Wholey Douglas R, Finch Michael, Christianson Jon B, Knutson David, Rockwood Todd, Warrick Louise
Division of Health Services Research and Policy, University of Minnesota, Minneapolis, Minn 55455, USA.
Am J Manag Care. 2003 Jun;9 Spec No 2:SP76-87.
To assess the measurement and scaling properties of survey items designed to measure health plan quality from a physician's perspective.
Prospective survey design with multivariate regression analysis.
Data were from 3798 physicians representing 23 health plans in 5 regions: Florida, New York, Colorado, Pennsylvania, and Washington. Scale reliability was assessed by using the Cronbach alpha. Generalist and specialist scales were compared with structural equation modeling. Multivariate analysis was used to examine internal validity by testing theoretically based hypotheses.
Scales constructed from the data were reliable, were stable across both generalist and specialist physicians, and demonstrated construct validity. Hypotheses about the relationship between physician experiences with a health plan and physician recommendations of the plan were confirmed, supporting construct validity.
The items on the survey instrument can be used with confidence to measure health plan quality from a physician perspective.
从医生的角度评估旨在衡量健康计划质量的调查项目的测量和量表特性。
采用多变量回归分析的前瞻性调查设计。
数据来自代表佛罗里达州、纽约州、科罗拉多州、宾夕法尼亚州和华盛顿州5个地区23个健康计划的3798名医生。使用克朗巴哈系数评估量表信度。通过结构方程模型比较全科医生和专科医生量表。多变量分析用于通过检验基于理论的假设来检验内部效度。
根据数据构建的量表可靠，在全科医生和专科医生中都很稳定，并显示出结构效度。关于医生对健康计划的体验与医生对该计划的推荐之间关系的假设得到证实，支持结构效度。
调查问卷中的项目可以放心地用于从医生角度衡量健康计划质量。
