背景:聊天机器人是一种计算机程序,旨在模拟与人类的对话。聊天机器人可能会提供快速,响应,和私人避孕信息;咨询;以及与产品和服务的联系,这可以提高避孕知识,态度,和行为。
目的:这篇综述旨在系统地整理和解释证据,以确定聊天机器人是否以及如何提高避孕知识,态度,和行为。避孕知识,态度,行为包括获取避孕信息,了解避孕信息,获得避孕服务,避孕吸收,避孕延续,和避孕沟通或谈判技巧。审查的第二个目的是确定和总结聊天机器人开发的最佳实践建议,以改善避孕效果。包括有证据的聊天机器人的成本效益。
方法:我们系统地搜索了同行评审和灰色文献(2010-2022年),以获取评估提供避孕信息和服务的聊天机器人的论文。如果他们以聊天机器人为特色并解决避孕问题,例如,激素避孕药的摄取。使用适当的质量评估工具评估文献的方法学质量。使用数据提取框架从包含的来源中提取数据。使用叙事综合方法来整理定性证据,因为定量证据太稀疏,无法进行定量综合。
结果:我们确定了15个来源,包括8篇原创研究论文和7篇灰色文献论文。这些来源包括16个独特的聊天机器人。这篇综述发现了以下关于聊天机器人的影响和功效的证据:强有力的随机对照试验表明,聊天机器人对使用避孕药的意图没有影响;一个小的,不受控制的队列研究表明,青春期女孩对避孕的吸收增加;和一份发展报告,使用低质量的方法,这表明对改善服务访问没有影响。也有低质量的证据表明,通过与聊天机器人内容的互动,避孕知识会增加。用户参与度参差不齐,一些聊天机器人吸引了广泛的受众,另一些则吸引了非常小的受众。用户反馈表明,聊天机器人的体验可能是可以接受的,方便,匿名,私人,但也不称职,不方便,和无情。关于开发聊天机器人以提高避孕知识的最佳实践指导,态度,行为与其他医疗保健领域的聊天机器人文献一致。
结论:我们发现关于聊天机器人提高避孕知识的证据有限且相互矛盾,态度,和行为。与替代技术相比,进一步研究了聊天机器人干预的影响,承认聊天机器人干预的多样性和不断变化的性质,并寻求确定与改善避孕效果相关的关键特征是必要的。这项审查的局限性包括关于这一主题的可用证据有限,缺乏对该领域聊天机器人的正式评估,以及缺乏对聊天机器人的标准化定义。
BACKGROUND: A chatbot is a computer program that is designed to simulate conversation with humans. Chatbots may offer rapid, responsive, and private contraceptive information; counseling; and linkages to products and services, which could improve contraceptive knowledge, attitudes, and behaviors.
OBJECTIVE: This
review aimed to systematically collate and interpret evidence to determine whether and how chatbots improve contraceptive knowledge, attitudes, and behaviors. Contraceptive knowledge, attitudes, and behaviors include access to contraceptive information, understanding of contraceptive information, access to contraceptive services, contraceptive uptake, contraceptive continuation, and contraceptive communication or negotiation skills. A secondary aim of the
review is to identify and summarize best practice recommendations for chatbot development to improve contraceptive outcomes, including the cost-effectiveness of chatbots where evidence is available.
METHODS: We systematically searched peer-reviewed and gray literature (2010-2022) for papers that evaluated chatbots offering contraceptive information and services. Sources were included if they featured a chatbot and addressed an element of contraception, for example, uptake of hormonal contraceptives. Literature was assessed for methodological quality using appropriate quality assessment tools. Data were extracted from the included sources using a data extraction framework. A narrative synthesis approach was used to collate qualitative evidence as quantitative evidence was too sparse for a quantitative synthesis to be carried out.
RESULTS: We identified 15 sources, including 8 original research papers and 7 gray literature papers. These sources included 16 unique chatbots. This
review found the following evidence on the impact and efficacy of chatbots: a large, robust randomized controlled trial suggests that chatbots have no effect on intention to use contraception; a small, uncontrolled cohort study suggests increased uptake of contraception among adolescent girls; and a development report, using poor-quality methods, suggests no impact on improved access to services. There is also poor-quality evidence to suggest increased contraceptive knowledge from interacting with chatbot content. User engagement was mixed, with some chatbots reaching wide audiences and others reaching very small audiences. User feedback suggests that chatbots may be experienced as acceptable, convenient, anonymous, and private, but also as incompetent, inconvenient, and unsympathetic. The best practice guidance on the development of chatbots to improve contraceptive knowledge, attitudes, and behaviors is consistent with that in the literature on chatbots in other health care fields.
CONCLUSIONS: We found limited and conflicting evidence on chatbots to improve contraceptive knowledge, attitudes, and behaviors. Further research that examines the impact of chatbot interventions in comparison with alternative technologies, acknowledges the varied and changing nature of chatbot interventions, and seeks to identify key features associated with improved contraceptive outcomes is needed. The limitations of this
review include the limited evidence available on this topic, the lack of formal evaluation of chatbots in this field, and the lack of standardized definition of what a chatbot is.