作者: Ayse Kuspinar , Nancy E. Mayo
DOI: 10.1007/S40273-014-0167-5
关键词: Validity 、 Psychometrics 、 Scale (social sciences) 、 Data mining 、 Convergent validity 、 Clinical psychology 、 Health Utilities Index 、 Reliability (statistics) 、 Content validity 、 Quality-adjusted life year 、 Medicine
摘要: The reliability and validity of generic utility measures have not yet been summarized in people with multiple sclerosis (MS). It is important to assess the psychometric properties these measures, ensure that values obtained by scoring system are valid for interpretation utilization clinicians, researchers policy makers. Therefore, objective this review was summarize evidence from published literature on MS. A structured search conducted using electronic databases. All potentially relevant abstracts full-text articles were read identify publications may be eligible inclusion review. meta-analysis combine correlation coefficient convergent validity. Schmidt–Hunter method, a weighted mean values, used. Heterogeneity, percentage total variation across studies due between-study differences rather than chance, assessed I 2 statistic. following identified: EQ-5D (n = 9)/EQ-5D-5 Level (EQ-5D-5L) (n = 1), followed Health Utilities Index Mark 3/2 (HUI2/HUI3) (n = 3), SF-6D (n = 2), Assessment Quality Life (AQOL) Well-Being (QWB) scale (n = 1). Ceiling floor effects present SF-6D, but HUI3. EQ-5D, HUI3 demonstrated excellent reliability. In terms discriminative ability, QWB able differentiate between moderately severely disabled MS patients, those who mildly disabled. AQOL HUI3, other hand, good as both all levels disability. As validity, HUI2/HUI3 highly correlated (r = 0.7) against measurement instruments evaluated impairments such disease severity, ambulation manual dexterity. small moderate correlations (r = 0.4) evaluating impairments, slightly stronger activity limitations/participation restrictions health-related quality life (HRQL) (r = 0.6). To our knowledge first study strongest when compared measures. However, only impairment excludes components HRQL participation restrictions. scale, do include items participation. lack content missing certain domains disease, well difficulty differentiating different addition MS-specific ‘bolt-ons’ development an specific measure possible areas exploration future research.