Despite its widespread global adoption and use, studies have not examined measurement invariance of the 31-question World Health Organization Quality of Life–HIV BREF scale among HIV/AIDS patients. The current study seeks to (a) evaluate the scale’s internal consistency reliability, and concurrent validity, and (b) test if the same latent construct of quality of life was consistently measured at two time-points, for a sample of HIV-positive women from two sites in Addis Ababa, Ethiopia. The study used data from two cross-sectional interviews with 926 HIV-positive women of reproductive age who participated in a quasi-experimental study. All participants were receiving antiretroviral therapy and related treatment support services from 51 service providers in two non-contiguous sub-cities. We used One-Way ANOVA, chi square test and Kruskal Wallis test to compare demographic characteristics and quality of life scores of study participants. Further, we used Cronbach’s coefficient alpha (α) to assess internal consistency reliability and Pearson product-moment correlation (r) to assess concurrent validity. Finally, multiple-group confirmatory factor analysis with maximum likelihood estimation was used to assess measurement invariance of the quality of life scale. Findings suggest that the WHOQOL-HIV BREF exhibited acceptable psychometric properties. There was evidence for strong internal consistency reliability demonstrated by Cronbach’s α >0.80 and coefficient ω >0.80. The scale’s six domains also exhibited good concurrent validity, with coefficient r = 0.63–0.82. In measurement invariance analysis, configural invariance was found for the scale’s six domains, suggesting measurement noninvariance for factor loadings, item intercepts and factor variances when comparing QOL among participants in the two sites at baseline and follow-up. Our findings suggest that the WHOQOL-HIV BREF has internal consistency reliability and concurrent validity in this study sample. The differences in the levels of observed responses for QOL across intervention and non-intervention groups at baseline and follow-up suggest participants had dissimilar understanding, interpretation, and responses to the scale’s items, which may have resulted from significant measurement, cultural, and developmental differences between study groups. The WHOQOL-HIV BREF exhibited internal consistency reliability, concurrent validity and configural invariance. However, strong invariance was not achieved, making it difficult to compare levels of QOL between groups in this study sample.

尽管它在全球范围内被广泛采用和使用，但研究并未检查包含 31 个问题的世界卫生组织生活质量 - HIV BREF 量表在 HIV/AIDS 患者中的测量不变性。目前的研究旨在 (a) 评估量表的内部一致性信度和同时效度，以及 (b) 测试是否在两个时间点一致测量了相同的潜在生活质量结构，对于 HIV 阳性女性样本来自埃塞俄比亚亚的斯亚贝巴的两个站点。该研究使用了对参与准实验研究的 926 名 HIV 阳性育龄妇女进行的两次横断面访谈的数据。所有参与者都接受了来自两个不相邻子城市的 51 家服务提供者的抗逆转录病毒治疗和相关治疗支持服务。我们使用了单向方差分析，卡方检验和 Kruskal Wallis 检验比较研究参与者的人口统计学特征和生活质量分数。此外，我们使用 Cronbach 系数 alpha (α) 来评估内部一致性可靠性和 Pearson 积矩相关性 (r) 来评估并发有效性。最后，使用最大似然估计的多组验证性因素分析来评估生活质量量表的测量不变性。结果表明，WHOQOL-HIV BREF 表现出可接受的心理测量特性。Cronbach's α > 0.80 和系数 ω > 0.80 证明了强大的内部一致性可靠性的证据。该量表的六个领域也表现出良好的并发效度，系数为 我们使用 Cronbach 系数 alpha (α) 来评估内部一致性信度和 Pearson 积矩相关性 (r) 来评估并发效度。最后，使用最大似然估计的多组验证性因素分析来评估生活质量量表的测量不变性。结果表明，WHOQOL-HIV BREF 表现出可接受的心理测量特性。Cronbach's α > 0.80 和系数 ω > 0.80 证明了强大的内部一致性可靠性的证据。该量表的六个领域也表现出良好的并发效度，系数为 我们使用 Cronbach 系数 alpha (α) 来评估内部一致性信度，使用 Pearson 积矩相关性 (r) 来评估并发效度。最后，使用最大似然估计的多组验证性因素分析来评估生活质量量表的测量不变性。结果表明，WHOQOL-HIV BREF 表现出可接受的心理测量特性。Cronbach's α > 0.80 和系数 ω > 0.80 证明了强大的内部一致性可靠性的证据。该量表的六个领域也表现出良好的并发效度，系数为 使用最大似然估计的多组验证性因素分析来评估生活质量量表的测量不变性。结果表明，WHOQOL-HIV BREF 表现出可接受的心理测量特性。Cronbach's α > 0.80 和系数 ω > 0.80 证明了强大的内部一致性可靠性的证据。该量表的六个领域也表现出良好的并发效度，系数为 使用最大似然估计的多组验证性因素分析来评估生活质量量表的测量不变性。结果表明，WHOQOL-HIV BREF 表现出可接受的心理测量特性。Cronbach's α > 0.80 和系数 ω > 0.80 证明了强大的内部一致性可靠性的证据。该量表的六个领域也表现出良好的并发效度，系数为r = 0.63–0.82。在测量不变性分析中，在量表的六个领域中发现了配置不变性，这表明在比较两个站点参与者在基线和随访时的 QOL 时，因子负荷、项目截距和因子方差的测量非不变性。我们的研究结果表明，WHOQOL-HIV BREF 在本研究样本中具有内部一致性信度和并发效度。干预组和非干预组在基线和随访时观察到的 QOL 反应水平的差异表明参与者对量表的项目有不同的理解、解释和反应，这可能是由于重要的测量、文化和发展研究组之间的差异。WHOQOL-HIV BREF 表现出内部一致性的可靠性，并发有效性和配置不变性。然而，并未实现强不变性，因此难以比较本研究样本中各组之间的 QOL 水平。