修订后的患者安全测量指标的验证：患者安全测量指标-30和患者安全测量指标-10。
Validation of revised patient measures of safety: PMOS-30 and PMOS-10.
作者信息
Louch Gemma, Reynolds Caroline, Moore Sally, Marsh Claire, Heyhoe Jane, Albutt Abigail, Lawton Rebecca
机构信息
Yorkshire Quality and Safety Research Group, Bradford Institute for Health Research, Bradford, UK
Yorkshire Quality and Safety Research Group, Bradford Institute for Health Research, Bradford, UK.
出版信息
BMJ Open. 2019 Nov 28;9(11):e031355. doi: 10.1136/bmjopen-2019-031355.
OBJECTIVES
There is growing evidence that patients can provide feedback on the safety of their care. The 44-item Patient Measure of Safety (PMOS) was developed for this purpose. While valid and reliable, the length of this questionnaire makes it potentially challenging for routine use. Our study aimed to produce revised, shortened versions of PMOS (PMOS-30 and PMOS-10), which retained the psychometric properties of the longer version.
PARTICIPANTS
To produce a shortened diagnostic measure, we analysed data from 2002 patients who completed PMOS-44, and examined the reliability of the revised measure (PMOS-30) in a sample of 751 patients. To produce a brief standalone measure, we again analysed data from 2002 patients who completed PMOS-44, and tested the reliability and validity of the brief standalone measure (PMOS-10) in a sample of 165 patients.
METHODS
The process of shortening the questionnaire involved a combination of secondary data analysis (eg, Standard Deviation and inter-item correlations) and a consensus group exercise to produce PMOS-30 and examine face validity. Analysis of PMOS-30 data examined reliability (eg, Cronbach's alpha). Further secondary data analysis (ie, corrected item-total correlations) produced PMOS-10, and primary data collection assessed its reliability and validity (eg, Cronbach's alpha, analysis of variance).
RESULTS
Fourteen items were removed to produce PMOS-30 and the percentage of negatively worded items was reduced from 57% to 33%. PMOS-30 demonstrated good internal reliability (α=0.89). The 10 items with the highest corrected item-total correlations across both PMOS-44 and PMOS-30 composed PMOS-10. PMOS-10 had good internal reliability (α=0.79), demonstrated convergent validity; however, discriminant validity was not established.
CONCLUSIONS
Two revised, shortened versions of the original PMOS-44 (PMOS-30 and PMOS-10) were produced to capture patient feedback about safety in hospital. The measures demonstrated good reliability and validity, and preserved the psychometric properties of the original measure.
目的
越来越多的证据表明患者能够对其医疗护理的安全性提供反馈。为此开发了包含44个条目的患者安全度量表（PMOS）。尽管该问卷有效且可靠，但其长度使其在常规使用中可能具有挑战性。我们的研究旨在生成修订后的、更简短的PMOS版本（PMOS - 30和PMOS - 10），并保留较长版本的心理测量特性。
参与者
为了生成一个更简短的诊断量表，我们分析了2002名完成PMOS - 44的患者的数据，并在751名患者的样本中检验了修订后量表（PMOS - 30）的信度。为了生成一个简短的独立量表，我们再次分析了2002名完成PMOS - 44的患者的数据，并在165名患者的样本中测试了简短独立量表（PMOS - 10）的信度和效度。
方法
缩短问卷的过程包括二次数据分析（如标准差和项目间相关性）以及一个共识小组活动，以生成PMOS - 30并检验其表面效度。对PMOS - 30数据的分析检验了信度（如克朗巴哈系数）。进一步的二次数据分析（即校正后的项目总分相关性）生成了PMOS - 10，并且通过收集原始数据评估了其信度和效度（如克朗巴哈系数、方差分析）。
结果
删除了14个条目以生成PMOS - 30，负面措辞条目的比例从57%降至33%。PMOS - 30显示出良好的内部信度（α = 0.89）。在PMOS - 44和PMOS - 30中校正后的项目总分相关性最高的10个条目构成了PMOS - 10。PMOS - 10具有良好的内部信度（α = 0.79），显示出收敛效度；然而，区分效度未得到证实。
结论
生成了原始PMOS - 44的两个修订后更简短版本（PMOS - 30和PMOS - 10），以获取患者对医院安全的反馈。这些量表显示出良好的信度和效度，并保留了原始量表的心理测量特性。
Zotero 插件