Harnessing the placebo effect: Exploring the influence of physician characteristics on placebo response

Abstract

Objective: Research on placebo/nocebo effects suggests that expectations can influence treatment outcomes, but placebo/nocebo effects are not always evident. This research demonstrates that a provider's social behavior moderates the effect of expectations on physiological outcomes.

Methods: After inducing an allergic reaction in participants through a histamine skin prick test, a health care provider administered a cream with no active ingredients and set either positive expectations (cream will reduce reaction) or negative expectations (cream will increase reaction). The provider demonstrated either high or low warmth, or either high or low competence.

Results: The impact of expectations on allergic response was enhanced when the provider acted both warmer and more competent and negated when the provider acted colder and less competent.

Conclusion: This study suggests that placebo effects should be construed not as a nuisance variable with mysterious impact but instead as a psychological phenomenon that can be understood and harnessed to improve treatment outcomes. (PsycINFO Database Record

(c) 2017 APA, all rights reserved).

Figures

Figure 1.

Participants’ postexam ratings of health care provider warmth and competence. Panel (A) displays rated competence (collapsing across warmth condition), showing a main effect such that participants rated the health care provider as more competent in the high competence condition. Panel (B) displays rated warmth (collapsing across competence condition), showing a main effect such that participants rated the health care provider as warmer in the high warmth condition. All questions related to warmth and competence were statements (e.g., “The medical assistant was warm,” with response options from 1 = strongly disagree to 7 = strongly agree). Standard errors represent ±1 SE N = 159. ** p < .01. *** p < .001.

Figure 2.

Difference in the size of the wheal as a function of expectations about the effects of a placebo cream delivered just before Minute 0 (Time 3), 9 min after the skin prick test. Panel (A) displays adjusted means at Time 5, 6 min after the application of the placebo cream and 15 min after the skin prick test. Nonoverlapping error bars indicate reliable differences at the p < .05 level. Panel (B) displays the longitudinal trajectory of wheal size from Time 3 to Time 5 for participants given positive expectations and negative expectations. The difference between the slopes is significant at the p < .05 level. All adjusted means are derived from the same multilevel longitudinal model, which controls for participants’ gender (−1 = male, +1 = female), ethnicity (−1 = white, +1 = nonwhite), and average wheal size at Times 1 and 2 (3 and 6 min after the skin prick test, respectively), grand-mean centered. Standard errors represent ±1 SE N = 159.

Figure 3.

Difference in wheal size as a function of positive and negative expectations, within each of the four health care provider interaction styles defined according to two dimensions: high or low warmth, and high or low competence. Panel (A) displays adjusted means at Time 5, 6 min after the application of the placebo cream and 15 min after the skin prick test. Nonoverlapping error bars indicate reliable differences at the p < .05 level. Panel (B) displays the longitudinal trajectory of wheal size from Time 3 to Time 5. The difference between the slopes in the high-warmth, high-competence condition compared to the both low condition is significant at the p < .05 level. All adjusted means are derived from the same multilevel longitudinal model, which controls for participants’ gender (−1 = male, +1 = female), ethnicity (−1 = white, +1 = nonwhite), and average wheal size at Times 1 and 2 (3 and 6 min after the skin prick test, respectively), grand-mean centered. Standard errors represent ±1 SE N = 159.

Figure 4.

Difference in wheal size of each provider interaction style relative to the average wheal size, controlling for expectations. Panel (A) displays the relevant comparisons for the placebo effect. Nonoverlapping error bars indicate reliable differences at the p < .05 level. Panel (B) displays the relevant comparisons for the nocebo effect. Both figures display adjusted means from the same longitudinal model as for Figures 1 and 2. Standard errors represent ±1 SE N = 159.