Addressing Vaccine Hesitancy: Survey and Experimental Evidence from Papua New Guinea

"The central findings of the representative phone survey in PNG illustrate the importance of trust in the COVID-19 vaccine and social norms in driving behavior."

In addition to structural barriers and limitations of health systems in low- and middle-income countries (LMICs), a key reason for the lack of use of COVID-19 vaccines has been hesitancy among eligible populations. By reporting on the results of an observation phone survey study, this paper examines the drivers of COVID-19 vaccine hesitancy in Papua New Guinea (PNG) and tests various means of increasing people's willingness to receive a COVID-19 vaccine. As of this writing, COVID-19 vaccination rates in PNG were among the lowest in the world, and existing evidence suggested extremely high levels of vaccine hesitancy by international standards.

To ground the study, the researchers provide a literature review that is broken into two parts. First, they review the small number of descriptive, observational studies of attitudes toward COVID-19 vaccines that come from LMICs. Second, they review the more explicitly causal literature that attempts to explain why vaccine hesitancy exists and whether it can be changed. They do not limit coverage either to COVID-19 or developing countries; neither has been studied enough. They use this broader literature to feed into the conceptual framework used in the experimental component of their own research.

The study draws on data collected through a broadly representative phone survey conducted in May and early June 2021 with 2,533 respondents and an online randomised survey experiment conducted in late June and throughout July 2021 with 2,392 participants in PNG. The survey included questions about people's willingness to get a COVID-19 vaccine and the factors that could be contributing to this, such as beliefs about the behaviour of others and people's most trusted sources of information about the vaccine. The results were weighted to match the general population. The online randomised survey experiment trialed whether messages informed by the findings of the phone surveys could reduce people's hesitancy about getting the COVID-19 vaccine. Specifically, the treatments that were tested referred to expert advice, social norms, and the relative safety of the vaccine. (The World Bank ran a series of Facebook experiments about vaccine hesitancy in the first half of 2021 in the Middle East and North Africa using a range of behaviourally informed messages. These treatments were the starting point for this study, but questions were tailored to suit the PNG context based on consultation with stakeholders.)

Both surveys showed that less than 20% of the respondents who were aware a vaccine existed were willing to be vaccinated. The main reason respondents stated for their hesitancy regarding the vaccine was concern about side effects; however, the majority also said health workers could change their mind, particularly if information was communicated face to face (in person).

Regression analysis shows that after controlling for demographic characteristics, respondents who thought their family and friends would get vaccinated were 65.9 percentage points more likely to plan on being vaccinated. A similar, but weaker, relationship existed when people were asked their beliefs about whether most people in PNG would be vaccinated. People's willingness to get vaccinated was also strongly related to whether they would encourage a family member to get vaccinated. For example, 57.4% of respondents who planned on getting vaccinated stated they would encourage a family member to get vaccinated, whereas around 3.1% of people who did not plan on being vaccinated said this was the case. This finding suggests that people's level of vaccine hesitancy is not only shaped by others' behaviour, but is also likely to impact the behaviour of others.

In addition to the role of beliefs about the behaviour of others and the role of encouraging others to get vaccinated, the phone survey illustrated that people's level of trust in the vaccine is strongly associated with their intention to get a COVID-19 vaccine. However, there was no clear relationship between other potential explainers of vaccine hesitancy, such as being concerned about COVID-19, people's most trusted source of information about the COVID-19 vaccine (including social media), their history of receiving vaccines for other diseases (for example, polio), and their willingness to be vaccinated.

The online experiment showed that providing a short message about the relative safety of the COVID-19 vaccine (designed with the aim of building trust in the vaccine) substantially increased people's willingness to get vaccinated. Only 19.3% of respondents in the control group (that is, where no intervention was provided) stated they would be willing to get the vaccine, whereas 30.6% of respondents who received the relative safety treatment said they would get vaccinated, representing an over 50% increase. The two other intervention arms that were provided (referring to expert advice and social norms) did not have a consistent impact on people's willingness to get vaccinated. The main findings of this experiment did not vary substantially across the demographics of respondents, though female respondents and those in the Southern region were more likely to respond to the social norms treatment. These types of respondents were also less likely to participate in the online experiment relative to the general population. As a result, after population weights were applied (in terms of sex, age, and location), the social norms treatment became statistically significant.

The study highlights three key areas for policymakers to prioritise as they attempt to reduce COVID-19 vaccine hesitancy in PNG:

1. Boosting trust in COVID-19 vaccines appears to be a fundamental step in increasing willingness to get vaccinated. Policymakers should inform the general population that COVID-19 vaccines can be effective at preventing severe illness, that there are very low risks of developing negative side effects, and that becoming ill with COVID-19 results in a non-trivial chance of death for those who are unvaccinated. The finding that the information treatment used in the relative safety treatment was clearly effective in reducing COVID-19 vaccine hesitancy "is good news. It suggests that at least some of the apprehension about COVID-19 vaccines present in PNG stems simply from the fact that people do not feel they know enough about the new vaccine, as opposed to entrenched 'anti-vaccination' positions of the sort found in other countries. As a result, it may prove feasible to shift people's views on COVID-19 vaccines" in PNG.
2. The importance of social norms in shaping people's willingness to be vaccinated is reaffirmed by the findings of the phone survey. Policymakers should pair communications with other policies that will encourage vaccine take-up in order to create a virtuous cycle where increasing take-up shifts norms and encourages more take-up. "It is possible that different types of information may be more effective in transmitting a norm-based message, and there is a strong case for further research in this area. However, it may well be that some sectors of the population will ultimately need to see firsthand their family, friends and others in their community receiving the vaccine before norms become truly effective in shaping attitudes. Although this would be a discouraging finding for those who hope to use information interventions to help increase vaccine uptake, it would be a positive finding in another sense: vaccinating those who are already willing to be vaccinated may itself help to change the attitudes of others as they see their fellow community members safely taking the vaccine."
3. Health workers have a prominent role to play in increasing people's willingness to receive a vaccine for COVID-19. "One potential complicating factor facing authorities is that vaccine hesitancy is comparatively high among health workers....This suggests that high priority should be given to educating local health workers about the benefits of the COVID-19 vaccine (including the low risk of side effects) so they can become advocates in their community for vaccination."

This study also illustrates that three areas that are often cited as contributing to vaccine hesitancy may play less of an instrumental role - at least in PNG - than commonly thought. They include: People in PNG do not seem to be hesitant because of misinformation about COVID-19 on social media; people in PNG already appear to be concerned about catching COVID-19 (so, there is a need to explicitly highlight that receiving the COVID-19 vaccine can prevent severe illness and death); and people's prior history of vaccination does not appear to be linked to their willingness to receive a COVID-19 vaccine (so, policymakers should develop a unique COVID-19 vaccination communication strategy, as opposed to rolling communications into existing efforts to minimise hesitancy about vaccines in general).

In conclusion, these findings suggest communication campaigns should recognise the distinct nature of the hesitancy people have around COVID-19 vaccines and convey that getting vaccinated is the single best way to prevent severe illness or death from COVID-19.