可穿戴技术对超重和肥胖成年人的生活方式干预：系统评价和Meta回归

　　Background

　　Globally, overweight and obesity are becoming a growing concern, and wearable technology combined with lifestyle intervention may offer an innovative solution.

　　This review aimed to (1) assess the effectiveness of lifestyle interventions delivered by wearable technology in improving weight loss and physical activity among overweight or obese adults and (2) explore the effects of covariates on intervention outcomes.

　　Systematic review, meta-analysis and meta-regression.

　　The criteria for inclusion in the review were that the trial must be a lifestyle modification intervention that utilised wearable technology and had a randomised control design and obese or overweight participants aged 18–64 years. Ten electronic databases were searched from inception to 8 December 2020. The Cochrane Risk of Bias Tool version 1 and Grading of Recommendations, Assessment, Development and Evaluations were adopted to rate risk of bias of individual trials and certainty of evidence, respectively. Stata 16 software was used to conduct the meta-analysis, subgroup analysis and meta-regression analysis.

　　Thirty trials comprising 5,391 adults from 11 countries were included. Meta-analyses found significant changes in weight (?1.08 kg, 95% confidence interval, CI: ?1.88, ?0.28), body mass index (?0.36 kg/m2, 95% CI: ?0.62, ?0.09), waist circumference (?1.12 cm, 95% CI: ?2.08, ?0.16), steps per day (1,243.51 steps, 95% CI: 111.51, 2375.51), steps per day change (456.18 steps, 95% CI: 40.61, 871.76), systolic pressure (?2.57 mmHg, 95% CI: ?4.57, ?0.56) and diastolic pressure (?2.10 mmHg, 95% CI: ?3.43, ?0.77). Significant differences were found between subgroups regarding region (Q?=?7.35, p?=?0.01), lifestyle component (Q?=?8.51, p?=?0.01) and registration protocol (Q?=?20.24, p < 0.01). The multivariable meta-regression model suggested that year of publication (adjusted β?=?0.20, 95% CI: 0.07, 0.34, p?=?0.003), mean age (adjusted β?=??0.94, 95% CI: ?0.14, 0.04, p < 0.001) and duration (adjusted β?=?0.17, 95% CI: 0.09, 0.25, p < 0.001) had significant effects on the mean difference of weight change.

　　The majority of the certainty of evidence was graded moderate to high, which suggests that interventions utilising wearable technology may improve body mass index, waist circumference and physical activity of participants. These findings may aid in the development of future health interventions. However, the current review was limited to self-selected samples and trials conducted in English.

　　摘要翻译（仅供参考）

　　在全球范围内，超重和肥胖日益受到关注，可穿戴技术与生活方式干预相结合可能会提供一种创新的解决方案。

　　本综述旨在 (1) 评估可穿戴技术提供的生活方式干预在改善超重或肥胖成年人体重减轻和身体活动方面的有效性，以及 (2) 探讨协变量对干预结果的影响。

　　系统评价、荟萃分析和荟萃回归。

　　纳入审查的标准是该试验必须是改变生活方式的干预措施，使用可穿戴技术并具有随机对照设计，参与者年龄在 18-64 岁之间。从开始到 2020 年 12 月 8 日，检索了 10 个电子数据库。分别采用 Cochrane 偏倚风险工具第 1 版和推荐、评估、制定和评价分级来评估个别试验的偏倚风险和证据的确定性。采用Stata 16软件进行meta分析、亚组分析和meta回归分析。

　　包括来自 11 个国家的 5,391 名成年人的 30 项试验。荟萃分析发现，在重量变化显著（-1.08千克，95％置信区间，CI：-1.88，-0.28）身体质量指数，（-0.36千克/米2，95％CI：-0.62，-0.09），腰周长 (-1.12 cm, 95% CI: -2.08, -0.16), 每天步数 (1,243.51, 95% CI: 111.51, 2375.51), 每天步数变化 (456.18, 95% CI: 40.61, 871.收缩压（-2.57 mmHg，95% CI：-4.57，-0.56）和舒张压（-2.10 mmHg，95% CI：-3.43，-0.77）。在区域（Q = 7.35，p ?= 0.01）、生活方式成分（Q = 8.51，p ?= 0.01）和注册协议（Q = 20.24，p< 0.01)。多变量元回归模型表明出版年份（调整后β ?= 0.20，95% CI：0.07、0.34，p ?= 0.003）、平均年龄（调整后β ?= -0.94、95% CI：-0.14、0.04，p < 0.001）和持续时间（调整后的β ?= 0.17, 95% CI: 0.09, 0.25, p < 0.001）对体重变化的平均差异有显着影响。

　　大多数确定性证据的等级为中到高，这表明利用可穿戴技术的干预措施可以改善参与者的体重指数、腰围和身体活动。这些发现可能有助于开发未来的健康干预措施。然而，目前的审查仅限于自选样本和以英语进行的试验。