Department of Epidemiology and Biostatistics, University of California, San Francisco, San Francisco, CA.
Currently, Department of Biostatistics, University of California, Berkeley, Berkeley, CA.
J Acquir Immune Defic Syndr. 2019 Dec;82 Suppl 3(Suppl 3):S199-S205. doi: 10.1097/QAI.0000000000002202.
Implementation science focuses on evaluating strategies for delivering evidence-based interventions to improve HIV prevention and treatment. The effectiveness of these implementation strategies is often context-dependent and reconciling the desire to produce generalizable knowledge in the face of these contextual interventions is a central challenge for implementation science researchers.
We provide an overview of the causal transportability theory and conceptualize context under this framework. We review how causal graphs can be used to illustrate the assumptions necessary to apply the results of a study to a new context, and we illustrate this approach using an example of a community adherence group intervention that aims to improve retention in HIV care. Finally, we discuss several key insights highlighted by the transportability theory that are relevant to implementation science researchers.
By adopting causal transportability to consider how context may affect the success of an implementation strategy, researchers can formally diagnose when the results of a study are likely to generalize to a given setting. Moreover, selection diagrams can highlight what additional measurements would be needed in a target population to estimate the effect of an implementation strategy in that target population without having to repeat the initial study.
Transportability translates intuition about context-dependent interventions and external validity into actionable and testable insight.
实施科学侧重于评估提供基于证据的干预措施的策略，以改善 HIV 的预防和治疗。这些实施策略的有效性往往取决于具体情况，面对这些干预措施，如何协调产生普遍适用知识的愿望，是实施科学研究人员面临的一个核心挑战。
我们提供了因果可传递性理论的概述，并在该框架下对背景进行了概念化。我们回顾了如何使用因果图来说明将研究结果应用于新环境所需的假设，并用一个旨在提高 HIV 护理保留率的社区依从性小组干预措施的示例来说明这种方法。最后，我们讨论了可传递性理论中突出的几个与实施科学研究人员相关的关键见解。
通过采用因果可传递性来考虑背景如何影响实施策略的成功，研究人员可以正式诊断研究结果在给定环境中是否可能推广。此外，选择图可以突出在目标人群中需要进行哪些额外的测量，以便在没有重复初始研究的情况下估计实施策略在该目标人群中的效果。
可传递性将关于基于具体情况的干预措施和外部有效性的直观认识转化为可操作和可测试的见解。
