Jaeschke R, Singer J, Guyatt G H
Department of Clinical Epidemiology and Biostatistics, McMaster University, Hamilton, Ontario, Canada.
Control Clin Trials. 1989 Dec;10(4):407-15. doi: 10.1016/0197-2456(89)90005-6.
In recent years quality of life instruments have been featured as primary outcomes in many randomized trials. One of the challenges facing the investigator using such measures is determining the significance of any differences observed, and communicating that significance to clinicians who will be applying the trial results. We have developed an approach to elucidating the significance of changes in score in quality of life instruments by comparing them to global ratings of change. Using this approach we have established a plausible range within which the minimal clinically important difference (MCID) falls. In three studies in which instruments measuring dyspnea, fatigue, and emotional function in patients with chronic heart and lung disease were applied the MCID was represented by mean change in score of approximately 0.5 per item, when responses were presented on a seven point Likert scale. Furthermore, we have established ranges for changes in questionnaire scores that correspond to moderate and large changes in the domains of interest. This information will be useful in interpreting questionnaire scores, both in individuals and in groups of patients participating in controlled trials, and in the planning of new trials.
近年来，生活质量测评工具已成为许多随机试验的主要研究结果。使用此类测评工具的研究人员面临的挑战之一是确定所观察到的任何差异的显著性，并将该显著性传达给将应用试验结果的临床医生。我们开发了一种方法，通过将生活质量测评工具的得分变化与整体变化评分进行比较，来阐明其变化的显著性。使用这种方法，我们确定了一个合理范围，最小临床重要差异（MCID）落在该范围内。在三项研究中，应用了测量慢性心肺疾病患者呼吸困难、疲劳和情绪功能的测评工具，当以七点李克特量表呈现回答时，MCID表现为每项得分的平均变化约为0.5。此外，我们还确定了问卷得分变化的范围，这些范围对应于感兴趣领域的中度和较大变化。这些信息将有助于解释个体和参与对照试验的患者群体的问卷得分，以及规划新的试验。
