Walters Stephen John, Stern Cindy, Robertson-Malt Suzanne
1The Joanna Briggs Institute, Faculty of Health Sciences, University of Adelaide, Australia 2University of Adelaide, Australia.
JBI Database System Rev Implement Rep. 2016 Apr;14(4):138-97. doi: 10.11124/JBISRIR-2016-2159.
There is a growing call by consumers and governments for healthcare to adopt systems and approaches to care to improve patient safety. Collaboration within healthcare settings is an important factor for improving systems of care. By using validated measurement instruments a standardized approach to assessing collaboration is possible, otherwise it is only an assumption that collaboration is occurring in any healthcare setting.
The objective of this review was to evaluate and compare measurement properties of instruments that measure collaboration within healthcare settings, specifically those which have been psychometrically tested and validated.
INCLUSION CRITERIA, TYPES OF PARTICIPANTS: Participants could be healthcare professionals, the patient or any non-professional who contributes to a patient's care, for example, family members, chaplains or orderlies. The term participant type means the designation of any one participant; for example 'nurse', 'social worker' or 'administrator'. More than two participant types was mandatory.
TYPES OF INTERVENTION(S)/PHENOMENA OF INTEREST: The focus of this review was the validity of tools used to measure collaboration within healthcare settings.
The types of studies considered for inclusion were validation studies, but quantitative study designs such as randomized controlled trials, controlled trials and case studies were also eligible for inclusion. Studies that focused on Interprofessional Education, were published as an abstract only, contained patient self-reporting only or were not about care delivery were excluded.
The outcome of interest was validation and interpretability of the instrument being assessed and included content validity, construct validity and reliability. Interpretability is characterized by statistics such as mean and standard deviation which can be translated to a qualitative meaning.
The search strategy aimed to find both published and unpublished studies. A three-step search strategy was utilized in this review. The databases searched included PubMed, CINAHL, Embase, Cochrane Central Register of Controlled Trials, Emerald Fulltext, MD Consult Australia, PsycARTICLES, Psychology and Behavioural Sciences Collection, PsycINFO, Informit Health Databases, Scopus, UpToDate and Web of Science. The search for unpublished studies included EThOS (Electronic Thesis Online Service), Index to Theses and ProQuest- Dissertations and Theses.
The assessment of methodological quality of the included studies was undertaken using the COSMIN checklist which is a validated tool that assesses the process of design and validation of healthcare measurement instruments.
An Excel spreadsheet version of COSMIN was developed for data collection which included a worksheet for extracting participant characteristics and interpretability data.
Statistical pooling of data was not possible for this review. Therefore, the findings are presented in a narrative form including tables and figures to aid in data presentation. To make a synthesis of the assessments of methodological quality of the different studies, each instrument was rated by accounting for the number of studies performed with an instrument, the appraisal of methodological quality and the consistency of results between studies.
Twenty-one studies of 12 instruments were included in the review. The studies were diverse in their theoretical underpinnings, target population/setting and measurement objectives. Measurement objectives included: investigating beliefs, behaviors, attitudes, perceptions and relationships associated with collaboration; measuring collaboration between different levels of care or within a multi-rater/target group; assessing collaboration across teams; or assessing internal participation of both teams and patients.Studies produced validity or interpretability data but none of the studies assessed all validity and reliability properties. However, most of the included studies produced a factor structure or referred to prior factor analysis. A narrative synthesis of the individual study factor structures was generated consisting of nine headings: organizational settings, support structures, purpose and goals; communication; reflection on process; cooperation; coordination; role interdependence and partnership; relationships; newly created professional activities; and professional flexibility.
Among the many instruments that measure collaboration within healthcare settings, the quality of each instrument varies; instruments are designed for specific populations and purposes, and are validated in various settings. Selecting an instrument requires careful consideration of the qualities of each. Therefore, referring to systematic reviews of measurement properties of instruments may be helpful to clinicians or researchers in instrument selection.
Systematic reviews of measurement properties of instruments are valuable in aiding in instrument selection. This systematic review may be useful in instrument selection for the measurement of collaboration within healthcare settings with a complex mix of participant types. Evaluating collaboration provides important information on the strengths and limitations of different healthcare settings and the opportunities for continuous improvement via any remedial actions initiated.
Development of a tool that can be used to measure collaboration within teams of healthcare professionals and non-professionals is important for practice. The use of different statistical modelling techniques, such as Item Response Theory modelling and the translation of models into Computer Adaptive Tests, may prove useful. Measurement equivalence is an important consideration for future instrument development and validation. Further development of the COSMIN tool should include appraisal for measurement equivalence. Researchers developing and validating measurement tools should consider multi-method research designs.
消费者和政府越来越多地呼吁医疗保健行业采用护理系统和方法来提高患者安全。医疗保健环境中的协作是改善护理系统的一个重要因素。通过使用经过验证的测量工具，可以采用标准化方法来评估协作，否则只能假定在任何医疗保健环境中都存在协作。
本综述的目的是评估和比较用于测量医疗保健环境中协作的工具的测量属性，特别是那些经过心理测量测试和验证的工具。
纳入标准、参与者类型：参与者可以是医疗保健专业人员、患者或任何为患者护理做出贡献的非专业人员，例如家庭成员、牧师或护工。术语“参与者类型”是指任何一名参与者的指定；例如“护士”、“社会工作者”或“管理人员”。必须有两种以上的参与者类型。
干预类型/感兴趣的现象：本综述的重点是用于测量医疗保健环境中协作的工具的有效性。
考虑纳入的研究类型是验证性研究，但随机对照试验、对照试验和案例研究等定量研究设计也符合纳入条件。专注于跨专业教育、仅以摘要形式发表、仅包含患者自我报告或与护理提供无关的研究均被排除。
感兴趣的结果是被评估工具的有效性和可解释性，包括内容效度、结构效度和信度。可解释性由均值和标准差等统计数据表征，这些数据可以转化为定性含义。
检索策略旨在查找已发表和未发表的研究。本综述采用了三步检索策略。检索的数据库包括PubMed、CINAHL、Embase、Cochrane对照试验中央注册库、Emerald全文数据库、MD Consult Australia、PsycARTICLES、心理学与行为科学文集、PsycINFO、Informit健康数据库、Scopus、UpToDate和科学引文索引。对未发表研究的检索包括EThOS（电子论文在线服务）、论文索引和ProQuest-学位论文数据库。
使用COSMIN清单对纳入研究的方法学质量进行评估，该清单是一种经过验证的工具，用于评估医疗保健测量工具的设计和验证过程。
开发了一个COSMIN的Excel电子表格版本用于数据收集，其中包括一个用于提取参与者特征和可解释性数据的工作表。
本综述无法对数据进行统计合并。因此，研究结果以叙述形式呈现，包括表格和图表以辅助数据展示。为了综合不同研究的方法学质量评估，对每个工具的评估是通过考虑使用该工具进行的研究数量、方法学质量评估以及研究之间结果的一致性来进行的。
本综述纳入了对12种工具的21项研究。这些研究在理论基础、目标人群/环境和测量目标方面各不相同。测量目标包括：调查与协作相关的信念、行为、态度、看法和关系；测量不同护理级别之间或多评估者/目标群体内的协作；评估跨团队的协作；或评估团队和患者的内部参与情况。研究产生了有效性或可解释性数据，但没有一项研究评估了所有的有效性和信度属性。然而，大多数纳入研究产生了一个因子结构或提及了先前的因子分析。生成了一个个体研究因子结构的叙述性综合，包括九个标题：组织环境、支持结构、目的和目标；沟通；对过程的反思；合作；协调；角色相互依存和伙伴关系；关系；新创建的专业活动；以及专业灵活性。
在众多测量医疗保健环境中协作的工具中，每个工具的质量各不相同；工具是针对特定人群和目的设计的，并在各种环境中得到验证。选择工具需要仔细考虑每个工具的质量。因此，参考工具测量属性的系统评价可能有助于临床医生或研究人员进行工具选择。
工具测量属性的系统评价有助于工具选择。本系统评价可能有助于选择用于测量具有复杂参与者类型组合的医疗保健环境中协作的工具。评估协作提供了关于不同医疗保健环境的优势和局限性以及通过任何补救行动进行持续改进机会的重要信息。
开发一种可用于测量医疗保健专业人员和非专业人员团队内部协作的工具对实践很重要。使用不同的统计建模技术，如项目反应理论建模以及将模型转化为计算机自适应测试可能会很有用。测量等效性是未来工具开发和验证的一个重要考虑因素。COSMIN工具的进一步开发应包括对测量等效性的评估。开发和验证测量工具的研究人员应考虑采用多方法研究设计。
Zotero 插件