Higginson Irene J, Donaldson Nora
Department of Palliative Care and Policy, King's College London, Weston Education Centre, Cutcombe Road, London SE5 9RJ, UK.
Health Qual Life Outcomes. 2004 Nov 29;2:68. doi: 10.1186/1477-7525-2-68.
Various scales have been used to assess palliative outcomes. But measurement can still be problematic and core components of measures have not been identified. This study aimed to determine the relationships between, and factorial structure of, three widely used scales among advanced cancer patients.
Patients were recruited who received home or hospital palliative care services in the south of England. Hope, quality of life and palliative outcomes were assessed by patients in face to face interviews, using three previously established scales - a generic measure (EQoL), a palliative care specific measure (POS) and a measure of hope (Herth Hope Index). Analysis comprised: exploratory factor analysis of each individual scale, and all scales combined, and confirmatory factor analysis for model building and validation.
Of 171 patients identified, 140 (81%) consented and completed first interviews; mean age was 71 years, 54% were women, 132 had cancer. In exploratory analysis of individual means, three out of the five factors in the EQoL explained 75% of its variability, four out of the 10 factors in POS explained 63% of its variability, and in the Hope Index, nine out of the 12 items explained 69% of its variability. When exploring the relative factorial structure of all three scales, five factors explained 56% of total combined variability. Confirmatory analysis reduced this to a model with four factors - self-sufficiency, positivity, symptoms and spiritual. Removal of the spiritual factor left a model with an improved goodness of fit and a measure with 11 items.
We identified three factors which are important outcomes and would be simple to measure in clinical practice and research.
已使用多种量表来评估姑息治疗的结果。但测量仍可能存在问题，且尚未确定测量的核心组成部分。本研究旨在确定晚期癌症患者中三种广泛使用的量表之间的关系及其因子结构。
招募在英格兰南部接受家庭或医院姑息治疗服务的患者。通过面对面访谈，使用三种先前建立的量表——一种通用测量量表（EQoL）、一种姑息治疗专用测量量表（POS）和一种希望测量量表（赫思希望指数），对患者的希望、生活质量和姑息治疗结果进行评估。分析包括：对每个单独量表以及所有量表合并后的探索性因子分析，以及用于模型构建和验证的验证性因子分析。
在确定的171名患者中，140名（81%）同意并完成了首次访谈；平均年龄为71岁，54%为女性，132名患有癌症。在个体均值的探索性分析中，EQoL的五个因子中的三个解释了其75%的变异性，POS的十个因子中的四个解释了其63%的变异性，在希望指数中，十二个项目中的九个解释了其69%的变异性。在探索所有三个量表的相对因子结构时，五个因子解释了总合并变异性的56%。验证性分析将其简化为一个具有四个因子的模型——自给自足、积极性、症状和精神。去除精神因子后留下一个拟合优度得到改善且有11个项目的测量量表的模型。
我们确定了三个重要的结果因子，在临床实践和研究中易于测量。
