Department of Psychology, University of Haifa, Mount Carmel, Haifa 31905, Israel.
Columbia University College of Physicians and Surgeons, New York State Psychiatric Institute, New York, NY, USA.
Psychol Med. 2019 Oct;49(14):2414-2420. doi: 10.1017/S0033291718003343. Epub 2018 Nov 13.
Patient expectancy is an important source of placebo effects in antidepressant clinical trials, but all prior studies measured expectancy prior to the initiation of medication treatment. Little is known about how expectancy changes during the course of treatment and how such changes influence clinical outcome. Consequently, we undertook the first analysis to date of in-treatment expectancy during antidepressant treatment to identify its clinical and demographic correlates, typical trajectories, and associations with treatment outcome.
Data were combined from two randomized controlled trials of antidepressant medication for major depressive disorder in which baseline and in-treatment expectancy assessments were available. Machine learning methods were used to identify pre-treatment clinical and demographic predictors of expectancy. Multilevel models were implemented to test the effects of expectancy on subsequent treatment outcome, disentangling within- and between-patient effects.
Random forest analyses demonstrated that whereas more severe depressive symptoms predicted lower pre-treatment expectancy, in-treatment expectancy was unrelated to symptom severity. At each measurement point, increased in-treatment patient expectancy significantly predicted decreased depressive symptoms at the following measurement (B = -0.45, t = -3.04, p = 0.003). The greater the gap between expected treatment outcomes and actual depressive severity, the greater the subsequent symptom reductions were (B = 0.49, t = 2.33, p = 0.02).
Greater in-treatment patient expectancy is associated with greater subsequent depressive symptom reduction. These findings suggest that clinicians may benefit from monitoring and optimizing patient expectancy during antidepressant treatment. Expectancy may represent another treatment parameter, similar to medication compliance and side effects, to be regularly monitored during antidepressant clinical management.
患者期望是抗抑郁药临床试验中安慰剂效应的一个重要来源，但之前所有的研究都是在开始药物治疗之前测量期望。关于期望在治疗过程中如何变化以及这种变化如何影响临床结果，我们知之甚少。因此，我们首次分析了抗抑郁药物治疗期间的治疗中期望，以确定其临床和人口统计学相关性、典型轨迹以及与治疗结果的关联。
数据来自两项抗抑郁药物治疗重度抑郁症的随机对照试验，其中有基线和治疗中期望评估。使用机器学习方法来识别治疗前临床和人口统计学因素对期望的预测作用。实施多层次模型来测试期望对随后治疗结果的影响，分解患者内和患者间的影响。
随机森林分析表明，虽然更严重的抑郁症状预示着更低的治疗前期望，但治疗中期望与症状严重程度无关。在每个测量点，治疗中患者期望的增加都显著预测了下一次测量时抑郁症状的减轻（B = -0.45，t = -3.04，p = 0.003）。期望的治疗结果与实际抑郁严重程度之间的差距越大，随后的症状减轻就越大（B = 0.49，t = 2.33，p = 0.02）。
治疗中患者期望越高，随后的抑郁症状减轻越大。这些发现表明，临床医生可能受益于在抗抑郁治疗期间监测和优化患者期望。期望可能代表另一个治疗参数，类似于药物依从性和副作用，在抗抑郁临床管理中需要定期监测。
