PDF(Pubmed)
Abstract：
BACKGROUND: The proliferation of digital technology has the potential to transform diabetes management. One of the critical aspects of modern diabetes management remains the achievement of glycemic targets to avoid acute and long-term complications.
OBJECTIVE: This study aims to describe the landscape of evidence pertaining to the relative effectiveness or efficacy and safety of various digital interventions for the self-management of type 2 diabetes mellitus (T2DM), with a primary focus on reducing glycated hemoglobin A1c (HbA1c) levels.
METHODS: A systematic literature review (SLR) was conducted by searching Embase, MEDLINE, and CENTRAL on April 5, 2022. Study selection, data extraction, and quality assessment were performed by 2 independent reviewers. Eligibility criteria for the SLR included randomized controlled trials (RCTs) and comparative observational studies evaluating interventions containing both human (eg, coaching) and digital components (eg, glucose meter) in adult patients with T2DM. The primary meta-analysis was restricted to studies that reported laboratory-measured HbA1c. In secondary analyses, meta-regression was performed with the intensity of coaching in the digital intervention as a categorical covariate.
RESULTS: In total, 28 studies were included in this analysis. Most studies (23/28, 82%) used the reduction of HbA1c levels as the primary end point, either directly or as a part of a multicomponent outcome. In total, 21 studies reported statistically significant results with this primary end point. When stratified into 3 intervention categories by the intensity of the intervention supporting the digital health technology (analyzing all 28 studies), the success rate appeared to be proportional to the coaching intensity (ie, higher-intensity studies reported higher success rates). When the analysis was restricted to RCTs using the comparative improvement of HbA1c levels, the effectiveness of the interventions was less clear. Only half (12/23, 52%) of the included RCTs reported statistically significant results. The meta-analyses were broadly aligned with the results of the SLR. The primary analysis estimated a greater reduction in HbA1c associated with digital interventions compared with usual care (-0.31%, 95% CI -0.45% to -0.16%; P<.001). Meta-regression estimated reductions of -0.45% (95% CI -0.81% to -0.09%; P=.02), -0.29% (95% CI -0.48% to -0.11%; P=.003), and -0.28% (95% CI -0.65% to 0.09%; P=.20) associated with high-, medium-, and low-intensity interventions, respectively.
CONCLUSIONS: These findings suggest that reducing HbA1c levels in individuals with T2DM with the help of digital interventions is feasible, effective, and acceptable. One common feature of effective digital health interventions was the availability of timely and responsive personalized coaching by a dedicated health care professional.
OBJECTIVE: This study aims to describe the landscape of evidence pertaining to the relative effectiveness or efficacy and safety of various digital interventions for the self-management of type 2 diabetes mellitus (T2DM), with a primary focus on reducing glycated hemoglobin A1c (HbA1c) levels.
METHODS: A systematic literature review (SLR) was conducted by searching Embase, MEDLINE, and CENTRAL on April 5, 2022. Study selection, data extraction, and quality assessment were performed by 2 independent reviewers. Eligibility criteria for the SLR included randomized controlled trials (RCTs) and comparative observational studies evaluating interventions containing both human (eg, coaching) and digital components (eg, glucose meter) in adult patients with T2DM. The primary meta-analysis was restricted to studies that reported laboratory-measured HbA1c. In secondary analyses, meta-regression was performed with the intensity of coaching in the digital intervention as a categorical covariate.
RESULTS: In total, 28 studies were included in this analysis. Most studies (23/28, 82%) used the reduction of HbA1c levels as the primary end point, either directly or as a part of a multicomponent outcome. In total, 21 studies reported statistically significant results with this primary end point. When stratified into 3 intervention categories by the intensity of the intervention supporting the digital health technology (analyzing all 28 studies), the success rate appeared to be proportional to the coaching intensity (ie, higher-intensity studies reported higher success rates). When the analysis was restricted to RCTs using the comparative improvement of HbA1c levels, the effectiveness of the interventions was less clear. Only half (12/23, 52%) of the included RCTs reported statistically significant results. The meta-analyses were broadly aligned with the results of the SLR. The primary analysis estimated a greater reduction in HbA1c associated with digital interventions compared with usual care (-0.31%, 95% CI -0.45% to -0.16%; P<.001). Meta-regression estimated reductions of -0.45% (95% CI -0.81% to -0.09%; P=.02), -0.29% (95% CI -0.48% to -0.11%; P=.003), and -0.28% (95% CI -0.65% to 0.09%; P=.20) associated with high-, medium-, and low-intensity interventions, respectively.
CONCLUSIONS: These findings suggest that reducing HbA1c levels in individuals with T2DM with the help of digital interventions is feasible, effective, and acceptable. One common feature of effective digital health interventions was the availability of timely and responsive personalized coaching by a dedicated health care professional.
摘要：
背景：数字技术的发展有可能改变糖尿病管理。现代糖尿病管理的关键方面之一仍然是实现血糖目标以避免急性和长期并发症。
目的：本研究旨在描述与2型糖尿病（T2DM）自我管理的各种数字干预措施的相对有效性或有效性和安全性有关的证据。主要关注降低糖化血红蛋白A1c（HbA1c）水平。
方法：通过搜索Embase进行系统文献综述（SLR），MEDLINE,和中央于2022年4月5日。研究选择，数据提取,质量评估由2名独立评审员进行.SLR的合格标准包括随机对照试验（RCTs）和评估包含两种人类干预措施的比较观察性研究（例如，教练)和数字组件(例如，血糖仪）在成人T2DM患者中。主要荟萃分析仅限于报道实验室测量的HbA1c的研究。在二级分析中，以数字干预中的指导强度作为分类协变量进行meta回归.
结果：总计，28项研究纳入本分析。大多数研究（23/28，82%）使用HbA1c水平的降低作为主要终点，直接或作为多组分结果的一部分。总的来说,21项研究报告了该主要终点的统计学显着结果。当根据支持数字健康技术的干预强度分为3个干预类别时（分析所有28项研究），成功率似乎与教练强度成正比(即，更高强度的研究报告了更高的成功率)。当分析仅限于使用HbA1c水平的比较改善的RCT时，干预措施的有效性尚不清楚.只有一半（12/23，52%）的纳入RCTs报告了统计学上显著的结果。荟萃分析与SLR的结果大致一致。主要分析估计，与常规护理相比，与数字干预相关的HbA1c降低更大（-0.31％，95%CI-0.45%至-0.16%；P<.001）。Meta回归估计降低-0.45%（95%CI-0.81%至-0.09%；P=.02），-0.29%（95%CI-0.48%至-0.11%；P=0.003），和-0.28%（95%CI-0.65%至0.09%；P=.20）与高，medium-,和低强度干预，分别。
结论：这些研究结果表明，借助数字干预降低T2DM患者的HbA1c水平是可行的，有效,并且可以接受。有效的数字健康干预措施的一个共同特征是由专门的医疗保健专业人员提供及时且响应迅速的个性化指导。
目的：本研究旨在描述与2型糖尿病（T2DM）自我管理的各种数字干预措施的相对有效性或有效性和安全性有关的证据。主要关注降低糖化血红蛋白A1c（HbA1c）水平。
方法：通过搜索Embase进行系统文献综述（SLR），MEDLINE,和中央于2022年4月5日。研究选择，数据提取,质量评估由2名独立评审员进行.SLR的合格标准包括随机对照试验（RCTs）和评估包含两种人类干预措施的比较观察性研究（例如，教练)和数字组件(例如，血糖仪）在成人T2DM患者中。主要荟萃分析仅限于报道实验室测量的HbA1c的研究。在二级分析中，以数字干预中的指导强度作为分类协变量进行meta回归.
结果：总计，28项研究纳入本分析。大多数研究（23/28，82%）使用HbA1c水平的降低作为主要终点，直接或作为多组分结果的一部分。总的来说,21项研究报告了该主要终点的统计学显着结果。当根据支持数字健康技术的干预强度分为3个干预类别时（分析所有28项研究），成功率似乎与教练强度成正比(即，更高强度的研究报告了更高的成功率)。当分析仅限于使用HbA1c水平的比较改善的RCT时，干预措施的有效性尚不清楚.只有一半（12/23，52%）的纳入RCTs报告了统计学上显著的结果。荟萃分析与SLR的结果大致一致。主要分析估计，与常规护理相比，与数字干预相关的HbA1c降低更大（-0.31％，95%CI-0.45%至-0.16%；P<.001）。Meta回归估计降低-0.45%（95%CI-0.81%至-0.09%；P=.02），-0.29%（95%CI-0.48%至-0.11%；P=0.003），和-0.28%（95%CI-0.65%至0.09%；P=.20）与高，medium-,和低强度干预，分别。
结论：这些研究结果表明，借助数字干预降低T2DM患者的HbA1c水平是可行的，有效,并且可以接受。有效的数字健康干预措施的一个共同特征是由专门的医疗保健专业人员提供及时且响应迅速的个性化指导。
