Nudges didn't work with vaccination intentions against COVID-19

Can vaccination intentions against COVID-19 be nudged? Elena Kantorowicz-Reznichenko, Jaroslaw Kantorowicz, Liam Wells. Behavioural Public Policy, July 8 2022. https://doi.org/10.1017/bpp.2022.20

Abstract: Once vaccines against COVID-19 became available in many countries, a new challenge has emerged – how to increase the number of people who vaccinate? Different policies are being considered and implemented, including behaviourally informed interventions (i.e., nudges). In this study, we have experimentally examined two types of nudges on representative samples of two countries – descriptive social norms (Israel) and saliency of either the death experience from COVID-19 or its symptoms (UK). To increase the legitimacy of nudges, we have also examined the effectiveness of transparent nudges, where the goal of the nudge and the reasons of its implementation (expected effectiveness) were disclosed. We did not find evidence that informing people that the vast majority of their country-people intend to vaccinate enhanced vaccination intentions in Israel. We also did not find evidence that making the death experience from COVID-19, or its hard symptoms, salient enhanced vaccination intentions in the UK. Finally, transparent nudges as well did not change the results. We further provide evidence for the reasons why people choose not to vaccinate, and whether different factors such as gender, belief in conspiracy theories, political ideology, and risk perception, play a role in people's intentions to vaccinate or susceptibility to nudges.

Discussion and conclusion

In this article, we aimed to examine a number of soft interventions to increase people's intention to vaccinate. Vaccination is currently considered the main solution to the global pandemic and vaccinating the majority of the population is a key public policy goal. At least in democratic countries, governments do not wish to force people to vaccinate, respecting their freedoms and rights over their bodies. Therefore, other methods to encourage vaccination are necessary. Nudging is one popular method, which has been used in many countries around the world for different public policies (e.g., increasing tax compliance, organ donation, savings). Therefore, it has been also discussed and considered in the context of vaccination against COVID-19.

We have experimentally examined three nudges, in two countries, which at the initial period of vaccination availability appeared to be leading in rates of vaccination. Those countries are also similar in terms of their public being generally supportive of different nudges (Reisch & Sunstein, Reference Reisch and Sunstein2016; Pe'er et al., Reference Pe'er, Feldman, Gamliel, Sahar, Tikotsky, Hod and Schupak2019). General support of nudges was also found in many other countries (e.g., Jung & Mellers, Reference Jung and Mellers2016; Sunstein et al., Reference Sunstein, Reisch and Rauber2018, Reference Sunstein, Reisch and Kaiser2019), thus constituting an instrument that has a potential to direct people's behaviour across different countries. First, we examined the effectiveness of a descriptive social norm on a representative sample of the Israeli (Jewish)Footnote¹⁸ population. In particular, we have utilized findings from psychology indicating that people follow the behaviour of others, in order to encourage vaccination intentions. Overall, the average intention of people to vaccinate was high, which is a promising result. However, the nudge itself did not make a difference. Also, a more ‘legitimate’ nudge which was transparent about its method and goal did not change people's choices.

We have also run an experiment to examine two additional nudges on a representative sample of the population in the UK. This experiment used the saliency nudge, utilizing findings from psychology that making certain factors more salient might affect how people treat probabilities, and in turn, which choices they make. The experiments made either the death experience very vivid and alarming or stressed the symptoms of COVID-19. Also, in the UK already in the baseline people had high intentions to vaccinate on average. The saliency nudges had very limited effect to none at all. In particular, when considering the full sample, none of the nudges changed the choices. Looking at the restricted sample (those participants who passed the manipulation check) showed a very small effect of the transparent death saliency nudge. Even though statistically significant, the small effect does not seem to be promising. In addition, we have examined the choices of different subgroups in the two countries but found no differences in the effectiveness of the nudge.

Our studies were well powered as we have based our sample sizes on a power analysis. Therefore, the null results are unlikely to be the result of lack of statistical power. One explanation might be that nudges are effective when people do not have strong preferences either way. In that case, it is not costly to follow a certain nudge. Some studies suggested that the inability of the nudge to change the behaviour of some people might be derived from a stronger preference of those people against the direction of the nudge (Bronchetti et al., Reference Bronchetti, Dee, Huffman and Magenheim2013; Beshears et al., Reference Beshears, Choi, Laibson, Madrian and Milkman2015; Jachimowicz et al., Reference Jachimowicz, Duncan, Weber and Johnson2019). For example, Bronchetti et al. (Reference Bronchetti, Dee, Huffman and Magenheim2013) raised the possibility that people with lower income are more resistant to default nudges that direct them to allocate money for savings because they already have plans how to use this money.

The context of the new vaccination is sensitive. On the one hand, there is an ongoing (threatening) pandemic with many uncertainties in respect of its long-term effects. On the other hand, the developed vaccine (which at the time of the study was not approved yet) is novel and entails many uncertainties with respect to the long-term effects. People are either more afraid of the former, or more of the latter. Therefore, it is difficult to affect their choices with nudges that target their intuitive system of decision making, instead of the deliberative process of decision making. In other words, it might be necessary to first address people's concerns, before having an effect on their behaviour. Our results seem to support the recently expressed opinion of one of the ‘founders’ of the concept of nudges himself, that nudges might not be a sufficient instrument to enhance vaccination to end the current pandemic,Footnote¹⁹ even though there is evidence that at least using reminders and simplifying the process of vaccination has a positive effect (Dai et al., Reference Dai, Saccardo, Han, Roh, Raja, Vangala, Modi, Pandya, Sloyan and Croymans2021).

Our results demonstrate that many people are willing to vaccinate. But there is also a smaller group which is hesitant. From our investigation of the reasons for this hesitation, it seems that the primary reason is the concern about the side effects. Even at the stage of our experiments it was clear that once the vaccine will be available, there will not be sufficient evidence of their long-term effects. This is due to the urgency in approving this vaccine and saving the world from the pandemic. Therefore, this concern is understandable.

Consequently, to encourage vaccination, governments should invest more in understanding people's concerns and trying to address them. For example, by investing in campaigns where people receive more information on the trade-offs between the uncertain long-term effects of the vaccine, and the uncertain (probably worse) long-term effects of contracting COVID-19. Relying solely on soft interventions such as nudges seems not to be sufficient.

One limitation of this study is that we focus on people's intentions rather than vaccination uptake, thus potentially facing the problem of intention-behaviour gap (Sheeran, Reference Sheeran2002). In the specific context of vaccinations for example, several studies have found that even when people intend to vaccinate, they do not always follow through (e.g., Bronchetti et al., Reference Bronchetti, Dee, Huffman and Magenheim2013; Chang et al., Reference Chang, Jacobson, Shah, Pramanik and Shah2021). Nevertheless, there are many studies demonstrating that an intention to vaccinate is generally a strong predictor of an actual uptake of vaccines (daCosta DiBonaventura & Chapman, Reference daCosta DiBonaventura and Chapman2005; Lehmann et al., Reference Lehmann, Ruiter, Chapman and Kok2014; Fall et al., Reference Fall, Izaute and Chakroun-Baggioni2018; Jensen et al., Reference Jensen, Ayers and Koskan2022). In the specific context of this article, at the time of the study we have found that 63.6% of the Israeli participants and 75.3% of the UK participants (control groups), respectively, either ‘strongly agreed’ or ‘agreed’ that they would vaccinate themselves against COVID-19. Looking at the most recent data to date of vaccination uptake we see that nearly 77 persons per 100 population have taken the first dose and around 69 per 100 are fully vaccinated in Israel. In the UK, almost 78 persons per 100 population have taken the first dose, and nearly 73 in 100 are fully vaccinated.Footnote²⁰ Therefore, the baseline intentions in our studies (which were expressed before the vaccine became available) seem to be overall aligned with the actual rate of vaccination as reported by WHO for the beginning of the year 2022.

Furthermore, even though we do not have a way to directly translate our results from intentions to behaviour, our findings seem to be conservative in this respect. From the studies of the intention-behaviour gap, it seems that the gap is mostly driven by people who intend to act but eventually fail to do so (Sheeran, Reference Sheeran2002).Footnote²¹ Therefore, it is reasonable to expect that if the investigated nudges in this article did not change people's intention, it probably would not change people's behaviour.

A related potential concern is that currently there is more knowledge about the effectiveness of the vaccine and its short-term safety.Footnote²² The fact that uncertainty regarding those two factors were the main reasons for some of our subjects not to intend to vaccinate, might suggest our results would change at this stage. However, the level of uncertainty was similar for all participants in our studies. And yet many indicated an intention to vaccinate. Those who were hesitant or reluctant did not seem on average to change their minds in response to the employed nudges. Whereas the new information may have on itself encouraged people to vaccinate, there is no immediate reason to believe it would influence the level of effectiveness of the nudge. For example, in Israel, even after the vaccines against COVID-19 were made available, and evidence of its efficacy and short-term safety had emerged, the two main concerns of the people who were still hesitant about vaccination remained its effectiveness and safety (e.g., Heller et al., Reference Heller, Chun, Shlomo, Gewirtz-Meydan, Acri, Kulkarni and Grinstein-Weiss2022). Also in the UK, the long-term safety was still a major concern for people who were choosing not to vaccinate themselves (Majeed et al., Reference Majeed, Papaluca and Molokhia2021). However, Majeed et al. (Reference Majeed, Papaluca and Molokhia2021) stated that the emerging data on the benefits of the vaccine also reduce vaccine hesitancy. Therefore, it might be reasonable to assume, that those who are still hesitating at this stage, more than a year after vaccines were introduced and with the current reliable information on the effectiveness and short-term safety of vaccines, are those who hold stronger preference against vaccinating. Consequently, our results might again be viewed as conservative, and suggest that other strategies, which are addressing the persisting concerns are needed, rather than simply using ‘system 1’ nudges to enhance vaccination.