Batterham R, Southern D, Appleby N, Elsworth G, Fabris S, Dunt D, Young D
Centre for Health Program Evaluation, Department of General Practice & Public Health, The University of Melbourne, West Heidelberg, VIC, Australia.
Soc Sci Med. 2002 Apr;54(8):1225-41. doi: 10.1016/s0277-9536(01)00092-2.
There are frequent calls to improve integration of health services, within and between primary and secondary care sectors. In Australia, general medical practitioners (GPs) are central to these endeavours. This paper aims to better conceptualise GP integration and to develop a model and index based on this. A conceptualisation of integration is proposed based on integration fundamentally as an activity or process not structure. Integration process is the frequency and quality of episodes of information exchange involving the GP and another practitioner or patient and aimed at fulfilling the objectives of the health care system with regard to patient care. These are both direct responses to structural forces and emergent GP capacities and dispositions. The content of this typology was studied using Concept Mapping in 11 groups of GPs, consumers and other practitioners. Clusters of related statements within thematic domains were used as the basis for a provisional model. This was tested using confirmatory factor analysis in a data set derived from a national probability sample of 501 GPs. Some re-specification of the model was necessary, with three integration process factors needing to be subdivided. One factor congeneric model assumptions were used to identify the constituent items for these factors. The result was a model in which 50 items measured nine integration process factors and 20 items measured five enabling factors. Two distinct but correlated higher order factors, relating to individual patient care and public (or community) health--in contrast to a single higher order factor for integration--were identified. The re-specified model was tested with a new sample of 151 GPs and exhibited strong psychometric properties. Reliability and validity were acceptable to this stage of the indices' development. Further testing of the index is necessary to demonstrate factor invariance of the indices in other contexts as well as their utility in cross-structural analysis. That said, the indices have immediate uses.
人们经常呼吁改善医疗服务的整合，包括初级和二级医疗部门内部以及它们之间的整合。在澳大利亚，全科医生（GP）是这些努力的核心。本文旨在更好地概念化全科医生的整合，并在此基础上开发一个模型和指数。本文提出了一种整合的概念化，其根本基础是将整合视为一种活动或过程而非结构。整合过程是涉及全科医生与另一名从业者或患者的信息交流事件的频率和质量，旨在实现医疗保健系统在患者护理方面的目标。这些既是对结构力量的直接反应，也是全科医生新出现的能力和倾向。使用概念映射法对11组全科医生、消费者和其他从业者进行了研究，以探讨这种类型学的内容。主题领域内相关陈述的聚类被用作临时模型的基础。在一个由501名全科医生的全国概率样本得出的数据集中，使用验证性因子分析对该模型进行了测试。对模型进行一些重新设定是必要的，需要细分三个整合过程因素。使用同属模型假设来确定这些因素的构成项目。结果得到一个模型，其中50个项目测量九个整合过程因素，20个项目测量五个促成因素。确定了两个不同但相关的高阶因素，分别与个体患者护理和公共（或社区）健康相关——这与整合的单一高阶因素形成对比。使用151名全科医生的新样本对重新设定的模型进行了测试，该模型表现出很强的心理测量特性。在指数发展的这个阶段，可靠性和有效性是可以接受的。有必要对该指数进行进一步测试，以证明其在其他背景下的因子不变性以及在跨结构分析中的效用。话虽如此，这些指数具有直接的用途。
