SSM: Population Health (Mar 2022)
Manufacturing doubt: Assessing the effects of independent vs industry-sponsored messaging about the harms of fossil fuels, smoking, alcohol, and sugar sweetened beverages
Abstract
Background Manufacturers of harmful products engage in misinformation tactics long employed by the tobacco industry to emphasize uncertainty about scientific evidence and deflect negative attention from their products. This study assessed the effects of one type of tactic, the use of “alternative causation” arguments, on public understanding. Methods In five trials (one for each industry) anonymized Qualtrics panel respondents were randomized to receive a message on the risk in question from one of four industry sponsored organizations (exposure), or from one of four independent organizations (control), on risks related to alcohol, tobacco, fossil fuel and sugar sweetened beverages. Logistic regression models were used to evaluate the effect of industry arguments about uncertainty on the primary outcome of public certainty about product risk, adjusting for age, gender and education. The results from all five trials were pooled in a random-effects meta-analysis. Findings In total, n=3284 respondents were exposed to industry-sponsored messaging about product-related risks, compared to n=3297 exposed to non-industry messages. Across all industries, exposure to industry-sponsored messages led to greater reported uncertainty or false certainty about risk, compared to non-industry messages [Summary odds ratio (OR) 1·60, confidence interval (CI) 1·28–1·99]. The effect was greater among those who self-rated as not/slightly knowledgeable (OR 2·24, CI 1·61–3·12), or moderately knowledgeable (OR 1·85, CI 1·38–2·48) compared to those very/extremely knowledgeable (OR 1·28, CI 1·03–1·60). Conclusions This study demonstrates that exposure to industry sponsored messages which appear intended to downplay risk significantly increases uncertainty or false certainty, with the effect being greater in less knowledgeable participants.