Behavioural change in post-COVID-19 vaccination
Since late 2021, mass COVID-19 vaccination has increased in many countries to rapidly increase herd immunity. Mixed and combined COVID-19 vaccine regimens, for instance, two doses of whole-pathogen inactivated vaccine (CoronaVac or BBIBP-CorV) or viral vector vaccine (AZD1222) followed by booster doses of mRNA-based vaccines (BNT162b2 or mRNA-1273), are generally used in Thailand. Uncertainty regarding vaccine novelty, unapproved efficacy of mixed regimens, and concerns regarding side effects may affect protective behaviours among vaccinated Thai people. Inconsistent with similar evidence globally, almost all vaccinated participants did not relax their protective behaviours after vaccination, while small groups of people exhibited an improved trend in exercising protective behaviours, especially avoiding social activity. The overall direction of the results indicated that non-HCWs, those living in urban areas, and people aged between 18 and 24 years were significant independent predictors of improvements in handwashing, physical distancing, and avoiding social activities. Interestingly, the variety of COVID-19 vaccine regimens was not significantly related to improvements in protective behaviours after vaccination.
Our results support the evidence from a previous study in China(13) conducted in March 2021, which revealed that the rate of mask-wearing did not reduce significantly after vaccination. In addition, another study from China showed that the scores of protective behaviours in the post-vaccination group were statistically higher than those in the pre-vaccination group (14). On the other hand, our results were inconsistent with the results of the study conducted in Israel(5) between 25 March and 7 April 2021, which reported decreased physical distancing and mask-wearing in specific populations following vaccination. Additionally, a study in Bangladesh(7) between 28 July to 13 August 2021 reported that the inclination to avoid distance, handshakes, use sanitiser and mask, visit crowded places, travel, and stay outside for longer periods had increased among vaccinated individuals. Moreover, a previous study led by Wright et al. reported that there was no clear evidence that compliance with social distancing had reduced in vaccinated UK adults relative to those who were not yet vaccinated(15). These results could be explained by public health policies in each country. In Thailand, the government policies regarding wearing masks, public gatherings, delaying the reopening, or limiting the opening time of entertainment places were strict(16). Likewise, in China, government policies regarding wearing masks in public places aimed for ‘zero-COVID’ and were very strictly implemented(17). Meanwhile, in other countries, especially Israel and the UK, the restrictions on public gatherings and wearing masks outdoors were lifted(18). Moreover, at the time of data collection in these studies, there was evidence regarding the COVID-19 vaccine’s efficacy to protect people from SARS-CoV-2 infection. But now, as new variants arise due to mutations of SARS-CoV-2, current COVID-19 vaccine regimens might not be able to cover all strains of SARS-CoV-2 infection. Moreover, the Thai population is eagerly concerned about this pandemic and thus is still strictly performing personal protective behaviours.
According to our analysis, non-HCWs exhibited significant improvements in handwashing and avoiding social activities after vaccination when compared with HCWs. In handwashing, improvements in non-HCWs were more than 4.31 folds compared with HCWs, which could be explained by public perception. Fujii et al.(19) suggested that higher perceived effectiveness might be a common factor to encourage protective behaviours in response to the COVID-19 pandemic. HCWs usually washed their hands as a routine hand hygiene practice, but the daily routine of handwashing in non-HCWs was less common. Moreover, avoidance of social activities in non-HCWs was 2.45 folds compared with HCWs. A possible explanation can be that during the COVID-19 situation, HCWs were on the frontline against COVID-19 and were unable to avoid social activities (i.e. meetings, teamwork). While the Thai government established a movement control order which encouraged people (non-HCWs) to work from home. Hence, it appeared that non-HCWs could avoid social activity more easily. Additionally, Thai vaccinated people received COVID-19 preventive recommendations during the observed adverse event after vaccination, which might have improved knowledge and protective behaviours in non-HCWs. The relevance of this assumption is in line with a previous study, where Fujii R(19) reported that people who changed their behaviours because of recommendations from doctors/public health officials were more likely to engage in handwashing/using hand sanitisers in China, Italy, and Korea. The same trend was observed in avoiding social gatherings in the USA. Importantly, this study revealed that after vaccination, approximately 85-90% of HCWs did not relax their protective behaviours, including physical distancing (95.8%), wearing masks (92.3%), and handwashing (88.1%), except avoiding social activity (82.5%). These results are inconsistent with those of a study by Rahamim-Cohen et al.(5) in Israel, where HCWs exhibited a minimal decrease in mask-wearing (4.3%) but a more widespread decrease in social distancing (43.5%).
Aside from that, improvements in avoiding social activities were 3.6 folds higher among people living in urban areas compared to those in rural areas. This result is inconsistent with a previous study, where Fujii R(19) reported that those who lived in urban areas were more likely to avoid social gatherings in Korea, but an inverse association was found in Italy and the UK. In the context of our study, a possible explanation might be that the enforcement of Thai government policies (i.e. lockdown, night-time curfew) in urban was stricter than in rural areas. Opening and closing services had a time limit(16). As a reason, it may be possible that participants who live in urban areas avoided social activity more than those living in rural areas.
Interestingly, our findings show that avoiding social activities was 3.93-fold higher among early adults than middle adults. It might be because early adults were at a studying age when the government enforced a strategy of closing onsite educational institutions across the country and replacing them with online classrooms. Thus, early adults could practice physical distancing more than the working-age population, who might have to go to the office. This result was inconsistent with the Rahamim-Cohen D study, which reported that people under the age of 50 were more likely to decrease mask-wearing and social distancing (28.1% and 56.1%, respectively) as compared with people over the age of 50 (17.2% and 41.8%, respectively)(5). The results can be seen against the backdrop that higher age was a known risk factor for a more severe COVID-19-associated illness and death than middle and young age(20-22). Thus, the higher age group could be more likely to engage in protective behaviour even after vaccination than the lower age group.
Strengths and limitations
This study is one of the few studies that assess the levels of protective behaviours in the light of various COVID-19 vaccine regimens which were available in Thailand. It is plausible that a few limitations may have influenced the results obtained. Firstly, due to the situation of COVID-19, we collected data using self-reported online questionnaires. The online questionnaire may have been unable to reach those without the Internet. Moreover, there are issues of reliability of responses in self-reported questionnaires. Secondly, the questions about protective behaviours sought a response on a 5-point Likert scale. It means that one level change in behavioural scale might not be significant for interpretation. Additionally, the term frequency among each individual might be different. There could be information bias, and observational study or behavioural monitoring tools are a better way to reduce bias. Thirdly, the questions on vaccine regimens were self-reported, and the possibility of invalid answers or recall bias should be considered. Lastly, in long-period data collection, variations in available types of COVID-19 vaccine, morbidity, and mortality rate by situation and time can affect periodic protective behaviour.
Implications and further studies
Our study findings shed some light on the levels of protective behaviours pre-and post-COVID-19 vaccination and predictive factors associated with improved protective behaviours. Further studies should obtain a large study sample, distribute online questionnaires more broadly via a cooperative institution, and advertise more in diverse work fields. The data collected on vaccine regimens from vaccine certificates or passports may help avoid invalid answers or recall bias. This study was conducted over a long period with uncertain fluctuating circumstances in Thailand; thus, data collection during a brief period with a greater quantity of responses would minimize information bias to effect behavioural change.