Skip to main content
NCBI home page
Bulletin of the World Health Organization logoLink to Bulletin of the World Health Organization
. 2022 Dec 1;101(2):111–120. doi: 10.2471/BLT.22.288655

Modelling COVID-19 vaccination status and adherence to public health and social measures, Eastern Mediterranean Region and Algeria

Modélisation du statut de vaccination contre la COVID-19 et de l'adhésion aux mesures sociales et de santé publique, Région de la Méditerranée orientale et Algérie

Modelización del estado de vacunación contra la COVID-19 y del cumplimiento de las medidas sociales y de salud pública, Región del Mediterráneo Oriental y Argelia

وضع نماذج لحالة التطعيم ضد كوفيد 19 والالتزام بالصحة العامة والتدابير الاجتماعية، منطقة شرق البحر المتوسط والجزائر

东地中海地区和阿尔及利亚 COVID-19 疫苗接种状况和公共卫生和社会措施合规性建模

Составление модели статуса вакцинации против COVID-19 и соблюдения мер общественного здравоохранения и социальных мер в странах Восточного Средиземноморья и Алжире

Zlatko Nikoloski a,✉, Robert Bain b, Manal K Elzalabany c, Peggy Hanna c, Tara Rose Aynsley c, Dalia Samhouri c, Leonardo Menchini b, Neha Kapil b, Amaya Gillespie b
PMCID: PMC9874377  PMID: 36733625

Abstract

Objective

To study the link between coronavirus disease 2019 (COVID-19) vaccination status and adherence to public health and social measures in Members of the Eastern Mediterranean Region and Algeria.

Methods

We analysed two rounds of a large, cross-country, repeated cross-sectional mobile phone survey in June–July 2021 and October–November 2021. The rounds included 14 287 and 14 131 respondents, respectively, from 23 countries and territories. Questions covered knowledge, attitudes and practices around COVID-19, and demographic, employment, health and vaccination status. We used logit modelling to analyse the link between self-reported vaccination status and individuals’ practice of mask wearing, physical distancing and handwashing. We used propensity score matching as a robustness check.

Findings

Overall, vaccinated respondents (8766 respondents in round 2) were significantly more likely to adhere to preventive measures than those who were unvaccinated (5297 respondents in round 2). Odds ratios were 1.5 (95% confidence interval, CI: 1.3–1.8) for mask wearing; 1.5 (95% CI: 1.3–1.7) for physical distancing; and 1.2 (95% CI: 1.0–1.4) for handwashing. Similar results were found on analysing subsamples of low- and middle-income countries. However, in high-income countries, where vaccination coverage is high, there was no significant link between vaccination and preventive practices. The association between vaccination status and adherence to public health advice was sustained over time, even though self-reported vaccination coverage tripled over 5 months (19.4% to 62.3%; weighted percentages).

Conclusion

Individuals vaccinated against COVID-19 maintained their adherence to preventive health measures. Nevertheless, reinforcement of public health messages is important for the public’s continued compliance with preventive measures.

Introduction

The introduction of vaccines for coronavirus disease 2019 (COVID-19) added another measure to the existing set of recommended preventive measures (wearing a mask in public, keeping a distance from other people and regular handwashing). The roll-out of the vaccines, however, raised concerns that vaccination may lead to lower adherence to the existing preventive measures. The advice from the World Health Organization (WHO) was to continue these public health and social measures after being vaccinated.1 However, evidence from other epidemics suggests that there is lower adherence to preventive measures when some level of protection exists (for example, individuals who use human immunodeficiency virus pre-exposure prophylaxis).2 This effect is further compounded by people losing motivation to follow recommended protective measures (so-called pandemic fatigue).3 With most countries relaxing the stringent restrictions imposed at the start of the pandemic, understanding the link between vaccination status and adherence to public health advice is important.

To date, a few studies from high-income countries have tried to understand the link between vaccination status and COVID-19 preventive behaviours. Relying on a longitudinal survey of vaccinated and unvaccinated individuals, a study from the United Kingdom of Great Britain and Northern Ireland4 found no evidence that vaccinated individuals decreased compliance relative to those who were not yet vaccinated. These findings were echoed in a study relying on a cross-sectional survey in 12 high-income countries.5

The Eastern Mediterranean Region (as defined by WHO) and the Middle East and North Africa Region (as defined by the United Nations Children’s Fund, UNICEF) together comprise an overlapping group of 23 countries and territories. Among these countries, only Algeria is not a Member of the WHO Eastern Mediterranean Regional Committee, but the WHO African Regional Committee. While vaccination uptake and barriers to uptake in these countries received much attention in 2021,6 the link between vaccination status and adherence to public health advice on COVID-19 prevention has not been sufficiently studied. Only one study, in Somalia, found a positive correlation between adherence to preventive behaviours and willingness to get vaccinated.7 However, there is little insight into what happens to COVID-19 preventive behaviours as vaccination rates across the region increase. We have identified 11 studies documenting the practice of the most common types of COVID-19 preventive behaviours in Egypt,8,9 the Islamic Republic of Iran,10,11 West Bank and Gaza Strip,12 Saudi Arabia13–16 and Somalia.7,17 The studies covered the general population,7,9–13,15,17 or medical professionals and medical students.8,14,16 There are a few common characteristics across these studies: (i) they were small-scale, cross-sectional surveys; (ii) except for the two studies in Islamic Republic of Iran, the available studies in the region relied on online data collection;18 and (iii) all of the studies pre-dated the main vaccination campaigns in the region.

Against this background, our objective was to study the relationship between COVID-19 vaccination status and adherence to public health and social measures throughout the Eastern Mediterranean and Middle East and North Africa. To account for the differences in vaccine availability across different countries, we also conducted a subregional analysis of countries by income group.

Methods

Setting

The analysis in this paper is based on a repeated cross-sectional survey of knowledge, attitudes and practices around COVID-19 among individuals in 23 countries and territories (Table 1). The survey was conducted by UNICEF (Middle East and North Africa Region) and WHO (Eastern Mediterranean Region).

Table 1. Participants recruited to the survey of COVID-19 vaccination status and adherence to preventive measures, Eastern Mediterranean Region and Algeria, June to November 2021.

Country or territory, by income group  No. of participants
Round 1: Jun–Jul 2021 Round 2: Oct–Nov 2021
High income 
Bahrain 354 350
Kuwait 501 511
Oman 500 503
Qatar 350 352
Saudi Arabia 761 755
United Arab Emirates 502 500
Middle income
Algeria 706 700
Djibouti 350 350
Egypt 1001 1059
Iran (Islamic Republic of) 1002 1030
Iraq 793 716
Jordan 552 520
Lebanon 500 504
Libya 520 511
Morocco 715 772
Pakistan 1026 1016
Tunisia 575 628
West Bank and Gaza Strip 359 350
Low income  
Afghanistan 775 713
Somalia 501 507
Sudan 769 761
Syrian Arab Republic 634 523
Yemen 541 500
Total 14 287 14 131
Open in a new tab

COVID-19: coronavirus disease 2019.

Note: Country groups are those of the 2021 World Bank income classification.19

Source: United Nations Children’s Fund (Middle East and North Africa Region) and World Health Organization (Eastern Mediterranean Region) survey of COVID-19 knowledge, attitudes and practices.

Data collection

We conducted the survey in two waves, first in June and July 2021 and then in October and November 2021. We based the survey on computer-assisted telephone interviews, using random digital dialling to sample working mobile numbers in each country. We hired a service provider company (GeoPoll, Denver, United States of America) to conduct the survey.

For the data collection we designed a structured questionnaire consisting of 31 standardized questions related to: (i) individual characteristics, such as demographic, employment, health and vaccination status; (ii) behavioural barriers, perceptions and beliefs about vaccines and COVID-19; and (iii) community factors, such as social norms and impact on households and health service utilization. We derived the questions from the global question bank provided by the Risk Communication and Community Engagement Collective Service, a collaborative partnership of key stakeholders from the public health and humanitarian sectors. The choice of questions was guided by a conceptual model of vaccine uptake.20 The model is well-recognized and embraces an ecological approach by including individual and community influences as well as wider policy and environmental factors and influences on health decisions (the model is shown in the online repository).21 The questionnaire was translated into national languages. We piloted the survey in all countries before starting data collection.

For the full data collection, trained enumerators used random digit dialling to generate a random sample of mobile phone respondents aged 18 years and older from the 22 countries and one territory. The analysis comprised separate samples of 14 287 individuals in the first round and 14 131 individuals in the second round. Sample sizes for each country were based on population size and mobile phone coverage. Participants’ details were anonymized after each round of data collection; it was therefore not possible to determine the overlap of participants in the first and the second round of the survey. However, even in the least populous countries the chances of selecting the same number with random digit dialling is very small. For example, Djibouti has a population of 1.1 million and we selected around 400 participants. Based on the proportion of adults in the population (63%) and the mobile phone coverage (43%), there are still around 300 000 eligible respondents in the country. We weighted the sample at regional level by gender and age, based on the United Nations (UN) demographics for the countries and territory.22 A detailed description of the derivation of the weights is presented in the online repository.21

Statistical analysis

The variables concerning adherence to public health and social measures were based on responses to the questions about the frequency of practising the following measures over the previous week: (i) wearing a mask in public; (ii) keeping a physical distance of at least 2 m from people in public; and (iii) washing hands with soap and water for 20 seconds. Respondents gave their responses on a 5-point Likert scale (1: all of the time; 5: never). We defined binary outcome variables for each of the three public health and social measures, taking a value of 1 if the respondent practised the measure all of the time or most of the time and 0 for sometimes, rarely or never. The variable concerning vaccination status was based on participants’ responses to the question about whether they had received at least one dose of COVID-19 vaccine. From the responses we created a binary variable with values of 1 if the respondent had been vaccinated and 0 if not vaccinated.

To study the correlates of each of the public health and social measures we developed three separate models. The binary variables for practice of public health and social measures were then used as dependent variables in a logit modelling analysis. In addition to the variable capturing vaccination status, the model included the following correlates: socioeconomic and demographic variables (such as age, gender, occupation); self-reported previous infection with severe acute respiratory syndrome coronavirus 2 (SARS-CoV-2); and knowledge about asymptomatic SARS-CoV-2 transmission (including attitudes towards being at risk). By asking about respondents’ risk beliefs, we were able to capture infection risk as well as self-reported vaccination status.

To account for country heterogeneity (for example, differences in the health-care systems), we included country dummy variables in the regression analysis. All regressions used the derived weights as described above (further details are in the online repository).21 We conducted the analysis on the first round of collected data and then repeated it on the second round of data. Given that our sample included respondents living in countries at different levels of economic development, we also analysed the countries grouped by the World Bank categories: high, middle and low income.19

We performed three additional analyses. First, we fitted our model onto a pooled data set comprising the two rounds of the survey, while also accounting for the time effect as well as an interaction variable between the time effect and the vaccination status variable. With this method we aimed to test if the link between vaccination status and adherence to preventive measures changed over time. Second, we conducted a propensity score matching analysis, which is an established technique to reduce selection bias in observational data, by matching treatment and control individuals (that is, vaccinated and non-vaccinated individuals) based on their observable characteristics.23,24 In doing so, we compared the practice of preventive measures of those respondents who were vaccinated (treated) with those who were not vaccinated (control group). In particular, we used a nearest neighbour matching estimator. A treatment group observation (vaccinated individual) thus matched with an observation from the control group (unvaccinated individual) that had the closest propensity score25,26 (further details are in the online repository).21 Finally, we conducted a multinomial logit analysis using untransformed public health and social measures variables (as categorical variables on a 5-point Likert scale).

We performed all analyses using Stata version 14.0 (StataCorp., College Station, USA); P-values less than 0.05 were considered significant throughout.

Ethical approval

The survey tool and protocols were approved by the WHO Regional Ethical Research Committee.

Results

Background characteristics

The samples in round 1 and round 2 were similar in terms of the distribution of age and gender (Table 2). Understandably, the numbers of respondents reporting that they had ever been infected with COVID-19 increased from 1950 in the first round to 2637 in the second round (weighted percentage of total respondents 13.9% and 19.0%, respectively). The number of respondents reporting that they were vaccinated increased from 2772 to 8766 between rounds (19.4% to 62.3%; weighted percentages). Finally, while the practice of physical distancing and handwashing was similar between the two rounds of the survey, the number of respondents reporting wearing a mask in public decreased from 9370 to 8674 (66.0% to 61.6%; weighted percentages). Similar findings emerged when the descriptive statistics were analysed by country income groups (available in the online repository).21 More specifically, the substantial increase in self-reported vaccination status was accompanied by only a small decrease in adherence to public health and social measures.

Table 2. Characteristics of respondents in the survey of COVID-19 vaccination status and adherence to preventive measures, Eastern Mediterranean Region and Algeria, June to November 2021.

Variable No. of respondents (weighted %)
Round 1: Jun–Jul 2021 (n = 14 287) Round 2: Oct–Nov 2021 (n = 14 131)
Gender and age
Female
    Age 18–24 years 1 432 (10.0) 1 416 (10.0)
    Age 25–34 years 1 810 (12.7) 1 790 (12.7)
    Age 35–49 years 1 986 (13.9) 1 964 (13.9)
    Age 50+ years 1 678 (11.7) 1 659 (11.7)
Male
    Age 18–24 years 1 521 (10.6) 1 505 (10.6)
    Age 25–34 years 1 982 (13.9) 1 960 (13.9)
    Age 35–49 years 2 172 (15.2) 2 149 (15.2)
    Age 50+ years 1 707 (11.9) 1 688 (11.9)
Occupation
Working in education sector 599 (4.2) 780 (5.5)
Working in health-care sector 357 (2.5) 623 (4.4)
Homemaker 2 437 (17.2) 3 050 (21.7)
Not currently in paid work 2 689 (19.0) 2 619 (18.6)
Working in other essential servicesa 6 684 (47.3) 5 602 (39.8)
Student 1 368 (9.7) 1 382 (9.8)
Do you have a chronic illness?
Yes 2 276 (16.0) 2 432 (17.3)
No 11 945 (84.0) 11 655 (82.7)
To your knowledge, are you or have you been infected with COVID-19? 
Yes 1 950 (13.9) 2 637 (19.0)
No 12 042 (86.1) 11 270 (81.0)
Have you received COVID-19 vaccination?
Yes 2 772 (19.4) 8 766 (62.3)
No 11 500 (80.6) 5 297 (37.7)
To what extent do you trust your local health-care providers to provide accurate information on COVID-19 vaccination and prevention?
Extremely 1 430 (15.0) 2 108 (19.4)
Very much 2 510 (26.2) 3 409 (31.4)
Moderately 2 837 (29.7) 2 931 (27.0)
Slightly 1 428 (14.9) 1 550 (14.3)
Not at all 940 (9.8) 851 (7.8)
Do you believe coronavirus can be transmitted by coming in direct contact with a person who has the virus but has no symptoms?
Yes 7 242 (75.7) 7 349 (70.7)
No 1 654 (17.3) 3 051 (29.3)
How likely do you believe that you will get infected with COVID-19?
Very likely 1 029 (10.8) 915 (8.9)
Likely 2 260 (23.6) 2 181 (21.3)
Neutral 1 647 (17.2) 1 839 (18.0)
Unlikely 1 344 (14.1) 1 802 (17.6)
Very unlikely 2 432 (25.4) 3 504 (34.2)
Over the past week how often have you:
Worn a mask in public 9 370 (66.0) 8 674 (61.6)
Kept at least 2 m away from people in public 7 286 (51.7) 7 046 (50.2)
Washed your hands with water and soap for 20 seconds 11 116 (78.3) 10 729 (76.3)
Open in a new tab

COVID-19: coronavirus disease 2019.

a Other services deemed essential during the pandemic, such as food sector, logistic.

Note: n is the total sample size in each round. We calculated percentages by applying the appropriate regional weights; data were missing in some categories. Inconsistencies arise in some values due to rounding. All variables were self-reported.

Source: United Nations Children’s Fund (Middle East and North Africa Region) and World Health Organization (Eastern Mediterranean Region) survey of COVID-19 knowledge, attitudes and practices.

Logit model analysis

Fig. 1 depicts the findings of the logit model analysis based on data of round 2 of the survey. Demographic status played a significant role in mask wearing, with young men being less likely relative to young women to wear a mask. More specifically, men aged 18–24 years were 0.5 times less likely (95% confidence interval, CI: 0.4–0.7), and men aged 25–34 years were 0.5 times less likely (95% CI: 0.4–0.7) to wear a mask in public compared with women aged 18–24 years. In addition, those reporting no previous COVID-19 infection were 0.7 times less likely (95% CI: 0.6–0.9) to wear a mask in public. Respondents who did not know that the SARS-CoV-2 could be passed on asymptomatically were 0.6 times less likely (95% CI: 0.5–0.8) to report wearing a mask in public. More importantly, the results show a strong link between self-reported COVID-19 vaccination status and wearing a mask in public. Those who had been vaccinated were 1.5 times more likely (95% CI: 1.3–1.8) to report wearing a mask in public relative to those who were not vaccinated.

Fig. 1.

Likelihood of adhering to preventive measures, Eastern Mediterranean Region and Algeria, October to November 2021

CI: confidence interval; COVID-19: coronavirus disease 2019; OR: odds ratio; Ref: reference group.

Note: All variables were self-reported.

Source: Round 2 (October and November 2021; n = 14 131 participants) of the United Nations Children’s Fund (Middle East and North Africa Region) and World Health Organization (Eastern Mediterranean Region) survey of COVID-19 knowledge, attitudes and practices.

Fig. 1

Open in a new tab

When considering physical distancing, we found some evidence that those with lower trust in the local health-care provider were less likely to practise physical distancing (Fig. 1). However, there is a strong link between vaccination status and physical distancing, suggesting that those who were vaccinated were 1.5 times more likely (95% CI: 1.3–1.7) to practise physical distancing relative to unvaccinated respondents.

Finally, when using handwashing with water and soap for 20 seconds as a dependent variable, younger men, those with lower trust in the local health-care provider, as well as those who did not consider themselves to be at risk of contracting SARS-CoV-2, were less likely to adhere to frequent handwashing. We also found that COVID-19 vaccinated respondents were 1.2 times more likely (95% CI: 1.0–1.4) to wash their hands regularly than were unvaccinated respondents. These findings were similar when the analysis was repeated on data from the round 1 survey (available in online repository).21

Income group analysis

Given the income heterogeneity of our sample, we next conducted a subregional analysis for high-, middle- and low-income countries. Fig. 2 illustrates the odds ratios of the logit model for our main variable of interest (vaccination status), while the logit models also controlled for the same set of variables used in Fig. 1 above. There was no statistically significant link between COVID-19 vaccination status and adherence to public health and social measures in the high-income countries. However, in the middle-income countries and territory there was a positive link between vaccination and practice of preventive measures, except for handwashing in round 2. In addition, the odds ratios for adherence to some of the public health and social measures were comparable across rounds. For example, in both rounds, those vaccinated were 1.5 times more likely to physically distance than those who were unvaccinated. These findings were similar when the analysis was repeated on the subsample of low-income countries (Fig. 2).

Fig. 2.

Likelihood of adhering to preventive measures, by vaccination status and country income group, Eastern Mediterranean Region and Algeria, June to November 2021

CI: confidence interval; COVID-19: coronavirus disease 2019; OR: odds ratio; Ref: reference group.

Notes: The models above control for the following set of correlates: age, gender, occupation status, existence of comorbidities, previous COVID-19 infection, trust in local health-care provider, knowledge about asymptomatic transmission of the virus, risk attitude towards COVID-19. Country groups are those of the 2021 World Bank income classification (Table 1). All variables were self-reported.

Source: Round 1 (June and July 2021; n = 14 287 participants) and round 2 (October and November 2021; n = 14 131 participants) of the United Nations Children’s Fund (Middle East and North Africa Region) and World Health Organization (Eastern Mediterranean Region) survey of COVID-19 knowledge, attitudes and practices.

Fig. 2

Open in a new tab

Additional analyses

Detailed results of all additional analyses are available in the online repository.21 As a robustness check, we repeated the analysis by pooling the two rounds of the survey. In addition, the time effect was statistically significant across all three public health and social measures and with odds ratios lower than 1, indicating, on average, a reduction in the practice of the selected preventive measures over time. Finally, in the case of mask wearing and physical distancing, the interaction term between the time effect and self-reported vaccination status was insignificant, suggesting that the link between vaccination and preventive measures had not changed over time. These results are consistent when the analysis was repeated on the subregional level by country income group.

Furthermore, when we used propensity score matching in the analysis, the results confirmed our main findings. Finally, the results were also similar when we used a multinomial logit analysis on untransformed health practice variables in the analysis.

Discussion

We found a robust link between self-reported COVID-19 vaccination status and adherence to all three public health and social measures in the Eastern Mediterranean Region and Algeria. Individuals vaccinated against COVID-19 were more likely to adhere to preventive behaviours compared with unvaccinated respondents in low- and middle-income countries and territory. This finding supports the general health motivation construct in the health belief model,27 and aligns with social identity theory.28 The theory proposes that people who practise one health behaviour (such as vaccination) are more likely to practise others (in this case, recommended measures to prevent the transmission of SARS-CoV-2). Furthermore, it may also be the case that, after receiving the vaccine at their vaccination appointment, individuals are reminded to continue their practice of mask wearing, physical distancing and handwashing.29 However, we did not find a statistically significant link between vaccination status and adherence to preventive measures in the high-income countries. The countries in the Gulf have been able to achieve a more rapid roll-out of COVID-19 vaccination programmes than low- and middle-income countries, resulting in a substantial increase in vaccination coverage in a relatively short period. According to official data, by the end of 2021 about three quarters of the adult population in the Gulf countries had received at least one dose of a COVID-19 vaccine.30 This high coverage may explain the lack of a statistically significant link between vaccination status and adherence to public health and social measures. Similar results have been reported from other advanced economies across the world.4,5,31

Nevertheless, the results of the pooled analysis suggest that adherence to preventive measures decreased slightly over the 5 months between surveys, which is somewhat consistent with people losing motivation to follow recommended protective measures.3 However, our analysis also indicates that, except for handwashing, the link between vaccination and the practice of preventive measures did not change between the two survey rounds, despite the surge in vaccination coverage in the region. The results are robust when we used alternative methods of analysis (propensity score matching and multinomial logit). Finally, we also found evidence that demographic characteristics, trust in the local health-care provider, as well as risk perception of COVID-19, were significant correlates of individuals’ adherence to preventive measures. Previously, it has been shown that declining prevalence and severity of COVID-19 was associated with lower adherence to preventive measures in Australia, the United Kingdom and the USA,32 as well as Somalia.7

We also found consistent evidence that demographic factors, trust in local health-care providers, as well as perception of COVID-19 risk, are significant correlates of adherence to preventive measures. Studies in the same region have shown that women and older individuals were more adherent to COVID-19 preventive measures.9,11,15,17 In addition, existing studies from the region indicated that lower trust in health-care providers was associated with lower adherence to preventive behaviours.13 Finally, individuals who believed they were at lower risk of contracting COVID-19 tended to show lower adherence to public health and social measures, consistent with existing evidence.33 It is worth pointing out that the risk of infection was different in the two rounds of the survey, in that lockdowns and other policies had reduced the risk of infection over time.

Our study has several limitations. The first limitation is in the study’s representativeness. People who did not have mobile phones or chose not to participate were not included in the study (although in 2020, the average mobile phone penetration in the region was 98 mobile phone subscriptions per 100 people).34 In addition, and given the cultural traditions of the region, more men than women tended to be included, using the sampling method applied here. To mitigate this bias, we used UN standard population demographic data to weight the raw data by age and gender to adjust for such differences. Second, the analysis in our study was based on self-reported data, which is vulnerable to various types of bias. We explored this bias through comparison with other data sources, such as vaccination status. Third, the survey was a repeated cross-sectional survey, so we caution against direct causal inferences, particularly as we gathered data on vaccination status and practice of preventive measures at the same time. Fourth, we only conducted regional and subregional analyses, as the sample was not stratified by subnational or administrative level, and country samples were too small to support individual analysis. Finally, public opinion about the risks of COVID-19 is dynamic. These data were collected during June and July and then October and November 2021. Interpretation of the data should consider factors affecting countries around that time, such as government policies and enforcement of restrictions, seasonal activities, traditions or conventions related to education calendars, cultural and religious events, and the media.

Overall, in the regions we studied, we found no evidence that the roll-out of vaccination programmes resulted in COVID-19 risk compensation, whereby individuals adjust their behaviour based on a lower perceived level of risk. Nevertheless, reinforcement of public health messages on prevention is still important for individuals’ compliance and, in an era when government-mandated restrictions are being lifted, adherence to public health and social measures are expected to drop.

Acknowledgements

The study was commissioned by UNICEF Regional Office for the Middle East and North Africa Region and WHO Regional Office for the Eastern Mediterranean. The data collection was conducted by GeoPoll. UNICEF’s contribution to this survey was made possible by support from UNICEF’s Access to COVID-19 Tools Accelerator (ACT-A) Appeal.

Competing interests:

None declared.

References

  • 1.COVID-19 advice for the public: getting vaccinated. Geneva: World Health Organization; 2022. Available from: https://www.who.int/emergencies/diseases/novel-coronavirus-2019/covid-19-vaccines/advice [cited 2022 Mar 22].
  • 2.Traeger MW, Schroeder SE, Wright EJ, Hellard ME, Cornelisse VJ, Doyle JS, et al. Effects of pre-exposure prophylaxis for the prevention of human immunodeficiency virus infection on sexual risk behavior in men who have sex with men: a systematic review and meta-analysis. Clin Infect Dis. 2018. Aug 16;67(5):676–86. 10.1093/cid/ciy182 [DOI] [PubMed] [Google Scholar]
  • 3.Petherick A, Goldszmidt R, Andrade EB, Furst R, Hale T, Pott A, et al. A worldwide assessment of changes in adherence to COVID-19 protective behaviours and hypothesized pandemic fatigue. Nat Hum Behav. 2021. Sep;5(9):1145–60. 10.1038/s41562-021-01181-x [DOI] [PubMed] [Google Scholar]
  • 4.Wright L, Steptoe A, Mak HW, Fancourt D. Do people reduce compliance with COVID-19 guidelines following vaccination? A longitudinal analysis of matched UK adults. J Epidemiol Community Health. 2022. Feb;76(2):109–15. 10.1136/jech-2021-217179 [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 5.Goldszmidt R, Petherick A, Andrade EB, Hale T, Furst R, Phillips T, et al. Protective behaviors against COVID-19 by individual vaccination status in 12 countries during the pandemic. JAMA Netw Open. 2021. Oct 1;4(10):e2131137. 10.1001/jamanetworkopen.2021.31137 [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 6.Qunaibi EA, Helmy M, Basheti I, Sultan I. A high rate of COVID-19 vaccine hesitancy in a large-scale survey on Arabs. eLife. 2021. May 27;10:e68038. 10.7554/eLife.68038 [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 7.Ahmed MAM, Colebunders R, Gele AA, Farah AA, Osman S, Guled IA, et al. COVID-19 vaccine acceptability and adherence to preventive measures in Somalia: results of an online survey. Vaccines (Basel). 2021. May 21;9(6):543. 10.3390/vaccines9060543 [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 8.Ahmed HM. Adherence to COVID-19 preventive measures among male medical students, Egypt. J Egypt Public Health Assoc. 2022. Feb 17;97(1):8. 10.1186/s42506-022-00103-7 [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 9.Abd Elhameed Ali R, Ahmed Ghaleb A, Abokresha SA. COVID-19 related knowledge and practice and barriers that hinder adherence to preventive measures among the Egyptian community. An epidemiological study in Upper Egypt. J Public Health Res. 2021. Mar 5;10(1):1943. 10.4081/jphr.2021.1943 [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 10.Abdollahzadeh H, Rahmanian V, Sharifi N, Zahedi R, Bajestani MJN, GholianAval M, et al. Risks assessment of Adherence to non-pharmaceutical measures towards COVID-19 among residents of Mashhad in the North-East of Iran during the awful wave of the epidemic. J Family Med Prim Care. 2021. Sep;10(9):3211–8. 10.4103/jfmpc.jfmpc_130_21 [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 11.Shati M, Alami A, Mortazavi SS, Eybpoosh S, Emamian MH, Moghadam M. Adherence to self-isolation measures by older adults during coronavirus disease 2019 (COVID-19) epidemic: a phone survey in Iran. Med J Islam Repub Iran. 2020. Nov 10;34:152. 10.47176/mjiri.34.152 [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 12.Al Zabadi H, Yaseen N, Alhroub T, Haj-Yahya M. Assessment of quarantine understanding and adherence to lockdown measures during the COVID-19 pandemic in Palestine: community experience and evidence for action. Front Public Health. 2021. Mar 2;9:570242. 10.3389/fpubh.2021.570242 [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 13.Al-Shammary AA, Hassan SU, Zahra A, Algahtani FBZ, Suleiman S. Role of community-based measures in adherence to self-protective behaviors during first wave of COVID-19 pandemic in Saudi Arabia. Health Promot Perspect. 2021. Feb 7;11(1):69–79. 10.34172/hpp.2021.10 [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 14.Albeladi FI, Alluli MM, Daghriri KA, Almalki YH, Wafi MY, Otaif FA, et al. Level of adherence to COVID-19 preventive measures among health care workers in Saudi Arabia. Cureus. 2021. Jun 27;13(6):e15969. 10.7759/cureus.15969 [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 15.Alghamdi AN, Alotaibi MI, Alqahtani ASA, Al Aboud D, Abdel-Moneim AS. Association between educational status and awareness of adherence to preventive measures for COVID-19 in Saudi Arabia. Asia Pac J Public Health. 2021. Jul;33(5):623–6. 10.1177/10105395211014637 [DOI] [PubMed] [Google Scholar]
  • 16.Almalki MJ. A cross-sectional study of the satisfaction with, adherence to, and perspectives toward COVID-19 preventive measures among public health students in Jazan, Saudi Arabia. Int J Environ Res Public Health. 2022. Jan 12;19(2):802. 10.3390/ijerph19020802 [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 17.Ahmed MAM, Siewe Fodjo JN, Gele AA, Farah AA, Osman S, Guled IA, et al. COVID-19 in Somalia: adherence to preventive measures and evolution of the disease burden. Pathogens. 2020. Sep 6;9(9):735. 10.3390/pathogens9090735 [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 18.Greenacre Z. The importance of selection bias in internet surveys. Open J Stat. 2016;6(03):397–404. 10.4236/ojs.2016.63035 [DOI] [Google Scholar]
  • 19.New World Bank country classifications by income level: 2021–2022 [internet]. Washington, DC: World Bank; 2021. Available from: https://blogs.worldbank.org/opendata/new-world-bank-country-classifications-income-level-2021-2022 [cited 2022 Mar 22].
  • 20.Quinn SC, Jamison AM, An J, Hancock GR, Freimuth VS. Measuring vaccine hesitancy, confidence, trust and flu vaccine uptake: results of a national survey of White and African American adults. Vaccine. 2019. Feb 21;37(9):1168–73. 10.1016/j.vaccine.2019.01.033 [DOI] [PubMed] [Google Scholar]
  • 21.Nikoloski Z, Bain R. Elzalabany MK, Hanna P, Aynsley TR, Samhouri D, et al. COVID-19 vaccination status and adherence to public health and social measures: a modelling study from the Eastern Mediterranean and Middle East and North Africa region. Supplementary materials [online repository]. London: Figshare; 2022. 10.6084/m9.figshare.21345258.v1 10.6084/m9.figshare.21345258.v1 [DOI]
  • 22.The United Nations world population prospects. New York: United Nations; 2019. Available from: https://population.un.org/wpp/ [cited 2022 Mar 22].
  • 23.Dehejia HR, Wahba S. Propensity score matching methods for non-experimental causal studies. Rev Econ Stat. 2002;84(1):151–61. 10.1162/003465302317331982 [DOI] [Google Scholar]
  • 24.Rosenbaum PR, Rubin DB. The central role of the propensity score observational studies for causal effects. Biometrika. 1983;70(1):41–55. 10.1093/biomet/70.1.41 [DOI] [Google Scholar]
  • 25.Caliendo M, Kopeinig S. Some practical guidance for the implementation of propensity score matching. J Econ Surv. 2008;22(1):31–72. 10.1111/j.1467-6419.2007.00527.x [DOI] [Google Scholar]
  • 26.Garrido MM, Kelley AS, Paris J, Roza K, Meier DE, Morrison RS, et al. Methods for constructing and assessing propensity scores. Health Serv Res. 2014. Oct;49(5):1701–20. 10.1111/1475-6773.12182 [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 27.Lam CN, Kaplan C, Saluja S. Relationship between mask wearing, testing, and vaccine willingness among Los Angeles County adults during the peak of the COVID-19 pandemic. Transl Behav Med. 2022. Mar 17;12(3):480–5. 10.1093/tbm/ibab150 [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 28.Hogg MA, Terry DJ, White KM. A tale of two theories: a critical comparison of identity theory with social identity theory. Soc Psychol Q. 1995;58(4):255–69. 10.2307/2787127 [DOI] [Google Scholar]
  • 29.Goldfarb JL, Kreps S, Brownstein JS, Kriner DL. Beyond the first dose – Covid-19 vaccine follow-through and continued protective measures. N Engl J Med. 2021. Jul 8;385(2):101–3. 10.1056/NEJMp2104527 [DOI] [PubMed] [Google Scholar]
  • 30.Coronavirus (COVID-19) vaccinations [internet]. Oxford: Our World in Data; 2022. Available from: https://ourworldindata.org/covid-vaccinations [cited 2020 Mar 28].
  • 31.Yamamura E, Koska Y, Tsutsui Y, Ohtake F. Effect of the COVID-19 vaccine on preventive behaviors: evidence from Japan. arXiv:2203.07660 [preprint]. 2022. 10.48550/arXiv.2203.07660 10.48550/arXiv.2203.07660 [DOI]
  • 32.MacIntyre CR, Nguyen PY, Chughtai AA, Trent M, Gerber B, Steinhofel K, et al. Mask use, risk-mitigation behaviours and pandemic fatigue during the COVID-19 pandemic in five cities in Australia, the UK and USA: a cross-sectional survey. Int J Infect Dis. 2021. May;106:199–207. 10.1016/j.ijid.2021.03.056 [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 33.Harper CA, Satchell LP, Fido D, Latzman RD. Functional fear predicts public health compliance in the COVID-19 pandemic. Int J Ment Health Addict. 2021;19(5):1875–88. 10.1007/s11469-020-00281-5 [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 34.World development indicators [internet]. Washington, DC: World Bank; 2022. Available from: https://databank.worldbank.org/source/world-development-indicators [cited 2022 Sep 15].

Articles from Bulletin of the World Health Organization are provided here courtesy of World Health Organization

RESOURCES