Does increased implementation support improve community clinics' guideline-concordant care? Results of a mixed methods, pragmatic comparative effectiveness trial
Does increased implementation support improve community clinics' guideline-concordant care? Results of a mixed methods, pragmatic comparative effectiveness trial
复制标题
DOI：
10.1186/s13012-019-0948-5
复制
发表时间：
2019-12-05
影响因子：
7.2
通讯作者：
Bulkley, Joanna
中科院分区：
文献类型：
--
作者：
Gold, Rachel;Bunce, Arwen;Bulkley, Joanna
背景：将护理指南传播到临床实践中仍然具有挑战性，部分原因是关于如何最好地帮助临床将新指南纳入常规护理的证据不足。在安全网社区卫生中心(CHCs)尤其如此。方法：这项务实的比较有效性试验采用平行混合方法设计。29家CHC诊所被随机分配，接受越来越密集的实施支持(实施工具包(ARM 1)；工具包+面对面培训+培训网络研讨会(ARM 2)；工具包+培训+网络研讨会+提供的实践促进(ARM 3))，目标是吸收电子健康记录(EHR)工具，重点是为糖尿病患者开出符合指南的心脏保护处方。研究人员比较了不同研究分支的结果，以测试增加支持是否会产生相加的改善，并与137名与研究诊所共享相同EHR的非研究CHC进行比较。CHC的EHR的定量数据被用来比较符合指南的ACE/ARB和他汀类药物处方的变化幅度，使用调整后的泊松回归。通过不同方法(如访谈、观察)收集的定性数据确定了影响量化结果的因素。结果：接受更高强度支持的CHC的结果不是以相加的模式改善的。任何CHC组的ACE/ARB处方均未见改善。他汀类药物的处方总体上有所改善，仅在第1组和第2组患者中显著高于未研究的患者。影响发现没有附加影响的因素包括：电子健康记录工具降低其实用性的方面，提供预期实施支持的障碍，以及研究设计要素，例如，无法调整所提供的支持。影响他汀类药物结果总体改善的因素可能包括对他汀类药物处方指南认识的长期趋势，有动机的诊所自愿参加研究的偏向，以及将临床工作人员集中在目标结果上的研究参与。结论：实施护理指南的努力应该：在提供实施支持时确保适应性，并进行形成性评估，以确定对给定诊所的这种支持的最佳形式；考虑研究数据收集如何影响采用；以及考虑诊所按计划使用/接受实施支持的能力的障碍。需要进行更多的研究，以支持在资源不足的环境中实施变革，如社区卫生中心。
Background: Disseminating care guidelines into clinical practice remains challenging, partly due to inadequate evidence on how best to help clinics incorporate new guidelines into routine care. This is particularly true in safety net community health centers (CHCs).Methods: This pragmatic comparative effectiveness trial used a parallel mixed methods design. Twenty-nine CHC clinics were randomized to receive increasingly intensive implementation support (implementation toolkit (arm 1); toolkit + in-person training + training webinars (arm 2); toolkit + training + webinars + offered practice facilitation (arm 3)) targeting uptake of electronic health record (EHR) tools focused on guideline-concordant cardioprotective prescribing for patients with diabetes. Outcomes were compared across study arms, to test whether increased support yielded additive improvements, and with 137 non-study CHCs that share the same EHR as the study clinics. Quantitative data from the CHCs' EHR were used to compare the magnitude of change in guideline-concordant ACE/ARB and statin prescribing, using adjusted Poisson regressions. Qualitative data collected using diverse methods (e.g., interviews, observations) identified factors influencing the quantitative outcomes.Results: Outcomes at CHCs receiving higher-intensity support did not improve in an additive pattern. ACE/ARB prescribing did not improve in any CHC group. Statin prescribing improved overall and was significantly greater only in the arm 1 and arm 2 CHCs compared with the non-study CHCs. Factors influencing the finding of no additive impact included: aspects of the EHR tools that reduced their utility, barriers to providing the intended implementation support, and study design elements, e.g., inability to adapt the provided support. Factors influencing overall improvements in statin outcomes likely included a secular trend in awareness of statin prescribing guidelines, selection bias where motivated clinics volunteered for the study, and study participation focusing clinic staff on the targeted outcomes.Conclusions: Efforts to implement care guidelines should: ensure adaptability when providing implementation support and conduct formative evaluations to determine the optimal form of such support for a given clinic; consider how study data collection influences adoption; and consider barriers to clinics' ability to use/accept implementation support as planned. More research is needed on supporting change implementation in under-resourced settings like CHCs.