Department of Community Health Sciences, Cumming School of Medicine, University of Calgary, 3D10-3280 Hospital Drive NW, Calgary, AB, T2N 4Z6, Canada.
Strategic Clinical Networks™, Provincial Clinical Excellence, Alberta Health Services, Calgary, AB, Canada.
Health Res Policy Syst. 2022 Dec 14;20(1):133. doi: 10.1186/s12961-022-00938-8.
Accurate, consistent assessment of outcomes and impacts is challenging in the health research partnerships domain. Increased focus on tool quality, including conceptual, psychometric and pragmatic characteristics, could improve the quantification, measurement and reporting partnership outcomes and impacts. This cascading review was undertaken as part of a coordinated, multicentre effort to identify, synthesize and assess a vast body of health research partnership literature.
To systematically assess the outcomes and impacts of health research partnerships, relevant terminology and the type/use of theories, models and frameworks (TMF) arising from studies using partnership assessment tools with known conceptual, psychometric and pragmatic characteristics.
Four electronic databases were searched (MEDLINE, Embase, CINAHL Plus and PsycINFO) from inception to 2 June 2021. We retained studies containing partnership evaluation tools with (1) conceptual foundations (reference to TMF), (2) empirical, quantitative psychometric evidence (evidence of validity and reliability, at minimum) and (3) one or more pragmatic characteristics. Outcomes, impacts, terminology, definitions and TMF type/use were abstracted verbatim from eligible studies using a hybrid (independent abstraction-validation) approach and synthesized using summary statistics (quantitative), inductive thematic analysis and deductive categories (qualitative). Methodological quality was assessed using the Quality Assessment Tool for Studies with Diverse Designs (QATSDD).
Application of inclusion criteria yielded 37 eligible studies. Study quality scores were high (mean 80%, standard deviation 0.11%) but revealed needed improvements (i.e. methodological, reporting, user involvement in research design). Only 14 (38%) studies reported 48 partnership outcomes and 55 impacts; most were positive effects (43, 90% and 47, 89%, respectively). Most outcomes were positive personal, functional, structural and contextual effects; most impacts were personal, functional and contextual in nature. Most terms described outcomes (39, 89%), and 30 of 44 outcomes/impacts terms were unique, but few were explicitly defined (9, 20%). Terms were complex and mixed on one or more dimensions (e.g. type, temporality, stage, perspective). Most studies made explicit use of study-related TMF (34, 92%). There were 138 unique TMF sources, and these informed tool construct type/choice and hypothesis testing in almost all cases (36, 97%).
This study synthesized partnership outcomes and impacts, deconstructed term complexities and evolved our understanding of TMF use in tool development, testing and refinement studies. Renewed attention to basic concepts is necessary to advance partnership measurement and research innovation in the field. Systematic review protocol registration: PROSPERO protocol registration: CRD42021137932 https://www.crd.york.ac.uk/prospero/display_record.php?RecordID=137932 .
在健康研究伙伴关系领域，准确、一致地评估结果和影响具有挑战性。更多地关注工具质量，包括概念、心理测量和实用特征，可以提高伙伴关系结果和影响的量化、测量和报告。本次级联审查是作为一项协调的、多中心努力的一部分进行的，旨在识别、综合和评估大量健康研究伙伴关系文献。
系统评估具有已知概念、心理测量和实用特征的伙伴关系评估工具中使用的健康研究伙伴关系、相关术语以及理论、模型和框架（TMF）的结果和影响。
从 2021 年 6 月 2 日起，在四个电子数据库（MEDLINE、Embase、CINAHL Plus 和 PsycINFO）中进行了搜索。我们保留了包含伙伴关系评估工具的研究，这些工具具有（1）概念基础（TMF 的参考），（2）经验、定量心理测量证据（至少有有效性和可靠性的证据）和（3）一种或多种实用特征。使用混合（独立提取-验证）方法从合格研究中逐字摘录结果、影响、术语、定义和 TMF 类型/使用，并使用汇总统计数据（定量）、归纳主题分析和演绎类别（定性）进行综合。使用多样性设计研究质量评估工具（QATSDD）评估研究方法的质量。
应用纳入标准后，共获得 37 项合格研究。研究质量评分较高（平均 80%，标准偏差 0.11%），但仍需改进（即方法学、报告、用户参与研究设计）。只有 14 项（38%）研究报告了 48 项伙伴关系结果和 55 项影响；大多数是积极的影响（43，90%和 47，89%，分别）。大多数结果是积极的个人、功能、结构和环境影响；大多数影响本质上是个人、功能和环境的。大多数术语描述了结果（39，89%），44 个结果/影响术语中有 30 个是独特的，但很少有明确定义（9，20%）。术语在一个或多个维度上复杂且混合（例如，类型、时间性、阶段、视角）。大多数研究都明确使用了与研究相关的 TMF（34，92%）。有 138 个独特的 TMF 来源，这些来源几乎在所有情况下都为工具构建类型/选择和假设检验提供了信息（36，97%）。
本研究综合了伙伴关系的结果和影响，对术语的复杂性进行了分解，并深化了我们对 TMF 在工具开发、测试和改进研究中的使用的理解。为了在该领域推进伙伴关系测量和研究创新，有必要重新关注基本概念。系统评价方案注册：PROSPERO 方案注册：CRD42021137932 https://www.crd.york.ac.uk/prospero/display_record.php?RecordID=137932 。
