Effects of likelihood framing on side effect expectations and nocebo side effects: Results from three experimental studies with a placebo analgesic cream

TL;DR

This study explored how different ways of presenting side effect likelihoods affect expectations and actual side effects in a placebo analgesic cream experiment.

Contribution

The study introduces a novel experimental approach to examine how framing influences expectations and nocebo effects in non-clinical settings.

Findings

Participants had higher side effect expectations with negative framing and frequency expressions.

No significant differences in itch occurrence were found across framing conditions.

Itch reports did not differ between groups, even when likelihood was increased or not mentioned.

Abstract

This research examined whether different framings of the likelihood of side effects influence their occurrence in response to a placebo analgesic.

Three independent experimental studies in non‐clinical samples were performed.

Study 1 investigated the effects of likelihood framing on side effect expectations, while Studies 2 and 3 investigated the effects of likelihood framing on side effect experiences in terms of nocebo side effects.

Study 1 (N = 244) showed that participants had higher side expectations when the side effect likelihood was negatively framed (e.g. 1% occurrence) than positively framed (e.g. 99% no occurrence). Side effect expectations were also more negative when frequencies (e.g. 1 in 100) rather than percentages were used. Study 2 tested the effects of positive vs. negative framing and percentages vs. frequencies on the occurrence of itch as a side effect of a placebo analgesic cream, which was applied in an experimental heat pain paradigm in a non‐clinical sample (N = 179). Only 11% reported itch, with no significant differences between conditions. Study 3 (N = 110) increased the stated likelihood of itch, yet only 17% reported experiencing it. Crucially, reports of itch did not differ between experimental groups and a control group where itch was not mentioned as a possible side effect.

This research failed to induce itch as a nocebo response. While examining likelihood framing in nocebo side effects is valuable, establishing an appropriate experimental model remains challenging. Future research directions and methodological considerations to move forward are discussed.

Figures