Patient education and counseling · 2026
Symptom checker formats to improve symptom management knowledge and trust in AI-enhanced advice: Qualitative insights.
СтатьяНазвание и аннотация: Перевод в очереди
Обработка статьи · этапы, ошибки и управление
Загрузка состояния…
Ошибки и ограничения
История · было → стало
Состояние обновляется каждые 5 секунд, пока блок открыт. Процент показывается только там, где известен объём работы. Пауза останавливает продолжение этапа; текущий запрос к источнику может завершиться.
Ayre Julie, Hudson Claire, Hudson Barton Momo, Gallagher Karen, Smith Jenna, Bamgboje-Ayodele Adeola, Scott Angelica, Woods Alison, Ng Christopher, Wickramasinghe Yuanee, Ma Ivan C K, Edlund Geoffrey, Butters Lorna, Nadesan Waren, Vu Trang, Kapoor Gauri, McCaffery Kirsten J
Аннотация
OBJECTIVE: Research consistently shows online symptom checkers could better meet the needs of diverse users. This study explored users' experiences of a symptom checker that had artificial intelligence (AI)-enhanced features.
METHODS: We recruited participants from an online trial that evaluated varied symptom checker formats for healthdirect's online symptom checker. The symptom checker formats included the standard version (existing tool), and AI-enhanced versions, which provided a rationale for the advice, more tailored, plain language symptom management advice, and links to relevant sources. Participants were invited to take part in cognitive interviews to provide further qualitative feedback. We used Framework analysis to generate themes.
RESULTS: Twenty-two people completed interviews. The first theme explored how people relied on cues such as links, government logos, and statements about clinical oversight, to make decisions about trusting the AI advice. The second theme described how people were more trusting of advice that matched their expectations. For unexpected or more urgent advice, participants emphasised the importance of clear explanations and links to further reading. The third theme described participants' preferences for clear and concrete steps for managing symptoms, whilst also needing to keep triage advice prominent to encourage appropriate care-seeking.
CONCLUSIONS: This study suggests using AI to improve online symptom checker content may help address barriers to their uptake and acceptability. More research is needed to ensure that symptom checkers keep pace with popular general-purpose AI tools, including evaluation of a range of designs using higher-fidelity prototypes.
PRACTICAL IMPLICATIONS: Symptom checker developers must carefully develop AI-enhanced designs that maintain trust, provide users with ways to verify information, and support appropriate action.
METHODS: We recruited participants from an online trial that evaluated varied symptom checker formats for healthdirect's online symptom checker. The symptom checker formats included the standard version (existing tool), and AI-enhanced versions, which provided a rationale for the advice, more tailored, plain language symptom management advice, and links to relevant sources. Participants were invited to take part in cognitive interviews to provide further qualitative feedback. We used Framework analysis to generate themes.
RESULTS: Twenty-two people completed interviews. The first theme explored how people relied on cues such as links, government logos, and statements about clinical oversight, to make decisions about trusting the AI advice. The second theme described how people were more trusting of advice that matched their expectations. For unexpected or more urgent advice, participants emphasised the importance of clear explanations and links to further reading. The third theme described participants' preferences for clear and concrete steps for managing symptoms, whilst also needing to keep triage advice prominent to encourage appropriate care-seeking.
CONCLUSIONS: This study suggests using AI to improve online symptom checker content may help address barriers to their uptake and acceptability. More research is needed to ensure that symptom checkers keep pace with popular general-purpose AI tools, including evaluation of a range of designs using higher-fidelity prototypes.
PRACTICAL IMPLICATIONS: Symptom checker developers must carefully develop AI-enhanced designs that maintain trust, provide users with ways to verify information, and support appropriate action.