Does Sex/Gender Play a Role in Placebo and Nocebo Effects? Conflicting Evidence From Clinical Trials and Experimental Studies

摘要

Sex has been speculated to be a predictor of the placebo and nocebo effect for many years, but whether this holds true or not has rarely been investigated. We utilized a placebo literature database on various aspects of the genuine placebo/nocebo response. In 2015, we had extracted 75 systematic reviews, meta-analyses, and meta-regressions performed in major medical areas (neurology, psychiatry, internal medicine). These meta-analyses were screened for whether sex/gender differences had been noted to contribute to the placebo/nocebo effect: in only 3 such analyses female sex was associated with a higher placebo effect, indicating poor evidence for a contribution of sex to it in RCTs. This was updated with another set of meta-analyses for the current review, but did not change the overall conclusion. The same holds true for 18 meta-analyses investigating adverse event (nocebo) reporting in RCT in the placebo arm of trials. We also screened our database for papers referring to sex/gender and the placebo effect in experimental studies, and identified 28 papers reporting 29 experiments. Their results can be summarized as follows: (a) Despite higher sensitivity of pain in females, placebo analgesia is easier to elicit in males; (b) It appears that conditioning is effective specifically eliciting nocebo effects; (c) Conditioning works specifically well to elicit placebo and nocebo effects in females and with nausea; (d) Verbal suggestions are not sufficient to induce analgesia in women, but work in men. These results will be discussed with respect to the question why nausea and pain may be prone to be responsive to sex/gender differences, while other symptoms are less. Lastly, we will discuss the apparent discrepancy between RCT with low relevance of sex, and higher relevance of sex in specific experimental settings. We argue that the placebo response is predominantly the result of a conditioning (learning) response in females, while in males it predominantly may be generated via (verbal) manipulating of expectancies. In RCT therefore, the net outcome of the intervention may be the same despite different mechanisms generating the placebo effect between the sexes, while in experimental work when both pathways are separated and explicitly explored, such differences may surface.