Swallow Veronica, Horsman Janet, Mazlan Eliza, Campbell Fiona, Zaidi Reza, Julian Madeleine, Branchflower Jacob, Martin-Kerry Jackie, Monks Helen, Soni Astha, Rodriguez Alison, Julian Rob, Dimitri Paul
Centre for Applied Health & Social Care Research, Sheffield Hallam University, Howard St, Sheffield City Centre, Sheffield, Sheffield, S1 1WB, United Kingdom, +44 (0)114 225 5555.
Children's Services, Leeds Teaching Hospitals NHS Trust, Leeds, United Kingdom.
JMIR Diabetes. 2025 Jul 23;10:e74032. doi: 10.2196/74032.
Transition to adult health care for young people and young adults (YP/YA) with type 1 diabetes mellitus (T1DM) starts around 11 years of age, but transition services may not meet their needs. A combination of self-management support digital health technologies exists, but no supportive chatbots with components to help YP/YA with T1DM were identified.
The aims of this study were to (1) evaluate the novel DigiBete Chatbot, the first user-led, developmentally appropriate, clinically approved transition chatbot for YP/YA with T1DM from four English diabetes services and (2) assess the feasibility of a future trial of the chatbot.
In a prospective, multimethod, nonrandomized feasibility and acceptability study in the UK National Health Service, YP/YA with T1DM from 4 hospital diabetes clinics (2 pretransition and 2 posttransition) were enrolled in a 6-week study to test the DigiBete Chatbot. During the study, YP/YA completed web-based, validated, and standardized questionnaires at baseline, 2 weeks, and 6 weeks to evaluate quality of life and anxiety and depression, along with chatbot usability and acceptability. Qualitative interviews involving YP/YA, parents, and health care professionals explored their views on the chatbot. Data were analyzed using descriptive statistics and framework analysis.
Eighteen YP/YA were enrolled. Qualitative interviews were conducted with 4 parents, 24 health care professionals, and 12 YP/YA. Questionnaire outputs and the emergent qualitative themes (living with T1DM, using the chatbot, and refining the chatbot) indicated that the measures are feasible to use and the chatbot is acceptable and functional. In addition, responses indicated that, with refinements that incorporate the feasibility results, the chatbot could beneficially support YP/YA during transition. Users scored the chatbot as "good" to "excellent" for being engaging, informative, and aesthetically pleasing, and they stated that they would use it again. The results suggest that, with some adaptations based on user feedback, the chatbot was feasible and acceptable among the YP/YA who enjoyed using it. Our reactive conversational agent offers content (messaging and additional multimedia resources) that is relevant for the target population and clinically approved. The DigiBete Chatbot addresses the identified lack of personalized and supported self-management tools available for 11-24 year olds with T1DM and other chronic conditions.
These results warrant chatbot refinement and further investigation in a full trial to augment it prior to its wider clinical use. Our research design and methodology could also be transferred to using chatbots for other long-term conditions. On the premise of this feasibility study, the plan is to rebuild the DigiBete Chatbot to meet identified user needs and preferences and progress to a national cohort study to assess the usability, feasibility, and acceptability of a modified chatbot, with a view to proceeding to rollout for national and international use on the established DigiBete platform.
1型糖尿病（T1DM）青少年及青年（YP/YA）向成人医疗保健的过渡始于11岁左右，但过渡服务可能无法满足他们的需求。虽然存在多种自我管理支持数字健康技术，但尚未发现具备帮助T1DM患者的YP/YA所需组件的支持性聊天机器人。
本研究的目的是：（1）评估新型DigiBete聊天机器人，这是首个由用户主导、适合其发育阶段且经临床批准的针对来自四个英国糖尿病服务机构的T1DM患者的YP/YA过渡聊天机器人；（2）评估该聊天机器人未来试验的可行性。
在英国国家医疗服务体系进行的一项前瞻性、多方法、非随机的可行性和可接受性研究中，来自4家医院糖尿病诊所（2家处于过渡前阶段，2家处于过渡后阶段）的T1DM患者的YP/YA参加了一项为期6周的研究，以测试DigiBete聊天机器人。在研究期间，YP/YA在基线、2周和6周时完成基于网络的、经过验证且标准化的问卷，以评估生活质量、焦虑和抑郁情况，以及聊天机器人的可用性和可接受性。涉及YP/YA、家长和医疗保健专业人员的定性访谈探讨了他们对聊天机器人的看法。使用描述性统计和框架分析对数据进行分析。
招募了18名YP/YA。对4名家长、24名医疗保健专业人员和12名YP/YA进行了定性访谈。问卷结果和新出现的定性主题（与T1DM共存、使用聊天机器人以及优化聊天机器人）表明，这些措施使用起来是可行的，聊天机器人是可接受且实用的。此外，反馈表明，结合可行性结果进行优化后，聊天机器人可以在过渡期间对YP/YA提供有益支持。用户对聊天机器人的参与度、信息性和美观性给出了“良好”到“优秀”的评分，并表示会再次使用。结果表明，根据用户反馈进行一些调整后，该聊天机器人在喜欢使用它的YP/YA中是可行且可接受的。我们的反应式对话代理提供与目标人群相关且经临床批准的内容（消息和其他多媒体资源）。DigiBete聊天机器人解决了已发现的11 - 24岁T1DM及其他慢性病患者缺乏个性化且有支持的自我管理工具的问题。
这些结果表明有必要对聊天机器人进行优化，并在全面试验中进一步研究，以便在更广泛的临床应用之前对其进行改进。我们的研究设计和方法也可用于将聊天机器人应用于其他长期疾病。在这项可行性研究的基础上，计划重新构建DigiBete聊天机器人以满足已确定的用户需求和偏好，并推进到全国队列研究，以评估改进后聊天机器人的可用性、可行性和可接受性，以期在既定的DigiBete平台上进行全国和国际推广使用。
