Skip to main content
NCBI home page
NIHPA Author Manuscripts logoLink to NIHPA Author Manuscripts
. Author manuscript; available in PMC: 2022 Dec 1.
Published in final edited form as: Health Psychol. 2021 Jun 17;40(12):887–896. doi: 10.1037/hea0001075

Mechanisms of Self-Persuasion Intervention for HPV Vaccination: Testing Memory and Autonomous Motivation

Austin S Baldwin 1,3, Hong Zhu 2,3, Catherine Rochefort 1, Emily Marks 2, Hannah M Fullington 2, Serena A Rodriguez 2,3, Sentayehu Kassa 4, Jasmin A Tiro 2,3
PMCID: PMC8678358  NIHMSID: NIHMS1727590  PMID: 34138615

Abstract

Objective:

Optimizing a self-persuasion intervention app for adolescent HPV vaccination requires investigating its hypothesized mechanisms. Guided by the experimental medicine approach, we tested whether (a) self-persuasion intervention components (verbalize vaccination reasons, choose HPV topics) changed putative mechanisms (memory, autonomous motivation) and (b) measures of the putative mechanisms were associated with HPV vaccination.

Methods:

These are secondary analyses from a randomized 2 (cognitive processing: verbalize reasons vs. listen) × 2 (choice: choose HPV topics vs. assigned) factorial trial (Tiro et al., 2016). Undecided parents (N = 161) with an unvaccinated child (11–17 years old) used the self-persuasion app, recalled reasons for vaccination (memory measure), and completed an autonomous motivation measure. Adolescent vaccination status was extracted from electronic medical records 12 months post-intervention.

Results:

The verbalize component resulted in greater recall accuracy of vaccination reasons (p < .001); however, the choose topics component did not increase autonomous motivation scores (p = .74). For associations with HPV vaccination, recall accuracy was not associated (ps > .51), but autonomous motivation scores significantly predicted vaccination (ps < .03), except when controlling for baseline motivation (p = .22).

Conclusion:

The intervention app engages parents in reasons for vaccination; however, memory may not be a viable mechanism of vaccination. Although the intervention did not affect autonomous motivation, associations with vaccination status suggest it is a viable intervention target for HPV vaccination but alternative strategies to change it are needed. Future testing of a refined app should examine implementation strategies to optimize delivery in clinical or community settings.

Keywords: self-persuasion, HPV vaccine, experimental medicine, memory, autonomous motivation

The current human papillomavirus (HPV) vaccine prevents infection from seven high-risk genotypes of HPV that lead to cervical and five other cancers. In the U.S., the 2-dose vaccine series is recommended for 11–12 year olds with catch-up vaccination recommended up to age 26. The HPV up-to-date vaccination rate is suboptimal (51.1%) falling well below Healthy People 2020’s goal of 80% (Walker et al., 2019). Suboptimal rates are due in part to parental hesitancy and ambivalence about the HPV vaccine (Thompson, Rosen, Vamos, Kadono, & Daley, 2017). Intervention strategies are needed to address parental vaccine hesitancy.

Self-persuasion, defined as the process of generating one’s own arguments for engaging in behavior, is a promising intervention strategy to address parental HPV vaccine hesitancy. Basic psychological research has demonstrated that self-generated arguments are more effective than arguments from an external source (Aronson, 1999; Baldwin, Rothman, Vander Weg, & Christensen, 2013; Petty & Wegener, 1998) and are similar to tailored arguments that are customized to individuals’ unique needs (Baldwin et al., 2013). Thus, a self-persuasion intervention may be a particularly effective and efficient approach to address HPV vaccine indecision and hesitancy because it leverages parents’ own, personally relevant reasons for the vaccine rather than arguments from an external source.

Optimizing a self-persuasion intervention approach to promote HPV vaccination for real-world settings requires: (1) translating basic research methods of self-persuasion into an intervention strategy that can be disseminated and sustained, and (2) investigating hypothesized mechanisms for how and why self-persuasion is effective. Consistent with the ORBIT model of behavioral intervention development (Czajkowski et al., 2015), we conducted a multi-stage project to develop and test a parent-targeted mHealth self-persuasion intervention app to promote adolescent HPV vaccination (Tiro et al., 2016). The project’s first stage maps to Phase 1a of the ORBIT model in which we tested feasibility of intervention components (Baldwin et al., 2017). The second stage maps to ORBIT’s Phase 1b in which we conducted a randomized factorial trial to test and refine the independent effects of intervention components, with some results from this stage reported here. We specifically chose an app format to maximize future dissemination potential and minimize burden on clinic staff. We tested it in a safety-net clinic population to extend its reach and generalizability as previous work on self-persuasion had been done primarily in lab-based experiments (Baldwin et al., 2013; Stone, Aronson, Crain, Winslow, & Fried, 1994) in which well-educated participants wrote self-generated arguments (Baldwin et al., 2013; Stice, Marti, Spoor, Presnell, & Shaw, 2008; Stone et al., 1994).

In the initial development stage, we adapted established methods of written self-persuasion into two separate self-persuasion components: (1) verbalize reasons for vaccination and (2) choose HPV vaccine topics (Baldwin et al., 2017; Tiro et al., 2016) to reach and engage parents regardless of their educational or health literacy background. We also translated the mHealth app into Spanish because over 18.3 million US children are Latino (24%) and 73% of the families speak Spanish in the home (Patten, 2016). Our formative work showed that verbalizing reasons for vaccination and choosing vaccine topics were both feasible and helpful to less educated English- and Spanish-speaking parents attending safety-net clinics (Baldwin et al., 2017). In the main trial analyses, we demonstrated the verbalization component had a significant effect on parents’ vaccination intentions (Tiro et al., under review). This secondary analysis examines putative mechanisms of the intervention components for how and why the intervention works. Understanding mechanisms is the critical next step in refining and optimizing the self-persuasion intervention components (Nielsen et al., 2018; Riddle et al., 2015).

Hypothesized Mechanisms of Self-Persuasion

Self-persuasion, generating one’s own reasons for engaging in a behavior, is characterized by two processes, deep cognitive processing of the content and selecting reasons among potential alternatives. These processes informed the design of the intervention components (i.e., verbalize reasons for HPV vaccination and choose vaccine topics) and suggest two putative mechanisms, respectively, by which the intervention might affect HPV vaccine decision-making and behavior. First, verbalizing vaccination reasons (vs. listening to reasons) should result in better memory for the reasons, one putative mechanism. Self-generated content is cognitively processed more deeply than content from an external source (Craik & Lockhart, 1972). Consistent with theories of persuasion (Petty & Cacioppo, 1986), when parents generate their own reasons for getting the HPV vaccine, these reasons are more likely to be convincing (Baldwin et al., 2013) and influence vaccination behavior because the content is more accessible in memory (Slamecka & Graf, 1978). Memory for health messages predicts different health behaviors (Farrelly et al., 2012; Grenard, Dent, & Stacy, 2013). Second, choosing HPV vaccine topics from among a set of potential topics (vs. being assigned topics) should result in greater autonomous motivation for vaccination, the other putative mechanism. Autonomous motivation for behavior reflects volition and perception that the behavior is personally important (Ryan, Patrick, Deci, & Williams, 2008). Choice is known to elicit autonomous motivation (Cordova & Lepper, 1996; Ryan et al., 2008). Thus, when parents select a subset of HPV vaccination topics from among a broader set of potential topics, they may be more likely to believe that HPV vaccination is personally important. Substantial evidence across various health behavior domains indicates that behavior is more likely when it is autonomously motivated (Gorin, Powers, Koestner, Wing, & Raynor, 2014; Ng et al., 2012). To date, memory and autonomous motivation have not been tested as mechanisms of HPV vaccination.

The experimental medicine approach (Nielsen et al., 2018; Riddle et al., 2015; Sheeran, Klein, & Rothman, 2017) provides a useful framework to test and evaluate mechanistic pathways in behavioral interventions. In contrast to traditional intervention mediation analyses that focus on whether the intervention has an indirect effect on the target behavior via putative mechanisms, the experimental medicine approach encourages separate tests of the mechanistic pathways to evaluate whether (a) intervention components effect a change in putative mechanisms and (b) putative mechanisms are associated with changes in the behavioral outcome. Evaluating both pathways separately provides clearer findings to unpack the ways in which the self-persuasion intervention and its putative mechanisms worked (or did not work), and to make decisions about how to refine it for future testing and dissemination. For example, if there is no evidence for associations between the putative mechanisms and HPV vaccination (i.e., Figure 1, path b1: memory → vaccination; path b2: autonomous motivation → vaccination), those results would raise questions about whether the hypothesized mechanisms are determinants of HPV vaccination and would suggest caution in continuing to pursue memory and/or autonomous motivation as intervention targets. Alternatively, if data support associations between the putative mechanisms and vaccination, but evidence does not support that intervention components changed the putative mechanisms (i.e., Figure 1, path a1: verbalize reasons → memory; path a2: choose topics → autonomous motivation), those results would suggest the components were ineffective in engaging the target mechanisms, and future studies should explore alternative strategies to change the mechanisms (see Rothman & Baldwin, 2019).

Figure 1.

Figure 1.

Open in a new tab

Hypothesized effects of the intervention components on target mechanisms and mechanism effects on vaccination

Current Study

This secondary analysis was guided by the experimental medicine approach (Nielsen et al., 2018; Riddle et al., 2015; Sheeran et al., 2017) to test hypothesized mechanisms of the self-persuasion intervention app for parental decision-making about adolescent HPV vaccination. Thus, this analysis extends beyond a traditional intervention efficacy test. For reasons outlined in the methods (below), it was infeasible to test mechanism effects on the primary trial outcome, vaccination intentions. Therefore, we tested the following hypotheses on the secondary outcome of HPV vaccination. First, two self-persuasion intervention components, verbalizing vaccination reasons and choosing HPV vaccine topics, will result in greater memory for vaccine reasons and autonomous motivation for vaccination, respectively (Figure 1, paths a1 and a2). Second, measures of the putative mechanisms will be significantly associated with HPV vaccine series initiation 12 months after exposure to the intervention (Figure 1, paths b1 and b2). Although not the focus of these analyses, we also report the direct effect of the intervention on vaccination (Figure 1, path c1: verbalize reasons → HPV vaccination and path c2: choose topics → HPV vaccination). In addition, we explored the interaction between the verbalize and choose components on the outcomes, although this was not hypothesized a priori.

Methods

For details about the factorial randomized controlled trial design, see Tiro et al. (2016). Briefly, participants were parents of pediatric patients in a safety-net healthcare system serving low-income, uninsured populations in Dallas County (TX). The system supports delivery of all guideline-recommended vaccines in two specific ways: leadership trains the provider team on the importance of strongly recommending vaccines at every visit and clinics have standing orders (i.e., prewritten orders and instructions empowering nurses to administer recommended vaccines). Parents who reported being undecided about vaccinating their child were randomized to one of four conditions in a 2 (cognitive processing: verbalize reasons vs. listen) × 2 (choice: choose HPV topics vs. assigned) factorial trial designed to manipulate two key components of self-persuasion (see description below and in Tiro et al., 2016). The primary outcome was HPV vaccination intentions and a secondary outcome was vaccination decision stage; findings are reported in Tiro et al. (under review). This paper tests the putative mechanisms of the self-persuasion intervention on the secondary outcome of 12-month vaccination behavior. The IRB committees of the University of Texas Southwestern Medical Center and Southern Methodist University approved the study and it was registered at clinicaltrials.gov (NCT02537756).

Participants

We used the electronic medical record (EMR) to identify families with a 11–17 year old child who had not received any HPV vaccine doses. We included parents who reported on a baseline telephone survey that they had never thought or were undecided about the HPV vaccine for their child. Exclusion criteria included sibling enrolled in the study, primary language not English or Spanish, no appointment in the prior 18 months (i.e., not current patient), no telephone access, impaired hearing or speech, or adolescent was pregnant. The safety-net system provides the HPV vaccine for free to all patients through the federal Vaccines for Children Program. Parents (N = 161) used the app and completed study procedures described below.

Procedures

Via telephone, bilingual research assistants (RAs) recruited parents, ascertained eligibility, collected a baseline survey, and scheduled an in-person research session at the child’s clinic. Due to the length of the research session and to avoid disrupting clinic procedures, the session was purposefully not scheduled adjacent to the adolescent patient’s clinic visit. Prior to each session, the project coordinator (EM) used a random assignment sequence list, developed following block randomization procedures, to assign the intervention condition in the app. Thus, the RA, who met the parent participant, remained blinded to condition.

The tablet-based app contained six separate tasks (see Baldwin et al., 2017, for development of these procedures):

  • Task A: All parents watched a 4-minute educational video about HPV and the vaccine to ensure all had some knowledge to reasonably complete the subsequent self-persuasion tasks.

  • Task B (Choose Vaccine Topics): Parents in the two choice conditions selected the three most interesting topics among six topics (protect health, prevent infection and cancer, protect future spouse or partner, lower concerns, most effective before sexually active, and future regret). Topics were previously rated as very helpful to parents in thinking about the HPV vaccine (Baldwin et al., 2017). Task B was skipped for parents randomized to the assigned topics conditions, and the app delivered three topics (protect health, lower concerns, future regret) reflecting known determinants of HPV vaccination.

  • Task C (Answer Questions): Parents in the two verbalize conditions audio-recorded answers to three question prompts based on the topics they chose or were assigned.

  • Task D (Verbalize Reasons): Parents in the two verbalize conditions then audio-recorded up to three reasons, summarized in their own words, for getting their child the vaccine.

  • Task E (Listen to Reasons): Parents in the two listen conditions heard three different vaccination reasons based on either the topics they chose or were assigned. These reasons were delivered using a personal narrative format based on formative work gathered from other parents’ attending the same safety-net clinics (Baldwin et al., 2017).

  • Task F: All parents reported their vaccination intentions and decision stage.

The number and order of tasks depended on the intervention condition. Specifically, Condition 1 (verbalize, choice) completed tasks A, B, C, D, and F; Condition 2 (verbalize, assigned) completed tasks A, C, D, and F; Condition 3 (listen, choice) completed tasks A, B, E, and F; and Condition 4 (listen, assigned) completed tasks A, E, and F.

Procedures to assess the two target mechanisms occurred after parents completed the self-persuasion tasks in the app and after parents reported their vaccination intentions. This temporal order of assessments is what precludes testing the effects of memory and autonomous motivation on intentions. We considered measuring the mechanisms before intentions; however, we opted against that assessment order because the measurement itself could have constituted an additional intervention and muddied the intervention components’ effects on the primary outcome. To assess memory of vaccination reasons, the app prompted parents to recall as much as they could about the reasons for vaccination they had previously verbalized or listened to and to audio-record their response. There was a 10-minute gap between the self-persuasion tasks and the recall task, a common procedure to distract participants’ attention and minimize cognitive rehearsal of the vaccination reasons prior to the recall task (see Kiviniemi & Rothman, 2006). Specifically, the RA told parents we were interested in their feedback about educational apps for children and asked them to sample several apps located on the tablet’s home screen while the RA ostensibly prepared for the final interview questions. After 10 minutes, the RA returned to the room and re-launched the HPV vaccination app to initiate the recall task. Afterward, the RA conducted an exit interview and assessed parents’ autonomous motivation for vaccinating their child, the other putative mechanism. Supplemental Table 1 reports average time spent completing self-persuasion and recall tasks by condition.

Measures

Mechanism: Memory for vaccination reasons.

To measure memory, we determined how accurately parents recalled the vaccination reasons they verbalized or heard previously (Kiviniemi & Rothman, 2006). Recall accuracy was operationalized as the number of topics in favor of vaccination parents stated in the recall task recording that corresponded to reasons parents recorded in the verbalize conditions (Task D) or topics presented as personal narratives in the listen conditions (Task E). Audio recordings of the recall task (all parents) and the verbalized reasons (parents in two verbalize conditions) were transcribed. Spanish audio recordings were transcribed in Spanish and then translated to English. Two independent coders then coded each transcript. First, coders noted whether the reasons were in favor of or against vaccination. Then, reasons in favor of vaccination were mapped to the six HPV vaccine-related topics (protect health, prevent infection and cancer, protect future spouse or partner, lower concerns, most effective before sexually active, and future regret) on which the question prompts (Task C) and personal narrative reasons (Task E) were based. If the pro-vaccination reason did not map to any of the six topics, coders placed it in the “other” category. Reasons against the vaccine were also mapped to different topics, but we did not account for them in the recall accuracy variable because the hypothesis focused on memory for reasons in favor of vaccination. Interrater agreement was high with agreement on 92% of the codes (concordance rate: M = 0.92, SD = 0.10, Median = 0.97, Mode = 1.00) and discrepancies were resolved through discussion.

Ten parents in the verbalize conditions did not have data for the recall accuracy variable. Six did not verbalize any reasons in favor of vaccination, three did not have audio recordings of vaccine reasons, and one did not offer any reasons for or against vaccination.

Mechanism: Autonomous motivation for vaccination.

We adapted the Treatment Self-Regulation Questionnaire (TSRQ), a well-validated measure of motivation in health behavior, for HPV vaccination (Denman, Baldwin, Marks, Lee, & Tiro, 2016). The stem was “The reason you would get [child’s name] the vaccine is because…” The four items in the autonomous motivation subscale were “…it is consistent with your goals as a parent”, “…you want to take responsibility for your child’s health”, “…it is important”, and “you believe it is the best thing for your child”. Responses were given on a 1 (strongly disagree) to 5 (strongly agree) scale. Items were assessed in the exit interview (α = .86) and the baseline questionnaire (α = .79), and inter-item reliability was good at both assessment points. Two parents did not complete the exit interview and thus did not provide responses to post-intervention autonomous motivation.

Adolescent vaccination status.

We extracted HPV vaccine dose dates from the EMR 12 months following the intervention and categorized the child’s vaccine series initiation status (1+ doses vs. 0 doses). Because parents used the app outside of a clinic visit, we selected a 12-month frame to give them opportunity to schedule and complete a visit with their child’s provider. Over three-quarters (n = 124; 77%) had a visit within 12 months of using the app.

Covariates.

From adolescent patients’ EMR, we extracted demographic characteristics (sex, age group, race/ethnicity) and healthcare utilization pattern (insurance type, clinic, recency of last clinic visit). We also documented parents’ preferred language, baseline autonomous motivation score, and time of recruitment into the trial after the study start date.

Analysis Plan

Summary statistics for parents’ and adolescents’ demographic and healthcare utilization characteristics were reported using counts and percentages for categorical variables, and using mean and standard deviation (SD) for continuous variables. Parent characteristics were compared across the two factorial conditions using Chi-square or Fisher’s exact test for categorical variables, and using t-tests for continuous variables.

To test the effect of the verbalize component on the putative mechanism—memory (Figure 1, path a1), we used a multivariate Poisson regression model with recall accuracy as the outcome variable, controlling for the choice condition, parent preferred language, child sex, child age group, clinic and time of recruitment into the trial. The model also included an “offset” variable (i.e., the number of different vaccination topics parents verbalized or heard) because the actual number of topics parents were exposed to varied by condition. Parents in the two listen conditions were exposed to three topics, whereas parents in the verbalize conditions were asked to provide three reasons, the actual number of reasons provided ranged from one to four. To test the effect of the choice component on the putative mechanism— autonomous motivation (Figure 1, path a2), we used a multivariate linear regression model with the post-intervention autonomous motivation score as the outcome variable, controlling for the baseline motivation score, the verbalize condition, and the same covariates included in the model testing effects on memory. To test the direct effect of the intervention components on 12-month HPV vaccination (Figure 1, paths c1 and c2), we used a logistic regression model with the two intervention conditions as the independent variables, controlling for the same covariates as the models testing effects on the mechanisms. To explore the interactive effects of the intervention conditions on the outcomes, we also included an interaction term between the two binary condition variables.

To test the effect of recall accuracy on 12-month HPV vaccination (Figure 1, path b1), we used a series of logistic regression models. All models used recall accuracy as the main independent variable, controlling for the number of topics verbalized or heard as an offset variable. The inclusion of the offset creates an independent variable that is interpreted as a relative ratio of recall accuracy. In Model 1, recall accuracy was the only independent variable, to examine its univariate association with vaccination status. In subsequent models, covariates were sequentially added to examine if the association between recall accuracy and vaccination status changed in the presence of the covariates. Model 2 added the verbalize and choice condition variables. In Model 3, the same set of covariates (parent preferred language, child sex, child age group, clinic, and time of recruitment into the trial) were added to Model 2. To test the effect of post-intervention autonomous motivation scores on 12-month vaccination status (Figure 1, path b2), we used a similar series of logistic regression models. Model 1 only included post-intervention autonomous motivation score (univariate effect). Subsequent models followed the same order of adding covariates as the recall accuracy models, except a fourth model that included baseline autonomous motivation scores.

Statistical tests were two-sided and the significance level was set at 5%. Linear regression coefficients, relative ratios, and odds ratios along with 95% confidence intervals (CI), and two-sided p-values were reported. Given the sample size (N=161), there is at least 80% power to detect small-to-moderate effects for all the outcomes (i.e., vaccination, recall, autonomous motivation). Analyses were conducted using SAS version 9.4 (SAS Institute Inc. Cary, NC).

Results

Sample Characteristics

Table 1 describes demographics, health care utilization, and study variables for the entire sample and collapsed across the two intervention factors. The only difference between the conditions was a significantly higher proportion of Spanish-speaking parents in the verbalize compared to listen conditions.

Table 1.

Sample characteristics by factorial conditions, Choose Topics vs. Assigned Topics and Verbalize vs. Listen

Characteristics All Groups N=161 (%) Choose Topics n=77 (%) Assigned Topics n=84 (%) Verbalize n=86 (%) Listen n=75 (%)
Child Sex
 Male 95 (59.0) 44 (57.1) 51 (60.7) 49 (57.0) 46 (61.3)
 Female 66 (41.0) 33 (42.9) 33 (39.3) 37 (43.0) 29 (38.7)
Age group
 11–12 y 121 (75.2) 62 (80.5) 59 (70.2) 65 (75.6) 56 (74.7)
 13–17 y 40 (24.8) 15 (19.5) 25 (29.8) 21 (24.4) 19 (25.3)
Race/Ethnicity
 Hispanic 124 (77.0) 55 (71.4) 69 (82.1) 70 (81.4) 54 (72.0)
 Non-Hispanic, black 36 (22.4) 21 (27.3) 15 (17.9) 15 (17.4) 21 (28.0)
 Non-Hispanic white/other 1 (0.6) 1 (1.3) 0 (0.0) 1 (1.2) 0 (0.0)
Language preference, parent *
 Spanish 110 (68.3) 48 (62.3) 62 (73.8) 66 (76.7) 44 (58.7)
 English 51 (31.7) 29 (37.7) 22 (26.2) 20 (23.3) 31 (41.3)
Clinic **
 Largest (Hatcher) 66 (41.0) 31 (40.3) 35 (42.7) 36 (41.9) 30 (40.0)
 Other clinics 95 (59.0) 46 (59.7) 49 (58.3) 50 (58.1) 45 (60.0)
Insurance type ¥
 Public 110 (68.3) 54 (70.1) 56 (66.7) 58 (67.4) 52 (69.3)
 Private 41 (25.5) 19 (24.7) 22 (26.2) 21 (24.4) 20 (26.7)
 No insurance 10 (6.2) 4 (5.2) 6 (7.1) 7 (8.1) 3 (4.0)
Recency of last clinic visit ¥
 0 to <6 months 84 (52.2) 35 (45.5) 49 (58.3) 46 (53.5) 38 (50.7)
 6 to <12 months 44 (27.3) 24 (31.2) 20 (23.8) 21 (24.4) 23 (30.7)
 12 to < 18 months 27 (16.8) 16 (20.8) 11 (13.1) 13 (15.1) 14 (18.7)
 18+ months 6 (3.7) 2 (2.6) 4 (4.8) 6 (7.0) 0 (0.0)
Time of recruitment into trial after start
 0 to <6 months 39 (24.2) 18 (23.4) 21 (25.0) 20 (23.3) 19 (25.3)
 7 to <12 months 51 (31.7) 24 (31.2) 27 (32.1) 30 (34.9) 21 (28.0)
 12 to <18 months 27 (16.8) 14 (18.2) 13 (15.5) 14 (16.3) 13 (17.3)
 18 to <24 months 26 (16.1) 11 (14.3) 15 (17.9) 13 (15.1) 13 (17.3)
 24 to 30 months 18 (11.2) 10 (13.0) 8 (9.5) 9 (10.5) 9 (12.0)
Baseline Autonomous Motivation M (SD) 4.01 (0.59) 4.07 (0.48) 3.96 (0.67) 4.09 (0.46) 3.92 (0.70)
Open in a new tab
*

Significantly higher proportion of Spanish-speakers in the verbalize vs. listen conditions (p < 0.05)

**

Regular clinic confirmed in initial screening call

¥

Measured at closest PCP clinic appointment prior to baseline survey

Direct Effects of the Intervention Components on 12-month Vaccination Status

Neither the verbalize reasons intervention component (odds ratio: 0.89, 95% CI: 0.45, 1.79, p = .75) or the choose vaccine topics component (odds ratio: 1.37, 95% CI: 0.69, 2.71, p = .36) had significant effects on 12-month vaccination status. Given that neither main effect was significant, we did not test the interaction effect.

Effects of the Intervention Components on Target Mechanisms

Verbalize reasons for vaccination component on memory (Path a1).

As hypothesized, the verbalize intervention component had a significant effect on parents’ recall accuracy of vaccination reasons. Model results are reported in Table 2. Although not hypothesized, there was also a significant effect of the choose vaccine topics component and a significant verbalize × choice interaction, indicating effects of the verbalize component depends on the choice condition and vice-versa. When HPV vaccine topics were assigned (reference category), parents verbalizing reasons (Condition #2) were 2.49 times more accurate in their recall (main effect rate ratio) compared to parents who listened to reasons (Condition #4). When parents chose the topics, those verbalizing reasons (Condition #1) were only 1.39 times more accurate (main effect 2.49 multiplied by interaction 0.56) compared to parents who listened to reasons (Condition #3). The interaction also indicates that among parents listening to HPV vaccine reasons (reference category), those choosing topics (Condition #3) were 1.95 times more accurate in their recall compared to parents assigned topics (Condition #4). Among parents verbalizing vaccine reasons, those choosing topics (Condition #1) were only 1.09 times more accurate (1.95 multiplied by 0.56) compared to parents assigned topics (Condition #2). Thus, the effect of the verbalize component on recall accuracy was strongest in the absence of choice, and the effect of the choice component was strongest in the absence of verbalization. This pattern suggests the two component effects on recall accuracy largely overlap with each other rather than being independently additive (i.e., verbalize plus choice did not result in the highest accuracy).

Table 2.

Effects of the intervention components on the target mechanisms

Vaccine Reasons Recall Accuracy
N (M, SD) Rate Ratio (95% CI) p
Intervention Components
 Verbalize (Conditions 1 & 3) 76 (1.71, 0.92) 2.49 (1.79, 3.46) <.001
 Listen (Conditions 2 & 4) 75 (1.27, 0.99) Reference
 Choose Topics (Conditions 1 & 2) 74 (1.66, 0.97) 1.95 (1.38, 2.75) <.001
 Assigned Topics (Conditions 3 & 4) 77 (1.32, 0.97) Reference
 Verbalize × Choose Interaction 0.56 (0.36, 0.87) .010
Post-Intervention Autonomous Motivation
N (M, SD) b estimate (95% CI) p
Intervention Components
 Choose Topics (Conditions 1 & 2) 76 (4.30, 0.57) 0.03 (−0.01, 0.20) .745
 Assigned Topics (Conditions 3 & 4) 83 (4.21, 0.69) Reference
 Verbalize (Conditions 1 & 3) 85 (4.29, 0.49) 0.01 (−0.17, 0.19) .913
 Listen (Conditions 2 & 4) 74 (4.22, 0.77) Reference
Open in a new tab

Note. Autonomous motivation was modeled using linear regression and the model included baseline autonomous motivation. Two participants did not complete exit interviews and thus are not included in the motivation models. Recall accuracy was modeled using generalized linear regression with Poisson distribution and included participant-level number of vaccine topics verbalized or heard as an offset variable. Ten participants in the verbalize conditions were excluded from the recall models because they did not have coded reasons. Both models also included parent preferred language, child sex, child age group, clinic, and time of recruitment into the trial as covariates.

Choose vaccine topics component on autonomous motivation (Path a2).

Contrary to the hypothesis, there was no effect of the choose intervention component on post-intervention autonomous motivation scores (Table 2). There was also no effect of the verbalize component. Unexpectedly, the mean values of autonomous motivation post-intervention (Table 2) increased relative to baseline values (in Table 1) across all four conditions. Given that neither main effect was significant, we did not test the interaction effect on autonomous motivation scores.

Effects of the Target Mechanisms on 12-month Vaccination Status

Associations between memory and HPV vaccine series initiation (path b1).

Contrary to the hypothesis, recall accuracy was not associated with HPV vaccine series initiation (top half of Table 3). This lack of effect was observed regardless of the covariates included in the models.

Table 3.

Tests of the associations between measures of the target mechanisms and 12-month vaccination status

12-month Vaccination Status
Odds Ratio (95% CI) p
Vaccine Reasons Recall
 Model 1: Recall Only 1.38 (0.53, 3.62) .513
 Model 2: Intervention Conditions Added 1.39 (0.47, 4.11) .549
 Model 3: Covariates Added 1.00 (0.32, 3.12) .999
Autonomous Motivation
 Model 1: Motivation Only 1.92 (1.09, 3.40) .025
 Model 2: Intervention Conditions Added 1.90 (1.07, 3.36) .029
 Model 3: Covariates Added 2.00 (1.08, 3.69) .028
 Model 4: Baseline Motivation Added 1.54 (0.77, 3.05) .221
Open in a new tab

Note. Successive models in the table added covariates, or a set of covariates, to the previous model to examine the effect adding them had on the association between the mechanism and vaccination status. Model 1’s: Mechanism as only predictor; Model 2’s: Mechanism + Intervention Conditions; Model 3’s: Mechanism + Intervention Conditions + Covariates; Model 4: Motivation + Intervention Conditions + Covariates + Baseline Motivation. The set of covariates included parent preferred language, child sex, child age group, clinic, and time of recruitment into the trial.

Associations between autonomous motivation and HPV vaccine series initiation (path b2).

As hypothesized, post-intervention autonomous motivation scores were significantly associated with HPV vaccine series initiation (bottom half of Table 3). When post-intervention autonomous motivation score was the only independent variable (Model 1), higher autonomous motivation scores significantly predicted vaccine initiation. Likelihood of vaccination increased nearly two times with each unit increase in autonomous motivation scores (odds ratio = 1.92). This effect remained consistent when adding both intervention conditions (Model 2) and additional covariates (Model 3). However, when baseline autonomous motivation was added (Model 4), the association between post-intervention autonomous motivation score and vaccine series initiation became insignificant and the odds ratio reduced in magnitude (adjusted odds ratio: 1.54, 95% CI: 0.77, 3.05); and the association between baseline autonomous motivation and vaccination was larger than post-intervention autonomous motivation score (odds ratio = 1.91) but did not reach the conventional level of significance (p = .084).

Discussion

Using the experimental medicine approach as a guiding framework, tests of the hypothesized mechanisms of the self-persuasion intervention app had mixed findings. One of the self-persuasion intervention components, verbalize reasons for vaccination, elicited a significant effect on its hypothesized mechanism (memory for vaccination reasons), whereas the other component, choose vaccine topics, did not affect its hypothesized mechanism (autonomous motivation). The intervention components did not have direct effects on 12-month vaccination. For the associations between the mechanisms and vaccination status, autonomous motivation was positively associated with vaccination but memory for vaccination reasons was not associated. These data expand understanding of the main trial findings for the self-persuasion app (Tiro et al., under review), and represent the first time that memory and autonomous motivation have been investigated as mechanisms of adolescent HPV vaccination. In terms of the experimental medicine approach, the findings reflect a “full test” of both mechanistic pathways (Sheeran et al., 2017, p. 591) rather than tests focused solely on one pathway (e.g., studies testing the intervention-to-mechanism effect; see Davidson, Mogavero, & Rothman, 2020).

Both intervention components (verbalize and choice) and the interaction between components had significant effects on parents’ ability to accurately recall their vaccination reasons. Neither the choice effect nor the interaction were hypothesized a priori. The interaction effect indicated absence of both components (i.e., parents listening to assigned reasons for vaccination) led to the lowest recall accuracy levels among the four conditions. Collectively, data indicate both intervention tasks effectively increased parents’ cognitive engagement with the app content on vaccination reasons. The findings on recall accuracy are consistent with the Elaboration Likelihood Model (Petty & Cacioppo, 1986) and evidence from the persuasion and health communication literatures that personally relevant information is: (1) more likely to be cognitively processed deeply and deliberately (Ko et al., 2011; Kreuter et al., 1999), and (2) more likely to be remembered (Kiviniemi & Rothman, 2006). Moreover, the findings suggest the verbalization or choice component (or both) could be used for behavioral interventions in which learning and understanding health information is critical because the app and its components allow individuals to control the pace and content of instruction. This approach is consistent with the tenets of information control theory (ICT; Eveland & Dunwoody, 2001) and might be particularly effective in (1) countering misinformation about vaccines (Larson, 2018) and (2) encouraging skeptical individuals to engage more substantively with trusted sources of information. Vaccine hesitancy is considered one of the top ten global health threats by the World Health Organization (WHO, 2019). Effective strategies to counter misinformation about vaccines are needed, particularly during the COVID-19 pandemic (Larson, 2018; WHO, 2020).

Contrary to our hypothesis, the choose vaccine topics component did not influence autonomous motivation (Figure 1, path a2), inconsistent with evidence from other domains (Cordova & Lepper, 1996; Ryan et al., 2008). In formative work, all six vaccine topics were rated as important (Baldwin et al., 2017); thus, perhaps there was not enough differentiation among the topics to elicit a strong choice effect (Baldwin et al., 2013). Another explanation for the absence of a choice effect is that using the app increased autonomous motivation regardless of intervention condition, potentially constraining detection of a choice effect. Tables 1 and 2 support this explanation as autonomous motivation increased from baseline assessment to post-intervention for all parents. Unfortunately, we did not have a group with baseline motivation assessed who had no exposure to the app to make a comparison and draw clear conclusions.

For the tests of mechanisms on 12-month HPV vaccination, recall accuracy was not associated with vaccination (Figure 1, path a2) in univariate or multivariate models. Prior literature is mixed with some evidence supporting the association between memory of relevant messages and behavior (Grenard et al., 2013; Farrelly et al., 2012) and others (like ours) not supporting an effect (Brick et al., 2016). Despite the intervention successfully engaging recall accuracy as hypothesized, findings indicate caution in pursuing memory for vaccination reasons as a target mechanism of HPV vaccination (see Rothman & Baldwin, 2019), at least when the opportunity to vaccinate is temporally distal. The effect of memory of vaccination reasons may be time sensitive and the 12-month follow-up being too long to observe an effect. The lengthy follow-up might also explain the lack of direct effects of the intervention components on vaccination. Alternatively, memory for vaccination reasons may have an indirect effect on vaccination, such as making parents more receptive to a provider’s recommendation (Gilkey et al., 2016) or persistent communication about HPV vaccination (Shay et al., 2018). To test both of these explanations, future studies could ask parents to use the app during their child’s clinic visit and test whether recall accuracy predicts same day vaccination and if effects are moderated when providers strongly recommend the vaccine.

The other target mechanism, autonomous motivation, significantly predicted 12-month HPV vaccination (Figure 1, path b2), which is consistent with studies in other health behavior domains (Gorin et al., 2014; Ng et al., 2012) but had not been previously observed in HPV vaccination. This pattern was robust even when covariates known to predict vaccination behavior were added (Model 3). However, post-intervention motivation did not predict vaccination when baseline autonomous motivation was included (Model 4). This pattern of findings suggests that the incremental change in autonomous motivation was not predictive of vaccination, but the variance shared between baseline and post-intervention motivation predicted vaccination status. The fact that autonomous motivation predicted vaccination suggests it is a viable mechanism to target in promoting adolescent HPV vaccination. Interventions in other domains demonstrate that autonomous motivation is a modifiable mechanism (Ng et al., 2012), but alternative intervention strategies from the self-persuasion components tested here are needed to successfully target it for HPV vaccination. For example, intervention messages that specifically frames adolescent HPV vaccination in terms of parents’ responsibility for their child’s health and conveys support for parents’ autonomy in that decision might prove to be effective (see Williams et al., 2006).

The findings have clear implications for understanding the conditions under which the intervention app and the mechanisms examined here might be more and less effective in influencing HPV vaccination. When the app is used outside of an immediate opportunity to vaccinate, recall of vaccination reasons is not an effective mechanism. However, the recall mechanism may have an effect if the app is used temporally adjacent to a clinic appointment. This possibility is consistent with meta-analytic findings that the timing of an opportunity to enact a behavior is a condition that moderates intervention effects (Ferrer & Cohen, 2019). In contrast, our findings indicate the effect of autonomous motivation on vaccination does not depend on an immediate opportunity, but the app should not be used to target this mechanism until alternative strategies to successfully elicit autonomous motivation are developed and tested.

Our findings also suggest concrete next steps for future testing and dissemination. There are two clear strategies to better understand the app’s effect and the mechanisms that should be investigated in future studies within an implementation science framework. First, when testing in clinic settings, implementation strategies need to be developed on how the app could be integrated into clinic protocols (Powell et al., 2015), so that delivery of the app is best timed for action. For example, clinic staff could send invitations through a patient web portal to families with upcoming clinic appointments or a tablet with the app could be made available to parents while they are waiting for their child’s visit. Directing parents to use it while waiting to see their child’s provider is likely feasible given average time to complete the video and self-persuasion tasks was 9–10 minutes (see Supplemental Table 1). Both these invitation methods would ensure that all parents receive consistent education about the importance of HPV vaccination, and prime them to have a focused discussion with their provider if they have any remaining concerns (Gilkey et al., 2016). Second, increasing the likelihood parents follow through to get their child the HPV vaccine after using the app may require targeting action planning skills (e.g., explain whether local clinics have walk-in immunization hours, scheduling immunization appointments), especially if the app is distributed in community settings without an immediate vaccination opportunity. Both strategies are best suited for evaluation through a hybrid dissemination-effectiveness study because it would advance understanding of an action planning intervention component and the implementation strategies needed for delivery in clinical settings.

The study has a few limitations. First, participants were not invited to use the app during a clinic visit. Thus, we do not know the effect on same day vaccination behavior or how the app’s effect might interact with a strong provider recommendation, a known determinant of HPV vaccination (Gilkey et al., 2016). Using the self-persuasion app just prior to a clinic visit might better prepare parents for a discussion with their provider and make them more receptive to a strong provider recommendation. Therefore, it is possible the app’s effect on vaccination behavior would be stronger if integrated within a clinic appointment when the family uses the app and then receives a strong provider recommendation with an opening for parents to discuss their concerns (Shay et al., 2018). Then the adolescent could receive the vaccine immediately if the parent decides in favor. Instead, we assessed the intervention conditions’ effect on HPV vaccination initiation after 12 months and 23% of the adolescents from the sample did not have a clinic appointment during that time. Second, it is possible the self-persuasion intervention has an effect through other mechanisms not examined here. For example, self-persuasion may elicit a positive affective response to the vaccination reasons, much like personal narratives in health communications (Shelby & Ernst, 2013). The positive affective response may serve as a mechanism of vaccination as it has been shown to be a robust determinant of various health behaviors (Williams, Rhodes, & Conner, 2018). Third, the study was conducted with mostly minority families accessing care through a safety-net healthcare system and effects may be different in majority white or privately insured populations. Future research should investigate alternate mechanisms and other target populations.

Conclusion

What can be learned about the intervention and hypothesized mechanisms given the mixed results? First, the intervention app effectively results in greater memory for vaccination reasons. Together with the main trial findings indicating the verbalize component increases vaccination intentions (Tiro et al., under review), the intervention app effectively influences parents’ thoughts about the HPV vaccine. Although the findings signal caution about whether memory for vaccination reasons is a viable mechanism of vaccination, testing the intervention more temporally adjacent to a vaccination opportunity should be addressed before definitively ruling out memory as an intervention target. Second, the current intervention does not increase autonomous motivation for HPV vaccination and alternative intervention strategies to target this promising mechanism need to be tested. Future research should also identify complementary strategies to implement the app within clinical workflows to further evaluate the app’s effects.

Supplementary Material

Supplemental Material 1

Acknowledgments

All phases of this study were supported by the National Institutes of Health (NIH), grant number R01CA178414. Additional support provided by the UTSW Center for Translational Medicine, through the NIH/National Center for Advancing Translational Sciences (UL1TR001105) and the Simmons Comprehensive Cancer Center (1P30 CA142543), and from the Lyda Hill Foundation. The authors have no conflicts of interest or financial relationships relevant to this article to disclose.

References

  1. Aronson E (1999). The power of self-persuasion. American Psychologist, 54, 875–884. 10.1037/h0088188 [DOI] [Google Scholar]
  2. Baldwin AS, Denman DC, Sala M, Marks EG, Shay LA, Fuller S, … Tiro JA (2017). Translating self-persuasion into an adolescent HPV vaccine promotion intervention for parents attending safety-net clinics. Patient Education and Counseling, 100, 736–741. 10.1016/j.pec.2016.11.014 [DOI] [PMC free article] [PubMed] [Google Scholar]
  3. Baldwin AS, Rothman AJ, Vander Weg MW, & Christensen AJ (2013). Examining causal components and a mediating process underlying self-generated health arguments for exercise and smoking cessation. Health Psychology, 32, 1209–1217. 10.1037/a0029937 [DOI] [PubMed] [Google Scholar]
  4. Brick C, McCully SN, Updegraff JA, Ehret PJ, Areguin MA, & Sherman DK (2016). Impact of cultural exposure and message framing on oral health behavior: Exploring the role of message memory. Medical Decision Making, 36, 834–843. 10.1177/0272989X15570114 [DOI] [PMC free article] [PubMed] [Google Scholar]
  5. Cordova DI, & Lepper MR (1996). Intrinsic motivation and the process of learning: Beneficial effects of contextualization, personalization, and choice. Journal of Educational Psychology, 88, 715–730. 10.1037/0022-0663.88.4.715 [DOI] [Google Scholar]
  6. Craik FIM, & Lockhart RS (1972). Levels of processing: A framework for memory research. Journal of Verbal Learning and Verbal Behavior, 11, 671–684. 10.1016/S0022-5371(72)80001-X [DOI] [Google Scholar]
  7. Czajkowski SM, Powell LH, Adler N, Naar-King S, Reynolds KD, Hunter CM, … Charlson ME (2015). From ideas to efficacy: The ORBIT model for developing behavioral treatments for chronic diseases. Health Psychology, 34, 971–982. 10.1037/hea0000161 [DOI] [PMC free article] [PubMed] [Google Scholar]
  8. Davidson KW, Mogavero JN, & Rothman AJ (2020). Using early phase studies to advance intervention research: The science of behavior change. Health Psychology, 39, 731–735. 10.1037/hea0000897 [DOI] [PubMed] [Google Scholar]
  9. Denman DC, Baldwin AS, Marks EG, Lee SC, & Tiro JA (2016). Modification and validation of the Treatment Self Regulation Questionnaire to assess parental motivation for HPV vaccination of adolescents. Vaccine, 34, 4985–4990. 10.1016/j.vaccine.2016.08.037 [DOI] [PMC free article] [PubMed] [Google Scholar]
  10. Eveland WP Jr, & Dunwoody S (2001). User control and structural isomorphism or disorientation and cognitive load? Learning from the Web versus print. Communication Research, 28, 48–78. 10.1177/009365001028001002 [DOI] [Google Scholar]
  11. Farrelly MC, Duke JC, Davis KC, Nonnemaker JM, Kamyab K, Willett JG, & Juster HR (2012). Promotion of smoking cessation with emotional and/or graphic antismoking advertising. American Journal of Preventive Medicine, 43, 475–482. 10.1016/j.amepre.2012.07.023 [DOI] [PubMed] [Google Scholar]
  12. Ferrer RA, & Cohen GL (2019). Reconceptualizing self-affirmation with the trigger and channel framework: Lessons from the health domain. Personality and Social Psychology Review, 23, 285–304. 10.1177/1088868318797036 [DOI] [PubMed] [Google Scholar]
  13. Gilkey MB, Calo WA, Moss JL, Shah PD, Marciniak MW, & Brewer NT (2016). Provider communication and HPV vaccination: The impact of recommendation quality. Vaccine, 34, 1187–1192. 10.1016/j.vaccine.2016.01.023 [DOI] [PMC free article] [PubMed] [Google Scholar]
  14. Gorin AA, Powers TA, Koestner R, Wing RR, & Raynor HA (2014). Autonomy support, self-regulation, and weight loss. Health Psychology, 33, 332–339. 10.1037/a0032586 [DOI] [PMC free article] [PubMed] [Google Scholar]
  15. Grenard JL, Dent CW, & Stacy AW (2013). Exposure to alcohol advertisements and teenage alcohol-related problems. Pediatrics, 131, e369–e379. 10.1542/peds.2012-1480 [DOI] [PMC free article] [PubMed] [Google Scholar]
  16. Kiviniemi MT, & Rothman AJ (2006). Selective memory biases in individuals’ memory for health-related information and behavior recommendations. Psychology & Health, 21, 247–272. 10.1080/14768320500098715 [DOI] [PubMed] [Google Scholar]
  17. Ko LK, Campbell MK, Lewis MA, Earp J, & DeVellis B (2011) Information processes mediate the effect of a health communication intervention on fruit and vegetable consumption. Journal of Health Communication, 16, 282–299. 10.1080/10810730.2010.532294 [DOI] [PMC free article] [PubMed] [Google Scholar]
  18. Kreuter MW, Bull FC, Clark EM, & Oswald DL (1999). Understanding how people process health information: a comparison of tailored and nontailored weight-loss materials. Health Psychology, 18, 487–494. 10.1037/0278-6133.18.5.487 [DOI] [PubMed] [Google Scholar]
  19. Larson HJ (2018). The biggest pandemic risk? Viral misinformation. Nature, 562, 309–310. 10.1038/d41586-018-07034-4 [DOI] [PubMed] [Google Scholar]
  20. Ng JYY, Ntoumanis N, Thøgersen-Ntoumani C, Deci EL, Ryan RM, Duda JL, & Williams GC (2012). Self-determination theory applied to health contexts: A meta-analysis. Perspectives on Psychological Science, 7, 325–340. 10.1177/1745691612447309 [DOI] [PubMed] [Google Scholar]
  21. Nielsen L, Riddle M, King JW, Aklin WM, Chen W, Clark D, … Weber W (2018). The NIH Science of Behavior Change Program: Transforming the science through a focus on mechanisms of change. Behaviour Research and Therapy, 101, 3–11. 10.1016/j.brat.2017.07.002 [DOI] [PMC free article] [PubMed] [Google Scholar]
  22. Patten E (2016) The nation’s Latino population is defined by its youth. Pew Research Center. Retrieved from https://www.pewresearch.org/hispanic/wp-content/uploads/sites/5/2016/04/PH_2016-04-20_LatinoYouth-Final.pdf. [Google Scholar]
  23. Petty RE, & Cacioppo JT (1986). The elaboration likelihood model of persuasion. In Petty RE & Cacioppo JT (Eds.), Communication and Persuasion: Central and Peripheral Routes to Attitude Change (pp. 1–24). 10.1007/978-1-4612-4964-1_1 [DOI] [Google Scholar]
  24. Petty RE, & Wegener DT (1998). Matching versus mismatching attitude functions: Implications for scrutiny of persuasive messages. Personality and Social Psychology Bulletin, 24, 227–240. 10.1177/0146167298243001 [DOI] [Google Scholar]
  25. Powell BJ, Waltz TJ, Chinman MJ, Damschroder LJ, Smith JL, Matthieu MM, Proctor EK, & Kirchner JE (2015). A refined compilation of implementation strategies: Results from the expert recommendation for implementing change (ERIC) project. Implementation Science, 10, 21. 10.1186/s13012-015-0209-1 [DOI] [PMC free article] [PubMed] [Google Scholar]
  26. Riddle M, & Science of Behavior Change Working Group (2015). News from the NIH: Using an experimental medicine approach to facilitate translational research. Translational Behavioral Medicine, 5, 486–488. 10.1007/s13142-015-0333-0 [DOI] [PMC free article] [PubMed] [Google Scholar]
  27. Rothman AJ, & Baldwin AS (2019). A person × intervention strategy approach to understanding health behavior. In Deaux K & Snyder M (Eds.), Handbook of personality and social psychology (2nd ed.; pp. 831–856). Oxford University Press. [Google Scholar]
  28. Ryan RM, Patrick H, Deci EL, & Williams GC (2008). Facilitating health behaviour change and its maintenance: Interventions based on self-determination theory. European Health Psychologist, 10, 2–5. [Google Scholar]
  29. Shay LA, Baldwin AS, Betts AC, Marks EG, Higashi RT, Street RL, … Tiro JA (2018). Parent-provider communication of HPV vaccine hesitancy. Pediatrics, 141, e20172312. 10.1542/peds.2017-2312 [DOI] [PMC free article] [PubMed] [Google Scholar]
  30. Sheeran P, Klein WMP, & Rothman AJ (2017). Health behavior change: Moving from observation to intervention. Annual Review of Psychology, 68, 573–600. 10.1146/annurev-psych-010416-044007 [DOI] [PubMed] [Google Scholar]
  31. Shelby A, & Ernst K (2013). Story and science: How providers and parents can utilize storytelling to combat anti-vaccine misinformation. Human Vaccines & Immunotherapeutics, 9, 1795–1801. 10.4161/hv.24828 [DOI] [PMC free article] [PubMed] [Google Scholar]
  32. Slamecka NJ, & Graf P (1978). The generation effect: Delineation of a phenomenon. Journal of Experimental Psychology: Human Learning and Memory, 4, 592–604. 10.1037/0278-7393.4.6.592 [DOI] [Google Scholar]
  33. Stice E, Marti CN, Spoor S, Presnell K, & Shaw H (2008). Dissonance and healthy weight eating disorder prevention programs: Long-term effects from a randomized efficacy trial. Journal of Consulting and Clinical Psychology, 76, 329–340. 10.1037/0022-006X.76.2.329 [DOI] [PMC free article] [PubMed] [Google Scholar]
  34. Stone J, Aronson E, Crain AL, Winslow MP, & Fried CB (1994). Inducing hypocrisy as a means of encouraging young adults to use condoms. Personality and Social Psychology Bulletin, 20, 116–128. 10.1177/0146167294201012 [DOI] [Google Scholar]
  35. Thompson EL, Rosen BL, Vamos CA, Kadono M, & Daley EM (2017). Human papillomavirus vaccination: What are the reasons for nonvaccination among U.S. adolescents? Journal of Adolescent Health, 61, 288–293. 10.1016/j.jadohealth.2017.05.015 [DOI] [PubMed] [Google Scholar]
  36. Tiro JA, Marks EG, Rochefort C, Fullington H, Rodriguez S, Zhu H, Lee SC, & Baldwin AS (2020). Factorial randomized controlled trial of parental self-persuasion for adolescent HPV vaccination. Manuscript under review. [Google Scholar]
  37. Tiro JA, Lee SC, Marks EG, Persaud D, Skinner CS, Street RL, … Baldwin AS (2016). Developing a tablet-based self-persuasion intervention promoting adolescent HPV vaccination: Protocol for a three-stage mixed-methods study. JMIR Research Protocols, 5, e19. 10.2196/resprot.5092 [DOI] [PMC free article] [PubMed] [Google Scholar]
  38. Walker TY, Elam-Evans LD, Yankey D, Markowitz LE, Williams CL, Fredua B, … Stokley S (2019). National, regional, state, and selected local area vaccination coverage among adolescents aged 13–17 years—United States, 2018. Morbidity and Mortality Weekly Report, 68, 718–723. 10.15585/mmwr.mm6833a2 [DOI] [PMC free article] [PubMed] [Google Scholar]
  39. Williams DM, Rhodes RE, & Conner MT (Eds.). (2018). Affective determinants of health behavior. Oxford University Press. [DOI] [PubMed] [Google Scholar]
  40. Williams GC, McGregor HA, Sharp D, Levesque C, Kouides RW, Ryan RM, & Deci EL (2006). Testing a self-determination theory intervention for motivating tobacco cessation: Supporting autonomy and competence in a clinical trial. Health psychology, 25, 91–101. 10.1037/0278-6133.25.1.91 [DOI] [PubMed] [Google Scholar]
  41. World Health Organization (2019). Ten threats to global health in 2019. https://www.who.int/news-room/spotlight/ten-threats-to-global-health-in-2019
  42. World Health Organization (2020). Immunizing the public against misinformation. https://www.who.int/news-room/feature-stories/detail/immunizing-the-public-against-misinformation

Associated Data

This section collects any data citations, data availability statements, or supplementary materials included in this article.

Supplementary Materials

Supplemental Material 1
NIHMS1727590-supplement-Supplemental_Material_1.pdf (129.3KB, pdf)

RESOURCES