AI Chatbot Intervention Could Improve Pneumococcal Vaccination Uptake in Older Adults

Results from a randomized controlled trial (NCT05772117) published by investigators in the Journal of the American Medical Association indicate that an online hybrid chatbot was more effective than a chatbot-delivered standard intervention in increasing pneumococcal vaccination uptake among community-dwelling older adults in Hong Kong, demonstrating potential for its use as a sustainable approach for promoting vaccination in this population.^1,2

How Can Chatbots Improve Pneumococcal Vaccination Uptake?

Older adults are at increased susceptibility to invasive pneumococcal disease and other severe complications if they contract Streptococcus pneumoniae (S pneumoniae). Fortunately, pneumococcal vaccines are available in the United States and worldwide and are recommended by immunization practices groups for older adults, especially those with comorbidities. Pneumococcal vaccines are effective and protect against over a dozen serotypes of S pneumoniae.³

Despite its effectiveness, countless at-risk individuals do not receive a pneumococcal vaccine. For example, in Hong Kong, only 35.3% of individuals aged 65 years and older received a vaccine, despite them being offered for free. This limited uptake requires rethinking established methods for increasing vaccination coverage, including utilizing evolving technologies that provide more tailored interventions for individuals.⁴

Chatbots, which are computer programs that simulate an authentic conversation with a user, have demonstrated the ability to reach target patients and increase uptake, especially when utilizing artificial intelligence (AI). The current authors assessed whether a hybrid chatbot that combines a rule-based system and AI-based components could increase pneumococcal vaccination uptake. According to the investigators, the hybrid model allowed for carefully designed interventions to be accurately delivered while providing AI-powered question and answer (Q&A) sessions.¹

“This effective engagement strategy enables chatbots to better understand users’ open-ended questions and to generate real-time answers,” the study authors wrote, saying that this combination was novel for a vaccine chatbot. In the randomized trial, investigators compared efficacy of the hybrid chatbot (the stage of change group) versus a standard chatbot delivered without a Q&A function (the standard intervention group) in older Hong Kong residents aged 65 and older.¹

How Was the Chatbot Designed?

For the intervention group, participants’ stage of change was assessed regarding pneumococcal vaccine uptake, and they were delivered codesigned stage of change-tailored interventions in 4 monthly sessions at months 0, 1, 2, and 3. At the beginning of each chatbot session, participants were asked whether they intended to receive the pneumococcal vaccine within the next year and whether they planned to do so within the next month. During the second session, the chatbot asked to confirm whether the participants had received the vaccine.¹

Depending on their response and stage of change, they were provided an online video that either aimed to increase their awareness of the vaccine’s effectiveness or increase self-efficacy and perceived benefits related to the vaccine, or they were given information about locations providing free pneumococcal vaccination and resources. The latter stage aimed to help individuals develop an action plan for vaccination.¹

Was Vaccine Uptake Improved?

In total, 374 individuals completed the baseline survey and were randomized to the stage of change or standard intervention groups. At 12 months following completion of the intervention, the dropout rates were 16.6% and 13.4% in each group, respectively. Ultimately, 55 stage of change participants and 35 standard intervention participants received a pneumococcal vaccine during the study period.¹

An intention-to-treat analysis revealed that pneumococcal vaccine uptake in the stage of change group was significantly higher than in the standard intervention group (29.4% vs 18.7%; relative risk, 1.57 [95% CI, 1.08–2.28]; absolute risk reduction, 0.11 [95% CI, 0.02–0.19]). Accordingly, as documented by the chatbot, more individuals in the stage of change group than in the standard intervention group completed at least 1 intervention episode (79.7% vs 57.8%; P < .001).¹

“The chatbot-delivered interventions addressed some main modifiable barriers to uptake of PV among older adults, including lack of awareness about benefits and low self-efficacy, and concerns about adverse effects, high cost, and inconvenience related to PV,” the study authors wrote in their discussion. Use of chatbot- and AI-based interventions to increase pneumococcal vaccination uptake could be feasible and effective in pharmacy settings in the United States and other countries around the world.¹

REFERENCES

1. Wang Z, Chen S, Poon J, et al. A hybrid chatbot to promote pneumococcal vaccination among older adults: a randomized clinical trial. JAMA Netw Open. 2025;8(10):e2535813. doi:10.1001/jamanetworkopen.2025.35813

2. A RCT evaluating an online intervention in increasing pneumococcal vaccination among older adults. ClinicalTrials.gov Identifier: NCT05772117. Last updated April 6, 2023. https://clinicaltrials.gov/study/NCT05772117?term=NCT05772117&rank=1

3. Halpern L. Meta-analysis: Pneumococcal vaccination reduces hospitalization, mortality in older adults. Pharmacy Times. Published October 3, 2025. Accessed October 10, 2025. https://www.pharmacytimes.com/view/meta-analysis-pneumococcal-vaccination-reduces-hospitalization-mortality-in-older-adults

4. Chan PS, Poon J, Han SC, et al. Changes in the pneumococcal vaccination iptake and its determinants before, during, and after the COVID-19 pandemic among community-living older adults in Hong Kong, China: repeated random telephone surveys. Vaccines (Basel). 2024;12(8):894. doi: 10.3390/vaccines12080894