Division of Endocrinology and Metabolism, Department of Medicine, University of Alberta, Edmonton, AB, Canada.
Alberta Diabetes Institute, University of Alberta, Edmonton, AB, Canada.
PLoS One. 2022 Aug 11;17(8):e0271096. doi: 10.1371/journal.pone.0271096. eCollection 2022.
User designed Automated Insulin Delivery systems (AID), termed Do-It-Yourself (DIY) AID include; AndroidAPS, OpenAPS and Loop. These unregulated systems provide challenges for healthcare providers worldwide, with potential legal and ethical barriers to supporting their use. We performed a scoping review of the currently available literature surrounding DIY AID systems, specifically to highlight the evidence available to facilitate healthcare providers to support persons with diabetes who may benefit from DIY AID.
Studies relating to DIY AID systems were searched in Embase, Medline, Web of Science, Scopus, Proquest and Cochrane library until 31st December 2021. Publications were screened through title and abstract to identify study type and AID system type described. A thematic synthesis methodology was used for analysis of studies of DIY AID use due to the heterogeneity in study designs (case reports, qualitative, cross-sectional and cohort studies), with similarity in outcome themes.
Following implementation of the search strategy, 38 relevant full texts were identified; comprising 12 case reports, 9 qualitative studies and 17 cohort studies, and data was also available from 24 relevant conference abstracts. No randomized studies were identified. Common themes were identified in the outcomes across the studies; glycemic variability, safety, quality of life, healthcare provider attitudes and social media.
There is extensive real-world data, but a lack of randomized control trial evidence supporting DIY AID system use, due to the user-driven, unregulated nature of these systems. Healthcare providers report a lack of understanding surrounding, and confidence in supporting, DIY AID despite impressive observational and user self-reported improvements in glycemic variability, without any reported safety compromises.
用户设计的自动化胰岛素输送系统（AID），称为 DIY AID，包括 AndroidAPS、OpenAPS 和 Loop。这些不受监管的系统为全球医疗保健提供者带来了挑战，在支持其使用方面存在潜在的法律和道德障碍。我们对围绕 DIY AID 系统的现有文献进行了范围综述，特别是为了强调可用于帮助可能从 DIY AID 中受益的糖尿病患者的医疗保健提供者的现有证据。
在 Embase、Medline、Web of Science、Scopus、Proquest 和 Cochrane 图书馆中搜索与 DIY AID 系统相关的研究，直到 2021 年 12 月 31 日。通过标题和摘要筛选出版物，以确定描述的研究类型和 AID 系统类型。由于研究设计（病例报告、定性、横断面和队列研究）存在异质性，而结局主题相似，因此采用主题综合方法分析 DIY AID 使用的研究。
实施搜索策略后，确定了 38 篇相关的全文，包括 12 篇病例报告、9 项定性研究和 17 项队列研究，还可以从 24 篇相关会议摘要中获得数据。未确定随机研究。研究中的结果确定了共同的主题；血糖变异性、安全性、生活质量、医疗保健提供者态度和社交媒体。
由于这些系统是用户驱动的、不受监管的，因此有大量的实际数据，但缺乏支持 DIY AID 系统使用的随机对照试验证据。尽管观察性研究和用户自我报告的血糖变异性改善令人印象深刻，而且没有报告安全性妥协，但医疗保健提供者报告说，他们对 DIY AID 的理解不足，并且缺乏支持 DIY AID 的信心。
