Active Implementation Research Network, Inc., Chapel Hill, NC, United States.
Front Public Health. 2024 Oct 16;12:1454268. doi: 10.3389/fpubh.2024.1454268. eCollection 2024.
Getting the science right for implementation is critical for making the processes for improving outcomes more predictable and effective in global public health. Unfortunately, "implementation science" has become a catchphrase for ideas, assumptions, and findings concerning the science to service gap and how to close it. The purpose of this paper is to explore the dimensions of a "science of implementation" that meets the definitions of a science and is focused on implementation variables (i.e., purposeful processes to put innovations into effect so that intended benefits can be realized). A science of implementation is important for accomplishing the goals related to improving the health and well-being of populations around the world. Much of public health involves interaction-based interventions. In a typology of science, interaction-based interventions are created by specifying the nature of certain exchanges between and among individual people or groups. The complexity of developing interaction-based independent variables requires meeting benchmarks for fidelity to assure the presence and strength of implementation independent variables. The paper presents information related to the following tenets: (1) A science of implementation is based on if-then predictions. Science is cumulative. As predictions are made, tested, and elaborated, the facts accumulate to form the knowledge base for science and practice. (2) Implementation variables are interaction-based inventions and, therefore, must be created and established so the specific set of activities related to implementation can be studied. (3) A science of implementation is based on theory that organizes facts, leads to testable predictions, and is modified or discarded based on outcomes. (4) A science of interaction-based implementation depends on frequent measures of independent and dependent variables specific to implementation methods and outcomes. Two examples illustrate the implications for theory, research, and practice. The paper advocates a paradigm shift to a new mental model that values fidelity over tailoring, has one size fits all as a goal, and is concerned with the function of evidence rather than the form of evidence based on RCTs. Global health fundamentally requires scaling implementation capacity so that effective innovations can be used as intended and with good effect to achieve population benefits.
为了使改善结果的过程在全球公共卫生领域更具可预测性和有效性，正确把握实施科学是至关重要的。不幸的是，“实施科学”已经成为一个流行语，涵盖了有关科学与服务差距以及如何弥合这一差距的理念、假设和发现。本文旨在探讨符合科学定义并专注于实施变量（即有目的地将创新付诸实践，以实现预期效益的过程）的“实施科学”的各个维度。实施科学对于实现改善全球人口健康和福祉的目标至关重要。许多公共卫生工作都涉及基于交互的干预措施。在科学分类中，基于交互的干预措施是通过指定个人或群体之间特定交互的性质来创建的。开发基于交互的自变量的复杂性需要满足保真度基准，以确保实施自变量的存在和强度。本文介绍了以下十个原则的相关信息：（1）实施科学基于如果-那么预测。科学是累积的。随着预测的提出、测试和阐述，事实不断积累，形成科学和实践的知识库。（2）实施变量是基于交互的发明，因此必须创建和建立这些变量，以便可以研究与实施相关的特定活动集。（3）实施科学基于组织事实、产生可测试预测并根据结果进行修改或摒弃的理论。（4）基于交互的实施科学依赖于针对实施方法和结果的特定实施变量和依赖变量的频繁测量。两个示例说明了对理论、研究和实践的影响。本文提倡向新的思维模式转变，这种模式重视保真度而不是定制化，以一刀切为目标，并关注证据的功能而不是基于 RCT 的证据形式。全球卫生从根本上需要扩大实施能力，以便有效创新能够按预期使用，并产生良好效果，从而实现人口效益。
