Who Wears the Face Mask? Preventive Measures Against COVID-19 in Latin America Before Vaccination

Who Wears the Face Mask? Preventive Measures Against COVID-19 in Latin

America Before Vaccination

Rentería, Elisenda

Gómez-Casillas, Amalia

Zueras, Pilar

Centre d’Estudis Demogràfics; ORCID 0000-0002-7072-8465; [email protected]

Centre d’Estudis Demogràfics; ORCID 0000-0002-3528-7844; [email protected]

University of Essex and Centre d’Estudis Demogràfics, ORCID 0000-0002-3345-

8865; [email protected].es

* Corresponding author

. CC-BY-NC-ND 4.0 International licenseIt is made available under a

perpetuity.

is the author/funder, who has granted medRxiv a license to display the preprint in(which was not certified by peer review)preprint

The copyright holder for thisthis version posted February 8, 2023. ; https://doi.org/10.1101/2023.02.07.23285570doi: medRxiv preprint

NOTE: This preprint reports new research that has not been certified by peer review and should not be used to guide clinical practice.

Who Wears the Face Mask?

Preventive Measures Against COVID-19 in Latin America Before Vaccination

Abstract

The COVID-19 pandemic outbreak imposed the use of the sanitary mask as a protective

measure to reduce the spread of the pandemic, recommended by the World Health

Organization. However, the use of the face mask has been uneven and determined by

individual, regional, cultural, and political factors. Based on data from the

Latinobarometer, we aim to understand the profile of people who used a mask in the

context of the COVID-19 pandemic in 18 countries of Latin America, between October

and November 2020, right before the mass vaccination campaigns. Results show that

women, older people, those with higher education, being employed and not working in

temporarily jobs, retirees, students, people with a centrist political ideology, and

Catholics, had a higher chance of using a face mask on a regular basis. People living in

Venezuela, Chile, Costa Rica and Brazil were the most likely to use face masks. These

results call attention to the need to understand social forces behind the willingness to

adopt non-pharmacological preventive measures in order to make them more effective in

health crisis emergencies.

Keywords: Sanitary face mask, COVID-19, Latin America, preventive measures,

socioeconomic status, political ideology

. CC-BY-NC-ND 4.0 International licenseIt is made available under a

perpetuity.

is the author/funder, who has granted medRxiv a license to display the preprint in(which was not certified by peer review)preprint

The copyright holder for thisthis version posted February 8, 2023. ; https://doi.org/10.1101/2023.02.07.23285570doi: medRxiv preprint

Introduction

The World Health Organization (WHO) recommended the use of face-masks as one of

the main measures to prevent COVID-19 transmission (1). However, the medical

discourse about preventive measures is mediated by a range of factors that affect their

implementation, such as gender constructions in health care (2), risk perception related to

age (3) and care attitudes linked to educational level (4). Political orientation and religious

affiliation are also important explanatory aspects of how the medical/epidemiological

discourses are decoded (5). In the specific context of COVID-19, the role of national

governments in recognizing the magnitude of the pandemic and their willingness to

implement the WHO health recommendations has also been decisive. The latter has been

especially relevant in Latin America, as political instability, low levels of investment in

public health systems and the idiosyncrasy of each government have defined different

public interventions and prevention messages sent to the population (6).

One of the studies that have addressed the individual determinants of prevention and

protection behaviors against the COVID-19 pandemic, concluded that the use of the face-

mask was related to the individual perception of risk, the feeling of responsibility and

solidarity, cultural and religious traditions, and the need to express their own identity (7).

In relation with the adoption of protective attitudes against respiratory epidemics, a study

prior to COVID-19, found that women were more inclined to apply non-pharmacological

preventive measures (like the use of mask, washing hands and avoid public transport),

while men were more prone to pharmacological measures (get vaccinated, use of

antivirals) (8). Regarding political ideology, being of right wing ideology in the United

States (US) was correlated with believing that the use of mask would not prevent the

expansion of the COVID-19 virus (9), and accordingly, the US was the only country were

government supporters were less likely to follow prevention health guidelines against

COVID-19 in a cross-country study (5), showing the importance of the government

position in individual prevention attitudes.

We did not find any work analyzing mask wearing determinants in Latin America.

However, an article warned about the limitations of the region to supply enough suitable

masks to the entire population and that these countries did not have previous experience

in the use of masks, contrary to Asian countries (10).

. CC-BY-NC-ND 4.0 International licenseIt is made available under a

perpetuity.

is the author/funder, who has granted medRxiv a license to display the preprint in(which was not certified by peer review)preprint

The copyright holder for thisthis version posted February 8, 2023. ; https://doi.org/10.1101/2023.02.07.23285570doi: medRxiv preprint

The use of a face-mask during the COVID-19 pandemic has not been a trivial act and it

is important to understand the individual, regional, cultural and political determinants that

influence this decision to obtain lessons and design more effective preventive campaigns

to manage future health crises. A transnational analysis in Latin America can allow us to

understand the different levels of influence that are interconnected in the individual

decision to wear a mask as a protective measure against the COVID-19 virus. This study

aims to discover the profile of people who wear masks, and understand the importance of

the cultural and sociopolitical context of individuals in Latin America. We carried out

logistic regression models controlling for all the demographic, socioeconomic and

country variables, relevant to understand the determinants of preventive behaviors.

Data and Methods

The Latinobarometer is a representative population survey conducted in 18 Latin

American countries, whose field work was between October 26 and November 17, 2020.

The sample includes 20,204 people, ranging from 1,000 to 1,200 adults interviewed by

country. We eliminated 127 minors, 89 people who did not answer the question about the

use of masks, and 3 people who did not report their educational level.

We performed a logistic regression of the binary dependent variable created from the

question about the use of mask on a regular basis to not get infected with the COVID-19

virus. The model included as independent variables the gender (male/female), age (18-

25; 26-40; 41-60, 61+), educational level (basic education not completed, basic education

completed, completed high school, completed higher education), current occupational

status (self-employed, salaried, retired or pensioned, unemployed and student), political

ideology (left, center-left, center-right, right), religion (Catholic, Evangelical, others,

none) and the country of residence (Argentina, Brazil, Bolivia, Chile, Colombia, Costa

Rica, Dominican Republic, Ecuador, El Salvador, Guatemala, Honduras, Mexico,

Nicaragua, Panama, Paraguay, Peru, Uruguay, Venezuela).

Results

Table 1 shows the results of the multivariate regression model, indicating women use

more the mask than men (OR=1.483; p-value<0.001). Younger people are less likely to

use a mask, although a greater effect is not observed among the population older than 60

years (OR= 1.562; p-value <0.001) compared to those between 41 and 60 years old

. CC-BY-NC-ND 4.0 International licenseIt is made available under a

perpetuity.

is the author/funder, who has granted medRxiv a license to display the preprint in(which was not certified by peer review)preprint

The copyright holder for thisthis version posted February 8, 2023. ; https://doi.org/10.1101/2023.02.07.23285570doi: medRxiv preprint

(OR=1.570; p-value <0.001). Those with higher educational level, have a higher chance

of wearing a mask, and differences are significant for people with higher education

(OR=2.161; p-value<0.001) and secondary education (OR=1.428; p-value<0.001).

Employees, temporarily unemployed people, retirees and students use the mask to a

greater extent than people who are self-employed. Self-perceived social class in the

bivariate analysis (results not shown here) indicates that the middle class is more likely

to use masks than the upper class, but in the multivariate analysis is not significant.

Regarding religion, Catholics have a higher chance of wearing the mask than Evangelicals

(OR=1.300; p-value<0.001) while people who declare no religion use it to a lesser extent

(OR=0.796; p-value <0.05). Results on political ideology indicate that people from the

center-right (OR=1.264; p-value<0.05) and center-left (OR=1.273; p-value<0.01) use the

mask to a greater extent than right-wing people. However, there are no significant

differences between people who is on the right with respect to those on the left or without

political ideology. Compared to Brazil, the reference category, the only country where

the use of masks is significantly higher is Venezuela (OR=2.070; p-value<0.01), although

Chile and Costa Rica also showed non-significant higher rates. In contrast, the use of the

mask is lower in Bolivia, Colombia, the Dominican Republic, Honduras, Mexico,

Nicaragua, Panama, Paraguay and Peru.

Discussion

Here we draw a profile of people who commonly used a mask between October and

November 2020, before the vaccination campaign started in Latin America. The results

of this analysis are consistent with previous studies that indicate that men are more prone

to risky behaviors than women (11). Although youngsters are less likely to use masks,

both the middle-age and oldest-age groups showed similar predisposition, indicating that

the age perception of the risk started earlier that the actual risk (12). Findings that having

a higher educational level implies a greater use of masks are consistent with previous

studies where higher educational levels are associated to healthier behaviors engagements

(4).

Regarding occupational categories, the fact that retirees or pensioners have a greater use

of the face mask than those who are self-employed is related to the fact that older people

are the most vulnerable group to the virus (3). Additionally, students might be more likely

to use them due to the mandatory use of face masks in educational centers. We also find

. CC-BY-NC-ND 4.0 International licenseIt is made available under a

perpetuity.

is the author/funder, who has granted medRxiv a license to display the preprint in(which was not certified by peer review)preprint

The copyright holder for thisthis version posted February 8, 2023. ; https://doi.org/10.1101/2023.02.07.23285570doi: medRxiv preprint

that employees, as well as the unemployed, are more likely to wear a mask than those

who are self-employed. In this sense, it should be noted that, although we control for the

current occupational situation, the survey does not have variables that allow us to control

which people can perform their daily lives without having to leave their home, where the

use of a face mask would not be so necessary. Social class does not show significant

results, but this might be a limitation of the self-perceive question to capture this

dimension. Unlike countries like the US, where political orientation shapes prevention

behaviors (5), the results of our study indicate that there is no left-right polarization of

mask use in Latin America. Interestingly, centrist people are the most likely to wear a

mask. In this sense, religion, which has also been present in discussions about the

pandemic (13), also shapes prevention behaviors. Those belonging to the Evangelical

religion wear a mask to a lesser extent than Catholics. These results open many questions

about the role of religions in promoting health behaviors and the arguments that certain

religions establish against medical discourse. Finally, Venezuela, Chile, Costa Rica and

Brazil are the countries in Latin America where the use of a mask is highest. In the case

of Brazil, a lack of rapid and equitable responses from the government might have led the

population to increase their prevention attitudes against the virus (14). Regarding

Venezuela, the country has deepened its political and economic crisis since COVID-19,

and there has been a lack of credibility from the citizenship in the reported cases because

of the shortage of virus tests (15).

The use of the face mask to prevent COVID-19 infection has been uneven across

individual, regional, cultural and ideological factors in Latin America. The analysis

showed the importance of understanding the demographic and socioeconomic profile of

those individuals more prone to use preventive measures to stop the spread of the

pandemic in the region, and the critical role of social forces–through religion and political

ideas–in shaping health-related attitudes in a health emergency crisis.

. CC-BY-NC-ND 4.0 International licenseIt is made available under a

perpetuity.

is the author/funder, who has granted medRxiv a license to display the preprint in(which was not certified by peer review)preprint

The copyright holder for thisthis version posted February 8, 2023. ; https://doi.org/10.1101/2023.02.07.23285570doi: medRxiv preprint

Table 1. Logistic regression over the regular use of face mask in Latin American

countries.

Variables Odds Ratio Std. err. Variables Odds Ratio Std. err.

Sex (Men) Political ideology (Right)

Women 1.483 (0.086) *** Center-Right 1.264 (0.07) *

Center-Left 1.273 (0.108) **

Age (18-25) Left 0.850 (0.144) *

26-40 1.185 (0.086) * None 1.227 (0.133)

41-60 1.570 (0.126) *** Missing 1.094 (0.127)

61+ 1.562 (0.165) ***

Country (Brazil)

Education (Incomplete Basic) Argentina 0.601 (0.117) *

Basic 1.119 (0.112) Bolivia 0.179 (0.03) ***

Secondary 1.428 (0.15) *** Chile 1.382 (0.326)

Tertiary 2.161 (0.262) *** Colombia 0.501 (0.092) ***

Costa Rica 1.251 (0.281)

Labor activity (self-employed) Dominican Rep. 0.503 (0.092) ***

Employee 1.429 (0.116) *** Ecuador 0.892 (0.182)

Unemployed 1.310 (0.134) * El Salvador 0.933 (0.191)

Retired 1.706 (0.283) ** Guatemala 0.976 (0.197)

Inactive 1.041 (0.075) Honduras 0.662 (0.124) *

Student 1.381 (0.209) * México 0.190 (0.032) ***

Nicaragua 0.341 (0.06) ***

Social Class (High) Panamá 0.649 (0.125) *

Medium 1.227 (0.13) Paraguay 0.401 (0.072) ***

Low 1.054 (0.106) Perú 0.659 (0.125) *

Missing 0.898 (0.13) Uruguay 0.788 (0.156)

Venezuela 2.070 (0.534) **

Religion (Evangelical)

Catholic 1.300 (0.094) *** Constante 6.433 (1.415) ***

Other 1.067 (0.149)

None 0.796 (0.068) ** N 19988

*** P<0.001; ** P<0.01, * P<0.05

. CC-BY-NC-ND 4.0 International licenseIt is made available under a

perpetuity.

is the author/funder, who has granted medRxiv a license to display the preprint in(which was not certified by peer review)preprint

The copyright holder for thisthis version posted February 8, 2023. ; https://doi.org/10.1101/2023.02.07.23285570doi: medRxiv preprint

Declaration of Competing Interest

This article adheres to ethical standards and complies with relevant guidelines for using

secondary survey data. Data upon the research is done is freely available and ensures the

confidentiality and privacy of the participants in accordance with relevant laws and

regulations. The analysis presented here are available upon request.

Funding

This work was supported by the Spanish Ministry of Science and Innovation, through

the Ramon y Cajal grant number [RYC-2017-22586]”, and the project of the National

R&D&I Plan COMORHEALTHSES [PID2020-113934RB-I00]; the Economic and

Social Research Council (ESRC) through the Research Centre on Micro-Social Change

(MiSoC) at the University of Essex, grant number ES/S012486/1; and support from

CERCA Programme Generalitat de Catalunya.

. CC-BY-NC-ND 4.0 International licenseIt is made available under a

perpetuity.

is the author/funder, who has granted medRxiv a license to display the preprint in(which was not certified by peer review)preprint

The copyright holder for thisthis version posted February 8, 2023. ; https://doi.org/10.1101/2023.02.07.23285570doi: medRxiv preprint

References

1. WHO. Advice on the use of masks in the context of COVID-19: interim guidance.

World Health Organization. 4 6, 2020.

2. Dean, K. Self-care components of lifestyles: The importance of gender, attitudes and

the social situation. Social Science & Medicine. 1989, Vol. 29, 2, pp. 137-152.

3. Soresen, R. J. D. et al. (COVID-19 Forecasting Team). Variation in the COVID-19

infection–fatality ratio by age, time, and geography during the pre-vaccine era: a

systematic analysis. The Lancet. February 24, 2022.

4. Cutler, D.M., Lleras-Muney, A. Understanding differences in health behaviors by

education. Journal of Health Economics. 2010, Vol. 29, 1, pp. 1-28.

5. Becher, M., Stegmueller, D., Brouard, S., & Kerrouche, E. Ideology and compliance

with health guidelines during the COVID‐19 pandemic: A comparative perspective.

Social science quarterly. 2021, Vol. 102, 5, pp. 2106-2123.

6. García, P.J., Alarcón, A., Bayer, A., Buss, P., Guerra, G., Ribeiro, H., Rojas, K.,

Saenz, R., Salgado de Snyder, N., Solimano, G., Torres, R., Tobar, S., Tuesca, R.,

Vargas, G., Atun, R. COVID-19 Response in Latin America. Am J Trop Med Hyg.

2020, 1.

7. Martinelli, L., Kopilaš, V., Vidmar, M., Heavin, C., Machado, H., Todorović, Z.,

Buzas, N., Pot, M., Prainsack, B., Gajović3, S. Face Masks During the COVID-19

Pandemic: A Simple Protection Tool With Many Meanings. Front Public Health, 8.

2021.

8. Moran, K. R., Del Valle, S. Y. A Meta-Analysis of the Association between Gender

and Protective Behaviors in Response to Respiratory Epidemics and Pandemics. PLoS

ONE. 2016, Vol. 11, 10, p. e0164541.

9. Bir, C., Widmar, N. O. Societal values and mask usage for COVID-19 control in the

US. Preventive Medicine. 2021, Vol. 153, 106784.

10. Moreno del Soto, D. Cardona Maya, W. D., Londoño Agudelo, E., Bueno-Sánchez,

J. C. The feasibility of generalized face mask usage during the COVID-19 pandemic: a

perspective from Latin America. Infection Control & Hospital Epidemiology. 2021,

Vol. 42, 2, pp. 245 – 246.

. CC-BY-NC-ND 4.0 International licenseIt is made available under a

perpetuity.

is the author/funder, who has granted medRxiv a license to display the preprint in(which was not certified by peer review)preprint

The copyright holder for thisthis version posted February 8, 2023. ; https://doi.org/10.1101/2023.02.07.23285570doi: medRxiv preprint

11. Lawlor, D.A., Ebrahim, S., Davey Smith, G. Sex matters: Secular and geographical

trends in sex differences in coronary heart disease mortality. BMJ. 2001, Vol. 323, pp.

541–545.

12. Zengh, S., Peng, F., Xu, B., et al. Risk factors of critical & mortal COVID-19 cases:

A systematic literature review and meta-analysis. . J Infect. 2020, Vol. 81, 2, pp. e16-

e25.

13. Barmania, S., Reiss, M. J. Health promotion perspectives on the COVID-19

pandemic: The importance of religion. Global Health Promotion. 2020, Vol. 28, 1, pp.

15-22.

14. Castro, M. S., Kim, S., Barbeira, L., Ribeiro, A. F., Gurzenda, S., Ribeiro, K. B.,

Abbott, E., Blossom, J., Rache, B., Singer, B. H. Spatiotemporal pattern of COVID-19

spread in Brazil. Science. 2021, Vol. 372, 6544, pp. 821-826.

15. Rodriguez-Morales, A. J., Figuera, M. E. COVID-19 in Colombia and Venezuela:

Two sides of the coin. American Journal of Public Health. 2022, Vol. 112, 56, pp. 5581-

5584.

. CC-BY-NC-ND 4.0 International licenseIt is made available under a

perpetuity.

is the author/funder, who has granted medRxiv a license to display the preprint in(which was not certified by peer review)preprint

The copyright holder for thisthis version posted February 8, 2023. ; https://doi.org/10.1101/2023.02.07.23285570doi: medRxiv preprint