Computer-Animated Relational Agents in Human Papillomavirus Vaccination Education

We conducted a pilot nonrandomized trial with participants recruited first to the relational agent and then to the control arm from the gynecologic oncology clinic between November 18, 2020, and February 5, 2021. Eligible participants had a diagnosis of cervical or vaginal cancer or precancer, were aged 18 years or older, and spoke English. The study was approved by the Wayne State University Institutional Review Board. The relational agent features an animated character, viewed on a preloaded tablet, that simulates face-to-face health counseling, 3 including basic education on HPV and HPV vaccination and brief motivational interviewing. 4 The control handout, developed for this study, provided the same basic information as the relational agent in a brochure format. Preintervention, postintervention, and 6-week follow-up intervention surveys were developed based on the Theory of Planned Behavior. 5 , 6 Responses were measured on a 4- or 7-point Likert scale. We recruited 30 participants (15 per group) for a medium effect size (0.5) for the future main trial, with 90% power at a two-sided 5% level. 7 As a pilot study, no formal statistical testing was carried out and all outcomes were descriptively summarized. Preintervention to postintervention change was compared using the Wilcoxon test–based effect size 8 for continuous variables.

Mean age of participants was 43.5 years (range 26–72 years). There was a 96.7% acceptability rate. Responding with a Likert scale score of 6 (satisfied) or 7 (very satisfied) for relatability, 14 of 15 (93%) relational agent participants liked the way the information was presented, 12 of 15 (80%) trusted the relational agent, and 13 of 15 (87%) found it easy to talk to the relational agent. On average, participants spent 12–15 minutes on the relational agent intervention. Theory of Planned Behavior domains of attitude toward HPV vaccination, 9 , 10 subjective norms, 11 , 12 perceived behavior control, 12 and intention to discuss HPV vaccination 13 , 14 were assessed in the preintervention and postintervention and follow-up surveys ( Table 1 ). For the Theory of Planned Behavior domain of behavioral control, patients in the relational agent group were more confident that they would be able to talk to their family members about HPV vaccination in the next 6 weeks (r=0.60). Regarding our primary outcome of intention to discuss HPV vaccination, relational agent participants were more likely to intend to discuss HPV vaccination after receiving the intervention than were control participants (r=0.66). Follow-up surveys were returned by 21 of 30 (70%) participants, 10 in the control group and 11 in the interactive HPV vaccination education program group. Nine (81%) relational agent participants and seven (70%) control participants reported that they discussed HPV vaccination with their family or friends. Of these, three (33%) relational agent participants and two (29%) control participants knew that their family or friends had subsequently discussed HPV vaccination with their health care professionals or received the HPV vaccine.

We found that using a computer-animated relational agent to educate participants about HPV and vaccination was a feasible and acceptable intervention among patients with cervical or vaginal precancer or cancer. The relational agent intervention facilitated various Theory of Planned Behavior constructs, most importantly the participants’ intention to discuss HPV vaccination with their family and community. Future studies will include a more ethnically diverse population; patients with other HPV-related diseases, including head and neck cancers; and nonaffected populations. We suggest that using interactive HPV vaccination education programs conducted by a relational agent should be tested in larger studies to determine whether it increases participants’ intention to discuss HPV vaccination and, ultimately, HPV vaccination rates.

Despite current prevention efforts, human papillomavirus (HPV) vaccination rates in the United States remain lower than in many other well-resourced countries 1 , 2 To reach the goal of eradicating HPV-associated disease, novel education strategies are needed. Our study explored the acceptability of an innovative approach using an interactive HPV vaccination education program administered by a computer-animated relational agent.