feasibility of three types of parent reminders to increase adolescent human papillomavirus (HPV) vaccination

Human papillomavirus (HPV) vaccines are safe and effective and prevent HPV types responsible for nearly all cervical cancers, most anal cancers, and a substantial proportion of other anogenital cancers. Routine vaccination is recommended for girls and boys aged 11–12 years in the U.S., yet as of 2019 only 67% of adolescents initiated and 45% completed the series by age 13. Parent reminders are low-cost and scalable, but HPV vaccination faces unique challenges, including multi-dose schedules and parental hesitancy, especially in resource-limited settings such as Federally Qualified Health Centers (FQHCs). This study aimed to evaluate the effectiveness and feasibility of three types of parent reminders (mailed letters, robocalls, text messages) for increasing receipt of the next needed HPV vaccine dose among 12-year-olds in a large multi-site FQHC. The authors hypothesized higher next-dose completion among those receiving reminders versus usual care and that text messages would be the most feasible reminder type.

Prior research on parent reminder/recall (letters, phone calls, text and email messages) has generally shown modest improvements in adolescent HPV vaccination. Many studies have used non-randomized designs and were conducted in public health agencies, managed care settings, or private practices, with relatively few focused on low-resource settings or on comparative effectiveness and implementation of multiple reminder types. Some randomized studies reported small to moderate effects on initiation or series completion, with stronger effects when multiple follow-up messages were used or when parents could choose their preferred reminder modality. Evidence for robocalls improving HPV vaccination has been mixed, despite prior success for other childhood immunizations.

Design and setting: Quasi-experimental study within a large multi-site Federally Qualified Health Center (FQHC) in Los Angeles County. Six clinics were matched into three pairs on baseline HPV vaccination rates, adolescent population size, and location. Within each pair, one clinic was randomly assigned to intervention and the other to usual care. Intervention clinics were randomly assigned to one of three reminder types: mailed letters, automated telephone calls (robocalls), or text messages. Timeframe: Intervention implemented November–December 2016; follow-up at 4 months. Population: Caregivers/parents of 12-year-old adolescents due for an HPV vaccine dose (either no prior doses or exactly one prior dose). Eligible patients were identified via electronic medical record (EMR) and population health management systems. Outcome: Receipt of the next needed HPV vaccine dose (dose 1 among unvaccinated or dose 2 among those with one prior dose) within 4 months. Interventions: Each reminder type used standardized scripts adapted from existing clinical messages; text messages were briefer due to character limits. Messages were delivered in English or Spanish according to patient preference. The same reminder modality could be sent up to two times: at baseline and approximately one month later if no dose was received or no appointment scheduled. Delivery systems: Mailed letters and robocalls were generated by the FQHC’s centralized call center; text messages were sent by a third-party vendor. Usual care: No parent-targeted HPV vaccine initiatives or reminders; routine vaccine administration during provider visits, with CDC Vaccine Information Sheet provided and timing information given for subsequent doses. Implementation/feasibility measure: A reminder was considered successfully delivered if there was no bounce-back (text), no returned mail (letter), or the robocall was completed or a voicemail could be left. Undelivered scenarios included text bounce-backs, unanswered/uncompleted robocalls, or returned mail. Data and covariates: Demographics (sex, race/ethnicity, insurance) abstracted from EMR. Statistical analysis: Vaccination rates presented as proportions. Logistic regression models estimated intervention effects overall and stratified by reminder type, sex, and dose number. Models (except those stratified by reminder type) included health center as a covariate; models (except those stratified by dose number) included dose number as a covariate. Software: Stata/SE 15.1. Sample size: 877 adolescents (approximately 47% female; >85% Latino; 61% Medi-Cal insured). IRB: Approved by UCLA IRB and the FQHC’s Research Committee.

- Overall effectiveness: 23% of adolescents in any reminder condition received the next HPV dose vs 12% in usual care at 4 months (11 percentage-point advantage; p = 0.046). - By reminder type: Text message reminders showed a significant effect (11 percentage-point advantage; p = 0.036). Mailed letters and robocalls did not show statistically significant effects in stratified analyses. - By sex: Significant effect among boys (15 percentage-point advantage; p = 0.006); not significant among girls (6 percentage-point advantage; p = 0.829). - Feasibility (delivery success): - Text messages: First reminder 83% delivered (116/139); second reminder 92% (107/116). Second reminders overall 92% delivered (107/117); 2% opt-out rate. - Robocalls: First reminder 67% delivered (80/120); second reminder 78% (79/101); lowest delivery success among modalities. - Mailed letters: Second reminders 92% delivered (118/128). - Population characteristics: 877 adolescents; ~47% female; >85% Latino; 61% insured by Medi-Cal.

Delivering parent reminders increased HPV next-dose receipt among adolescents in an FQHC setting, with the overall effect largely driven by text message reminders and by improvements among boys. Text messages were both effective and feasible, achieving high delivery success and low opt-out rates, and integrating well within existing clinic systems. Robocalls were least feasible, likely due to high rates of unanswered or blocked calls and parent preferences for concise, retrievable text communications. The stronger effect among boys suggests that a simple cue to action may be particularly impactful where baseline awareness or provider recommendation is lower, whereas additional barriers may exist for parents of girls. Implementing reminders at the clinic level, rather than individually, provided insights into real-world feasibility and sustainability by leveraging existing centralized infrastructure for outreach.

Parent reminders, particularly text messages, are a feasible and effective strategy to improve adolescent HPV vaccination in resource-limited clinical settings. Future research should confirm these findings in larger samples, evaluate costs, optimize message frequency and modality preferences, and compare reminders with other clinic-based strategies to promote vaccination.

Findings derive from a single multi-site FQHC serving a mostly Latino population, which may limit generalizability. The intervention targeted parents of 12-year-olds; results may not extend to older adolescents or adults eligible for catch-up vaccination. Sample size limited power for direct comparisons across reminder types. The feasibility measure (successful delivery) was crude and may overestimate true receipt if undelivered notices were delayed or not recorded; it did not confirm whether messages were read or heard. Implementation burden perceptions (e.g., mailed letters being more resource-intensive than texts) and parent receptivity to robocalls may also affect scalability.