共享决策对 2 型糖尿病成年患者血糖控制效果的系统评价和 Meta 分析。

BACKGROUND: In diabetes care and management guidelines, shared decision-making (SDM) implementation is explicitly recommended to help patients and health care providers to make informed shared decisions that enable informed choices and the selection of treatments. Despite widespread calls for SDM to be embedded in health care, there is little evidence to support SDM in the management and care of diabetes. It is still not commonly utilized in routine care settings because its effects remain poorly understood. Hence, the current systematic review and meta-analysis aimed to evaluate the effectiveness of SDM for glycaemic control among type 2 diabetes adult patients. METHODS: Literature sources were searched in MEDLINE, PubMed, Cochrane library and HINARI bibliographic databases and Google Scholar. When these records were searched and reviewed, the PICO criteria (P: population, I: intervention, C: comparator, and O: outcome) were applied. The extracted data was exported to RevMan software version 5.4 and STATA 17 for further analysis. The mean differences (MD) of glycated hemoglobin (HbA1c) were pooled using a random effect model (REM), and sub-group analysis were performed to evaluate the effect size differences across the duration of the follow-up period, modes of intervention, and baseline glycated hemoglobin level of patient groups. The sensitivity analysis was performed using a leave-one-out meta-analysis to quantify the impact of each study on the overall effect size in mean difference HbA1c%. Finally, the statistically significant MD of HbA1c% between the intervention groups engaged in SDM and control groups received usual care was declared at P ˂0.05, using a 95% confidence interval (CI). RESULTS: In the database search, 425 records were retrieved, with only 17 RCT studies fulfilling the inclusion criteria and were included in the meta-analysis. A total of 5416 subjects were included, out of which 2782(51.4%) were included in trial arms receiving SDM and 2634(48.6%) were included in usual diabetes care. The Higgins (I2) test statistics were calculated to be 59.1%, P = 0.002, indicating statistically significant heterogeneity was observed among the included studies, and REM was used as a remedial to estimate the pooled MD of HbA1c% level between patients who participated in SDM and received usual care. As a result, the pooled MD showed that the SDM significantly lowered HbA1c by 0.14% compared to the usual care (95% CI = [-0.26, -0.02], P = 0.02). SDM significantly decreased the level of HbA1c by 0.14% (95% CI = -0.28, -0.01, P = 0.00) when shared decisions were made in person or face-to-face at the point of care, but there was no statistically significant reduction in HbA1c levels when patients were engaged in online SDM. In patients with poorly controlled glycaemic level (≥ 8%), SDM significantly reduced level of HbA1c by 0.13%, 95% CI = [-0.29, -0.03], P = 0.00. However, significant reduction in HbA1c was not observed in patients with ˂ 8%, HbA1c baseline level. CONCLUSIONS: Overall, statistically significant reduction of glycated hemoglobin level was observed among T2DM adult patients who participated in shared decision-making compared to those patients who received diabetes usual care that could lead to improved long-term health outcomes, reducing the risk of diabetes-related complications. Therefore, we strongly suggest that health care providers and policy-makers should integrate SDM into diabetes health care and management, and further study should focus on the level of patients' empowerment, health literacy, and standardization of decision supporting tools to evaluate the effectiveness of SDM in diabetes patients.

背景：在糖尿病护理和管理指南中，明确推荐实施共同决策（SDM），以帮助患者和医疗保健提供者做出知情的共同决策，从而做出明智的选择并选择治疗方案。尽管广泛呼吁将 SDM 嵌入医疗保健中，但在糖尿病的管理和护理方面，几乎没有证据支持 SDM。它仍然没有在常规护理环境中广泛使用，因为其效果仍然知之甚少。因此，目前的系统评价和荟萃分析旨在评估 SDM 对 2 型糖尿病成年患者血糖控制的有效性。

方法：在 MEDLINE、PubMed、Cochrane 图书馆和 HINARI 书目数据库以及 Google Scholar 中搜索文献来源。当搜索和审查这些记录时，应用了 PICO 标准（P：人群，I：干预，C：比较，O：结局）。提取的数据导出到 RevMan 软件版本 5.4 和 STATA 17 进行进一步分析。使用随机效应模型（REM）汇总糖化血红蛋白（HbA1c）的平均差异（MD），并进行亚组分析，以评估随访期间持续时间、干预模式以及患者组基线糖化血红蛋白水平的效应大小差异。使用单因素剔除荟萃分析进行敏感性分析，以量化每项研究对总体平均差异 HbA1c%效应大小的影响。最后，使用 95%置信区间（CI），宣布在接受 SDM 干预的试验组与接受常规护理的对照组之间的 HbA1c%的统计学显著 MD 为 P ˂0.05。

结果：在数据库搜索中，检索到 425 条记录，只有 17 项 RCT 研究符合纳入标准，并纳入荟萃分析。共有 5416 名受试者纳入研究，其中 2782 名（51.4%）纳入接受 SDM 的试验组，2634 名（48.6%）纳入常规糖尿病护理组。计算 Higgins（I2）检验统计量为 59.1%，P = 0.002，表明纳入研究之间存在显著的异质性，使用 REM 作为补救措施来估计接受 SDM 和接受常规护理的患者之间 HbA1c%水平的汇总 MD。结果表明，与常规护理相比，SDM 显著降低 HbA1c 水平 0.14%（95%CI = [-0.26，-0.02]，P = 0.02）。当在护理点以面对面或亲自方式进行共同决策时，SDM 可显著降低 0.14%的 HbA1c 水平（95%CI = -0.28，-0.01，P = 0.00），但当患者参与在线 SDM 时，HbA1c 水平没有统计学显著降低。在血糖控制水平较差（≥8%）的患者中，SDM 可显著降低 HbA1c 水平 0.13%，95%CI = [-0.29，-0.03]，P = 0.00。然而，在 HbA1c 基线水平 ˂ 8%的患者中，未观察到 HbA1c 水平的显著降低。

结论：总体而言，与接受常规糖尿病护理的患者相比，参与共同决策的 2 型糖尿病成年患者的糖化血红蛋白水平显著降低，这可能导致长期健康结局改善，降低糖尿病相关并发症的风险。因此，我们强烈建议医疗保健提供者和政策制定者将 SDM 纳入糖尿病医疗保健和管理中，并进一步研究应侧重于患者赋权、健康素养和决策支持工具的标准化，以评估 SDM 在糖尿病患者中的有效性。