Department of Psychology, Wayne State University, Detroit, MI, USA.
Center for Health Assessment Research and Translation, University of Delaware, Newark, DE, USA.
Qual Life Res. 2022 Jul;31(7):2223-2233. doi: 10.1007/s11136-022-03114-1. Epub 2022 Mar 14.
The present study tested the fit and comparability of the tripartite model of health (Physical Health, Mental Health, and Social Health) proposed by the NIH PROMIS for adults with SCI and TBI.
Participants were 630 adults with spinal cord injury (SCI; n = 336) and traumatic brain injury (TBI; n = 294) who completed 8 PROMIS short forms. The Physical Health domain is composed of the Physical Function, Pain Interference, and Fatigue scales. The Mental Health domain included the Depression, Anxiety, and Anger scales. Social Health included the Social Emotional Support scale.
Confirmatory factor analyses supported the tripartite model of health over a unifactorial model of health for both SCI and TBI groups. Measurement invariance testing indicated the tripartite model met the level of configural and metric invariance for TBI and SCI groups, suggesting comparable structure and factor loadings. Failure to meet the level of scalar invariance indicated unequal intercepts across groups. Physical Function was identified as the source of noninvariance, and a partial scalar invariance model permitting different Physical Function intercepts across conditions was supported.
Consistent with theory, findings supported construct validity of the PROMIS tripartite structure of health composed of Physical, Mental, and Social Health. PROMIS measures appeared to tap domains of health consistent with what is accepted for SCI and TBI populations, although the measurement of Physical Function was not equivalent across groups. Findings support the utility of PROMIS broadly as well as the need for condition-optimized measurement.
本研究测试了 NIH PROMIS 提出的适用于 SCI 和 TBI 成年人的健康三因素模型（身体健康、心理健康和社会健康）的适配性和可比性。
参与者为 630 名脊髓损伤（SCI；n=336）和创伤性脑损伤（TBI；n=294）成年人，他们完成了 8 项 PROMIS 短表。身体健康域由身体功能、疼痛干扰和疲劳量表组成。心理健康域包括抑郁、焦虑和愤怒量表。社会健康包括社会情感支持量表。
验证性因子分析支持 SCI 和 TBI 组的健康三因素模型优于健康单因素模型。测量不变性测试表明，三因素模型满足 TBI 和 SCI 组的结构和因子负荷的配置和度量不变性水平，表明结构和因子负荷具有可比性。未能达到标度不变性水平表明组间截距不等。身体功能被确定为非不变性的来源，支持允许不同条件下身体功能截距不同的部分标度不变性模型。
与理论一致，研究结果支持 PROMIS 健康三因素结构（身体、心理和社会健康）的结构有效性。PROMIS 测量似乎可以测量与 SCI 和 TBI 人群接受的一致的健康领域，尽管身体功能的测量在不同组之间并不等效。研究结果支持 PROMIS 的广泛应用，以及对条件优化测量的需求。
