Zengul Ayse, Evans Eric, Hall Allyson, Qu Haiyan, Willig Amanda, Cherrington Andrea, Thirumalai Mohanraj
Department of Nutrition Sciences, University of Alabama at Birmingham, Birmingham, AL, United States.
Department of Health Services Administration, University of Alabama at Birmingham, Birmingham, AL, United States.
JMIR Res Protoc. 2021 Sep 10;10(9):e31695. doi: 10.2196/31695.
Diabetes mellitus is a major health problem among people with physical disabilities. Health coaching has been proven to be an effective approach in terms of behavioral changes, patient self-efficacy, adherence to treatment, health service use, and health outcomes. Telehealth systems combined with health coaching have the potential to improve the quality of health care by increasing access to services. Treatment fidelity is particularly important for behavior change studies; however, fidelity protocols are inadequately administered and reported in the literature.
The aim of this study is to outline all the intervention fidelity strategies and procedures of a telecoaching intervention-artificial intelligence for diabetes management (AI4DM)-which is a randomized controlled trial to evaluate the feasibility, acceptability, and preliminary efficacy of a telehealth platform in adults with type 2 diabetes and permanent impaired mobility. AI4DM aims to create a web-based disability-inclusive diabetes self-management program. We selected the National Institutes of Health Behavior Change Consortium (NIH BCC) fidelity framework to describe strategies to ensure intervention fidelity in our research.
We have developed fidelity strategies based on the five fidelity domains outlined by the NIH BCC-focusing on study design, provider training, treatment delivery, treatment receipt, and enactment of treatment skills. The design of the study is grounded in the social cognitive theory and is intended to ensure that both arms would receive the same amount of attention from the intervention. All providers will receive standardized training to deliver consistent health coaching to the participants. The intervention will be delivered through various controlling and monitoring strategies to reduce differences within and between treatment groups. The content and structure of the study are delivered to ensure comprehension and participation among individuals with low health literacy. By constantly reviewing and monitoring participant progress and protocol adherence, we intend to ensure that participants use cognitive and behavioral skills in real-world settings to engage in health behavior.
Enrollment for AI4DM will begin in October 2021 and end in October 2022. The results of this study will be reported in late 2022.
Developing and using fidelity protocols in behavior change studies is essential to ensure the internal and external validity of interventions. This study incorporates NIH BCC recommendations into an artificial intelligence embedded telecoaching platform for diabetes management designed for people with physical disabilities. The developed fidelity protocol can provide guidance for other researchers conducting telehealth interventions within behavioral health settings to present more consistent and reproducible research.
ClinicalTrials.gov NCT04927377; http://clinicaltrials.gov/ct2/show/NCT04927377.
INTERNATIONAL REGISTERED REPORT IDENTIFIER (IRRID): PRR1-10.2196/31695.
糖尿病是身体残疾者面临的一个主要健康问题。健康指导已被证明在行为改变、患者自我效能感、治疗依从性、医疗服务利用和健康结果方面是一种有效的方法。远程医疗系统与健康指导相结合，有可能通过增加服务可及性来提高医疗质量。治疗保真度对于行为改变研究尤为重要；然而，保真度方案在文献中的管理和报告并不充分。
本研究的目的是概述远程指导干预——糖尿病管理人工智能（AI4DM）的所有干预保真度策略和程序，这是一项随机对照试验，旨在评估远程医疗平台在2型糖尿病且永久性行动不便的成年人中的可行性、可接受性和初步疗效。AI4DM旨在创建一个基于网络的、包容残疾人士的糖尿病自我管理项目。我们选择了美国国立卫生研究院行为改变联盟（NIH BCC）的保真度框架来描述确保我们研究中干预保真度的策略。
我们根据NIH BCC概述的五个保真度领域制定了保真度策略，重点关注研究设计、提供者培训、治疗实施、治疗接受以及治疗技能的运用。该研究的设计基于社会认知理论，旨在确保两组都能从干预中获得同等程度的关注。所有提供者将接受标准化培训，以便为参与者提供一致的健康指导。干预将通过各种控制和监测策略来实施，以减少治疗组内部和之间的差异。研究的内容和结构经过精心设计，以确保健康素养较低的个体能够理解并参与其中。通过不断审查和监测参与者的进展及方案依从性，我们旨在确保参与者在现实环境中运用认知和行为技能来参与健康行为。
AI4DM的招募将于2021年10月开始，2022年10月结束。本研究结果将于202年年底报告。
在行为改变研究中制定和使用保真度方案对于确保干预的内部和外部有效性至关重要。本研究将NIH BCC的建议纳入了一个为身体残疾者设计的、嵌入人工智能的糖尿病管理远程指导平台。所制定的保真度方案可为其他在行为健康环境中进行远程医疗干预的研究人员提供指导，以呈现更一致且可重复的研究。
ClinicalTrials.gov NCT04927377；http://clinicaltrials.gov/ct2/show/NCT04927377。
国际注册报告识别码（IRRID）：PRR1-10.2196/31695。
