Public awareness of and opinions on the use of mathematical transmission modelling to inform public health policy in the United Kingdom

Abstract

AbstractMathematical transmission modelling is a key component of scientific evidence used to inform public health policy and became particularly prominent during the COVID-19 pandemic. As key stakeholders, it is vital that the public perception of this set of tools is better understood. To complement a previously published article on the science-policy interface by the authors of this study, novel data were collected via responses to a survey via two methods: via an online panel (“representative” sample) and via social media (“non-probability” sample). Many identical questions were asked separately for the period “prior to” compared to “during” the COVID-19 pandemic.All respondents were increasingly aware of the use of modelling in informing policy during the pandemic, with significantly higher levels of awareness among social media respondents than online panel respondents. Awareness generally stemmed from the news media and social media during the pandemic. Transmission modelling informing public health policy was perceived as more reliable during the pandemic compared to the pre-pandemic period in both samples, with awareness being positively associated with reliability within both samples and time points, except for social media during the pandemic. Trust in government public health advice remained high across samples and time periods overall but was lower in the period of the pandemic compared to the pre-pandemic period. The decay in trust was notably greater among social media respondents. Many respondents from both samples explicitly made the distinction that their trust was reserved for “scientists” and not “politicians”. Almost all respondents, regardless of sample, believed governments have responsibility for the communication of modelling to the public.These results provide an important reminder of the potentially skewed conclusions that could be drawn from non-representative samples.