Smallest detectable change and test-retest reliability of a self-reported outcome measure: Results of the Center for Epidemiologic Studies Depression Scale, General Self-Efficacy Scale, and 12-item General Health Questionnaire
Smallest detectable change and test-retest reliability of a self-reported outcome measure: Results of the Center for Epidemiologic Studies Depression Scale, General Self-Efficacy Scale, and 12-item General Health Questionnaire
复制标题
DOI：
10.1111/jep.12795
复制
发表时间：
2017-12-01
影响因子：
2.4
通讯作者：
Hirao, Kazuki
中科院分区：
文献类型：
--
作者：
Ohno, Shotaro;Takahashi, Kana;Hirao, Kazuki
基本原理、目的和目标本研究旨在检查流行病学研究中心抑郁量表 (CES-D)、一般自我效能量表 (GSES) 和 12 项一般健康问卷 (GHQ-12) 的最小可检测变化 (SDC) 和重测可靠性。方法我们在基线和两周后对 154 名年轻人进行了测试。我们使用双向随机效应模型计算了重测可靠性的组内相关系数 (ICC)，以获得一致性。然后，我们使用 ICC 公式计算了测量标准误差 (SEM) 以确保一致性。使用一致性 SEM 来计算个体水平 (SDCind) 和群体水平 (SDCgroup) 的 SDC 值。 结果 研究参与者包括 137 名年轻人。所有自我报告结果测量量表的 ICC 均超过 0.70。 CES-D 的 SEM 为 3.64，导致 SDCind 为 10.10 分，SDCgroup 为 0.86 分。 GSES 的 SEM 为 1.56，导致 SDCind 为 4.33 分，SDCgroup 为 0.37 分。双峰评分的 GHQ-12 的 SEM 为 1.47，导致 SDCind 为 4.06 分，SDCgroup 为 0.35 分。 GHQ-12 的 Likert 评分的 SEM 为 2.44，导致 SDCind 为 6.76 分，SDCgroup 为 0.58 分。结论为了确认变化不是测量误差的结果，自我报告的结果测量量表的分数需要改变大于这些 SDC 值的量。这对于临床医生和流行病学家评估结果时具有重要意义。
Rationale, aims, and objectivesThis study aims to examine the smallest detectable change (SDC) and test-retest reliability of the Center for Epidemiologic Studies Depression Scale (CES-D), General Self-Efficacy Scale (GSES), and 12-item General Health Questionnaire (GHQ-12).MethodWe tested 154 young adults at baseline and 2weeks later. We calculated the intra-class correlation coefficients (ICCs) for test-retest reliability with a two-way random effects model for agreement. We then calculated the standard error of measurement (SEM) for agreement using the ICC formula. The SEM for agreement was used to calculate SDC values at the individual level (SDCind) and group level (SDCgroup).ResultsThe study participants included 137 young adults. The ICCs for all self-reported outcome measurement scales exceeded 0.70. The SEM of CES-D was 3.64, leading to an SDCind of 10.10 points and SDCgroup of 0.86 points. The SEM of GSES was 1.56, leading to an SDCind of 4.33 points and SDCgroup of 0.37 points. The SEM of GHQ-12 with bimodal scoring was 1.47, leading to an SDCind of 4.06 points and SDCgroup of 0.35 points. The SEM of GHQ-12 with Likert scoring was 2.44, leading to an SDCind of 6.76 points and SDCgroup of 0.58 points.ConclusionTo confirm that the change was not a result of measurement error, a score of self-reported outcome measurement scales would need to change by an amount greater than these SDC values. This has important implications for clinicians and epidemiologists when assessing outcomes.