Development and feasibility testing of an artificially intelligent chatbot to answer immunization-related queries of caregivers in Pakistan: A mixed-methods study

Abstract: Objective “Zero-dose children” remain a crucial burden to global health, particularly in low- and middle-income countries (LMICs) where vulnerable children have limited access to immunizations. This review explores the transformative potential of artificial intelligence (AI) in addressing this challenge and examines how AI can enhance routine childhood immunization. Methodology This narrative review synthesizes literature on AI, digital health innovations, vaccine delivery, and public health informatics. Literature was sourced from PubMed, Scopus, and Google Scholar, as well as institutional and authoritative reports such as the World Health Organization, the United Nations Children's Fund (UNICEF), and Gavi, the Vaccine Alliance. Results AI applications have demonstrated utility in identifying zero-dose populations, optimizing vaccine delivery, and supporting data-driven decision-making. Tools such as digital health passports, predictive analytics, and real-time monitoring platforms (e.g. UNICEF's real-time vaccination monitoring and analysis) enhance immunization tracking and reduce inequities. AI-driven chatbots, mobile applications, and social listening tools address vaccine hesitancy by providing tailored communication and combating misinformation. Furthermore, geographic information systems (GISs) and AI-based behavioral analysis improve outreach to underserved populations. However, challenges remain, including data quality, scalability, and ethical considerations. Conclusion AI solutions allow efficient identification of potential pockets of non-vaccinated populations, enhance decision-making processes based on data, and reduce vaccine skepticism due to AI-based interventions. Examples of successful implementation of AI include digital health passports, continuous monitoring mechanisms, and GISs for vaccine administration and coverage. This review also discusses the challenges and ethical considerations associated with AI implementation in LMICs, as well as the ethical implications. Using the insights derived from the review, this paper calls for the targeted utilization of AI toward the realization of a healthier future for children in need to achieve the target of reducing the number of zero-dose children by 50% by the year 2030, not only in areas afflicted by immunization disparities but across the globe.

Abstract: Importance There are few robust evaluations assessing the efficacy of chatbots to improve pneumococcal vaccination (PV) uptake among adults 65 years of age or older. Objective To evaluate the relative efficacy of a hybrid chatbot in increasing PV uptake among Hong Kong residents aged 65 years or older. Design, Setting, and Participants This partially masked, parallel-group randomized clinical trial was conducted between May 1, 2023, and November 30, 2024 in Hong Kong, China. Participants were aged 65 years or older, had a Hong Kong identity card, could speak and comprehend Cantonese, were smartphone and WhatsApp users, and had no prior PV uptake. Participants were recruited through random telephone calls and were randomized to either the stage of change group or the standard intervention group. Interventions In the stage of change group, the rule-based component of the hybrid chatbot assessed participants’ stage of change regarding PV uptake and then delivered stage of change–tailored interventions at months 0, 1, 2, and 3. The natural language processing component of the hybrid chatbot provided real-time answers to participants’ PV-related questions. In the standard intervention group, the chatbot sent participants a link to access a standard online video covering PV information at months 0, 1, 2, and 3. Main Outcomes and Measures The primary outcome was self-reported PV uptake at month 12, which was validated by the research team. The secondary outcome was participants’ stage of change measured at month 0 and month 12 by using validated questions, with a score of 1 = precontemplation, 2 = contemplation, 3 = preparation, and 4 = action. Results A total of 374 participants (213 female [57.0%]; mean [SD] age, 69.6 [3.1] years) were randomized to either the stage of change group (n = 187) or the standard intervention group (n = 187). The intention-to-treat analysis showed that the validated PV uptake rate was higher in the stage of change group than in the standard intervention group (29.4% vs 18.7%; P = .01). The mean (SD) stage of change score was higher in the stage of change group than in the standard intervention group (2.2 [1.3] vs 1.9 [1.1]; P = .02). More participants in the stage of change group than in the standard intervention group completed at least 1 intervention session (79.7% vs 57.8%; P < .001). Conclusions and Relevance In this randomized clinical trial, the hybrid chatbot was more efficacious than the standard intervention in increasing PV uptake among older adults in Hong Kong. A hybrid chatbot may be a sustainable PV promotion for older adults. Trial Registration ClinicalTrials.gov Identifier: NCT05772117

TL;DR: Focus of immunization microplans at the subdistrict level are needed to achieve universal immunization goals.

TL;DR: The child immunization card is an inexpensive yet effective instrument for systematically recording the vaccines received by a child but is too often underutilized or inappropriately used by parents and health care workers and therefore does not always fulfil its intended purpose.

TL;DR: The findings add to the existing literature on innovative, low cost reminders for health and make several suggestions for enhancing practical implementation of these tools.

TL;DR: In this article , the authors evaluated the efficacy of ChatGPT, an advanced natural language processing model, in adapting and synthesizing clinical guidelines for diabetic ketoacidosis (DKA) by comparing and contrasting different guideline sources.