“Solidarity is No One-Way Street” – The Effects of COVID-19 Framings on Attitudes Towards Cost-Sharing for Inpatient Treatment in Germany

In 2022, SARS-CoV-2 vaccines were available in Germany. Nevertheless, ∼22 % of the population remained unvaccinated. Meanwhile, hospitalization costs due to severe COVID-19 were significant. In consequence, public debate emerged over whether unvaccinated individuals should contribute towards their (often significant) hospitalization costs. However, it remained unclear what public attitudes were towards this suggestion, which is important as political consequences, such as fierce popular resistance and pertinent costs, can significantly outweigh the benefits, if popular support is lacking in democratic societies. We, therefore, assessed public support for requiring unvaccinated individuals to contribute towards their COVID-19 hospitalization costs and examined how different frames influence this support. A randomized, controlled survey experiment was conducted in July 2022 (n = 6,725 Germans). Respondents answered whether unvaccinated individuals hospitalized due to COVID-19 should contribute to treatment costs and, if so, the proportion they deem adequate. Participants were randomized to one of five study arms: control group (no framing) or one of four experimental groups with frames highlighting individual or collective, and economic or health, consequences of severe COVID-19 or Long COVID. Due to the strong moral polarisation between SARS-CoV-2 vaccination advocates and opponents, which likely also affects views on cost-sharing, this analysis can be understood as a least likely design, since, under these circumstances, no significant effect of any kind of political framing can be assumed. In the control group, 46.6 % supported cost-sharing by unvaccinated, with an average contribution of 45.5 %. Economic framing had a modest but statistically significant positive effect: one frame increased support by 27 % (p = 0.01) compared to the control (no a framing message). Framing did not significantly influence the size of the contribution deemed appropriate. Compared to some of the controls, and in line with the assumption of the least-likely design, the found framing effects were not particularly strong. The most influential predictor was vaccination status: fully vaccinated individuals were 96-times more likely to support cost-sharing than unvaccinated peers. Framing effects were moderated by political affiliation and personality traits. Long COVID frames had no significant overall effect but produced divergent effects across political affiliations. For example, Long COVID Frame 1 decreased support among supporters of the Christian Democrats (CDU/CSU) but increased it among supporters of the Social Democrats (SPD). While support for requiring unvaccinated individuals to share COVID-19 treatment costs is primarily driven by personal vaccination status, messaging effects also depend on political orientation and personality traits. Since even small changes in public health compliance can already make an important difference with regard to reducing infectious disease transmission or achieving herd immunity, these comparatively small framing effects that we find are nevertheless important – particularly as they present one of the very few relatively cost-effective and simple options that policymakers can take to combat a pandemic and they stand in stark contrast to harsh policies that involve severe restrictions on personal freedoms (such as compulsory vaccination, curfews, etc.). Public health strategies aiming to increase vaccine uptake or justify cost-sharing policies may benefit from targeted communication strategies that align with audience-specific values and partisanship.