脊柱关节炎与类风湿关节炎健康相关生活质量的决定因素——来自COMOSPA与COMORA研究的数据
基于国际横断面COMOSPA和COMORA研究数据,采用多变量线性回归和决策树分析,评估脊柱关节炎(SpA)和类风湿关节炎(RA)中影响健康相关生活质量(HRQoL,以EQ-5D-3L评估)的因素。结果显示,在SpA和RA中,身体功能(BASFI和MHAQ)是HRQoL的主要贡献因素,其次为疾病活动度和工作生产力损失。
要点
- 在SpA和RA中,身体功能(BASFI和MHAQ)是EQ-5D-3L所测HRQoL的最主要决定因素。
- 决策树显示,SpA和RA中EQ-5D-3L的首要区分变量分别为BASFI和MHAQ,其后为工作生产力损失和疾病活动度。
- 多变量线性回归中,SpA的HRQoL与BASFI、ASDAS、工作生产力损失评分、NSAID治疗、bDMARD治疗、大学教育及放射学骶髂关节炎相关。
核心结论
- 在SpA和RA中,身体功能是EQ-5D-3L所测HRQoL的主要贡献因素,其次为疾病活动度和工作生产力损失。
研究设计与样本
- 研究使用国际横断面COMOSPA和COMORA研究的数据,共纳入3984例SpA患者和3920例RA患者。
- HRQoL采用EuroQOL五维度三级量表(EQ-5D-3L)评估。
- 首先采用多变量线性回归模型分析EQ-5D-3L(因变量)与人口学和临床变量(自变量)的关联。
- 其次采用卡方自动交互检测器(CHAID)构建决策树,进行无偏分层多变量分析(因变量:EQ-5D-3L)。
主要发现
- 在SpA中,HRQoL与BASFI(调整B=-0.006;95%CI=-0.007至-0.005)、ASDAS(-0.052;-0.071至-0.033)、工作生产力损失评分(-0.002;-0.003至-0.002)、NSAID治疗(-0.052;-0.083至-0.020)、bDMARD治疗(-0.051;-0.082至-0.021)、大学教育(-0.051;-0.075至-0.027)及放射学骶髂关节炎(0.035;0.004至0.030)相关。
- 在RA中,HRQoL与改良健康评估问卷(MHAQ)(-0.220,-0.253至-0.188)、DAS28-CRP-3v(-0.027,-0.036至-0.018)、工作生产力损失评分(-0.003,-0.003至-0.002)、骨侵蚀存在(-0.042,-0.065至-0.020)、每日饮酒≥3单位(0.012,0.001至0.024)及csDMARD治疗(0.034,0.001至0.066)相关。
- 决策树显示,在SpA和RA队列中,BASFI和MHAQ是对EQ-5D-3L区分力最强的首要变量,其后为工作生产力损失和疾病活动度。
局限性
- 原文未报告局限性。
以上解读依据论文摘要(全文未开放获取或许可不允许翻译转载)。
原文
OBJECTIVES: To assess the hierarchy of outcomes contributing to health-related quality of life (HRQoL) in spondyloarthritis (SpA) and rheumatoid arthritis (RA). METHODS: Data from the international cross-sectional COMOSPA and COMORA studies were used. HRQoL was assessed using the EuroQOL 5-dimension 3-level (EQ-5D-3 L). First, multivariable linear regression models were used to identify associations between EQ-5D-3 L (dependent variable) and several demographic and clinical variables (independent variables). Second, a decision tree was built using Chi-square Automatic Interaction Detector, a method of unbiased hierarchical multivariable analysis (dependent variable: EQ-5D-3 L). RESULTS: In total, 3984 patients with SpA and 3920 patients with RA were included. In SpA, HRQoL was associated with BASFI (adjusted B=-0.006; 95%CI=-0.007 to -0.005), ASDAS (-0.052; -0.071 to -0.033), work productivity loss score (-0.002; -0.003 to -0.002), NSAID treatment (-0.052; -0.083 to -0.020), bDMARD treatment (-0.051; -0.082 to -0.021), university education (-0.051; -0.075 to -0.027) and radiographic sacroiliitis (0.035; 0.004 to 0.030). In RA, HRQoL was associated with modified Health Assessment Questionnaire (MHAQ) (-0.220, -0.253 to -0.188), DAS28-CRP-3v (-0.027, -0.036 to -0.018), work productivity loss score (-0.003, -0.003 to -0.002), presence of erosions (-0.042, -0.065 to -0.020), alcohol consumption ≥3 units/day (0.012, 0.001 to 0.024)) and csDMARD treatment (0.034, 0.001 to 0.066). The decision tree revealed BASFI and MHAQ as first variables with the most discriminative power on EQ-5D-3 L, followed by work productivity loss and disease activity, in both SpA and RA cohorts. CONCLUSION: In SpA and RA, physical function is the main contributor to HRQoL measured by EQ-5D-3 L, followed by disease activity and work productivity loss.
引用
- Carvalho PD, Vieira-Sousa E, Hmamouchi I, et al. Determinants of health-related quality of life in spondyloarthritis and rheumatoid arthritis - data from the COMOSPA and COMORA studies. Semin Arthritis Rheum. 2022. PMID: 36115216. doi:10.1016/j.semarthrit.2022.152086
供执业医师工作参考,不替代临床判断;数值与分级以指南原文为准。