Face mask use among pedestrians during the COVID-19 pandemic in southwest Iran: An observational study on 10440 people

Background: Many countries have recommended the use of face masks for general population in public places to reduce the risk of COVID-19 transmission. This study aims to estimate the prevalence of face mask usage and investigate about different types of face mask and their distribution among pedestrians in southwest Iran during the Covid-19 pandemic. Methods: This cross-sectional study was conducted in August 2020 in Ahvaz, southwest Iran. Using a multistage sampling method, a total of 10440 pedestrians selected from 8 urban districts and 92 neighborhoods of the city. The data gathered by observation method. Percentage, mean and standard deviation were used to describe the variables. Chi-square test, sher exact test and Chi-square for trend used to assess relationship between two categorical variables. We used unconditional logistic regression model to control confounders. Results: The mean ± SD age of the observed pedestrians was 32.2 ±15.1 years and 67.9% of them were male. The overall prevalence of face mask usage was 45.6% (95% CI, 44.6 – 46.5). In general, as the age increased, the prevalence of face mask use signicantly increased (p for trend <0.001). Women used face masks signicantly higher than men (60.2% vs. 38.7%, p <0.001). Among the pedestrians who used the mask, 75.6 % wore facemask correctly. The most common type of facemask used by the pedestrians were surgical(medical) masks (63.8%). In total, the prevalence of facemask usage was signicantly higher during a.m. (49.4%) compared to p.m. (43.9%), (p<0.001). Besides, in our study, 1.7% and 0.3 % of Pedestrians had worn gloves and shielded respectively. Women used shields and gloves signicantly higher than men (3.6 % vs. 0.7%, p <0.001). Also, women used shields more than men (0.5 % vs. 0.3%, p =0.036). Conclusion: We concluded that the prevalence rate of face mask use in Ahvaz was fairly low especially in men and younger people.


Background
Severe acute respiratory syndrome corona virus 2 (SARS-CoV-2) or corona virus disease 2019 (COVID- 19) was initially reported from Wuhan, China on 31 December 2019. The World Health Organization (WHO) declared COVID-19 to be a pandemic on 11 March 2020. It is an ongoing global pandemic now [1].
According to the current evidence, Corona virus is mainly transmitted between individuals via respiratory droplets and contact routes, primarily from person to person during coughing, sneezing, talking [2].
Because there is not any effective treatment or vaccine against COVID-19 yet, personal protective measures including personal protective equipment (PPE) like masks, respirators (i.e. N95 or FFP2), shields and gloves can be used to prevent the infection [3,4] . Social distancing and maintaining hand hygiene are the key strategies to prevent COVID-19 transmission (rational) whereas effectiveness of face Page 3/14 mask usage by the healthy people in the community against COVID-19 is controversial, though increasingly recommended [5,6]. The WHO had not yet recommended mass masking for healthy individuals in the all communities to prevent transmission of COVID-19 in its interim guidance of April 6, 2020, although it suggests the general public to wear a fabric mask in communities with widespread transmission, and especially in settings where physical distancing is di cult to maintain [2]. In the other hand, US Centre for Disease Control (CDC) recommends that people wearing masks in public places to reduce the risk of COVID-19 transmission, when around people outside of their household, especially in settings that social distancing cannot be maintained [4]. However, the use of mask alone is not su cient to protect a person against COVID-19. It is also necessary to maintain a minimum physical distance from others, frequently washing hands and to avoid touching face (2). Although it is partly unknown the degree that masks protect against droplets/aerosols of respiratory system, [6] but even with a limited protective effect, face masks can reduce the risk of transmission of COVID-19 in the general public [3,[7][8][9] . They also can presumptively diminish the viral load, resulting to decrease the severity and risk of the death [10].
Masks in point of view ltration e ciency and breathability, can be different. Hence, it is important to use each type of mask in proper setting and situation. Medical masks and respirators are recommended to provide care to suspected or con rmed COVID-19 patients, not in public settings whereas wearing nonmedical (Fabric) masks are recommended in public settings [2,4].
A national o cial report announced incidence rate of COVID-19 in Ahvaz is among the highest in Iran [13]. Regarding no data is available on frequency of face mask use in Ahvaz yet, this study aims to estimate the prevalence of face mask usage and investigate about different types of face mask and their distribution among pedestrians in Ahvaz, southwest Iran during the Covid-19 pandemic. This study also aims to assess the acceptance rate of the face mask practice worn by pedestrians. We believe these preliminary ndings will help policymakers, managers and health professionals to design and implement their interventional programs.

Type of Study
This population-based cross-sectional study was conducted during 10 days, from 2 to 11 August 2020 in Ahvaz, southwest Iran. A total of 10440 pedestrians selected from 8 urban districts and 92 neighborhoods of the city. Pedestrians mask use behavior was accessed via observation. The data gathering was based on observation of passers-by in street because the observation method usually is more accurate and more valid than the self-reporting approach for assessing behaviors. This study was approved by Ethics Committee of Ahvaz Jundishapur University of Medical Sciences (IR.AJUMS.REC.1399.396).
Inclusion and Exclusion criteriaInclusion criterion for this study was all pedestrians ≥ 2 years old in the city. Exclusion criteria for the study were: 1) fully covered face so that the observer cannot detect whether pedestrian wears the mask and, 2) be exposed to the observer for a short time so that it is not possible to record the required information.

Study Setting
The metropolitan city of Ahvaz is located in the southwest Iran. Ahvaz is the capital of Khuzestan province. It's population according to the 2019 census is 1292752. Ahvaz has a subtropical hot desert climate with long summers and short winters. The temperature sometimes exceeds 50 °C during summers and the humidity in sometimes reaches more than 90%.

Data collection
In this study, The data were gathered by 8 well trained observers. Observation stations were determined according to urban divisions of the city and allocated proportional to the population size, living in each district and neighborhood. These stations were determined from detailed maps of urban divisions and were selected from the crowded passages of each neighborhood. At each station, data of 60 pedestrians were collected including gender, approximate age, use of facemask, gloves and shield, type of facemask, and correct use of facemask. Insu cient coverage of the mouth and nose, wearing facemask upside down or inside-out were de ned as "incorrect or unacceptable" use of the mask. Observation was performed during the busy hours of each area from 9.00 to 13.00 and 17.00 to 23.00.

Sample size and Sampling method
In order to determine the minimum sample size, we used the formula for estimating a population proportion. For this purpose, α= 0.05, p=0.5, d=0.04 and a design effect equal to 1.6 were considered. A minimum sample size of 960 estimated for each district. Regarding to the unequal size of the districts and using a proportional to size sampling method, the nal sample size needed for this study was estimated 10,440 pedestrians. In total,174 clusters of 60 people from 93 urban neighborhood of Ahvaz were assessed in this study.
We used a Multistage sampling method in this research. At the rst stage, each of 8 urban districts was considered as a stratum. Then de ned number of clusters were assigned to each neighborhood. Each cluster was consisting of 60 pedestrians. The location of the observation stations was determined by a targeted sampling strategy from the busy passages of each neighborhood. A non-probability convenience sampling method was used for the last stage. In such a way that at each observation station, after deploying the observer, the closest pedestrian to the observer was selected as the rst sample and entered to the study. After nished recording the data of the rst person, the next closest person to the observer selected as the next sample and this was continued until the total number of selected persons in each cluster reach to 60.

Statistical analysis
Descriptive statistical measures including mean, standard deviation and percent used to describe the data. The estimated prevalence rates presented with 95% con dence interval (95% CI). Chi-square test, sher exact test and Chi-square for trend used to assess relationship between two categorical variables.
We used unconditional logistic regression model to control potential confounders. Odds ratios used to assess strength of the associations. The Statistical signi cance was declared if the p-value was less than 0.05. The analyses were carried out with SPSS version 22.

Ethics
The Ethics Committee of Ahvaz Jundishapur University of Medical Sciences (AJUMS.REC.1399.396) con rmed the morality and ethics of the study.

Results
A total number of 10440 pedestrians were assessed in terms of facemask usage. The mean ± SD age of the observed pedestrians was 32.2 ±15.1 years. Among the studied individuals, 7072(67.9%) were male. Demographic characteristic and frequency of personal protective measures among of the assessed pedestrians are shown in Table 1.
Among the observed pedestrians, 4749 people had used facemasks. The overall prevalence of facemask usage was 45.6% (95% CI, 44.6 -46.5). The prevalence rates of facemask usage by the assessed factors and their 95% con dence intervals are presented in Table 2. The highest prevalence of facemask usage was seen in the age group of 70 years and older, and the lowest was seen in the age group of less than 10 years, 71.7% and 26.6% respectively. In general, as the age increased, the prevalence of facemask usage signi cantly increased too (p for trend <0.001). This trend was obviously seen for men but women showed different pattern so that facemask usage in age groups under 10 and, 70 and older were low while the prevalence of facemask usage among the other age groups were higher and almost the same (graph 1). In total, women used facemasks signi cantly higher than men (60.2% vs. 38.7%, p <0.001).Please insert graph1.
The pedestrians in the eastern area of Ahvaz city worn facemasks signi cantly more than the western area (48.9% vs. 41.5%, p<0.001). The prevalence of facemask usage among pedestrians of the eight districts of Ahvaz were highly different, so that the highest prevalence was observed in district two and the lowest was seen in district six, 63.6% and 20.9% respectively. The prevalence of facemask usage was even more different among the neighborhoods. It ranged between 1.7% to 78.3%.
The most common type of facemask used by the observed pedestrians were surgical(medical) masks (63.8%) and the lowest were the ltered masks including N95 respirators (15.4%). Older pedestrians wore ltered masks more than younger pedestrians while younger pedestrians used cloth mask more frequently (p= 0.002). However, we did not nd any signi cant association between the type of used facemasks and gender (p= 0.44)Among the pedestrians who used the mask, 75.6 % wore facemask correctly, 12.6% did not cover completely their mouths and noses, 9.9% wore inside-out and 1.9% wore upside-down. Acceptable facemask practice was signi cantly higher in women than men (78% vs.73.9%; p =0.017) whereas there was no statistically signi cant difference between the age and correct wearing of facemasks (p = 0. 19). Besides, in our study, 1.7% and 0.3 % of Pedestrians had worn gloves and shielded respectively. Women used gloves signi cantly higher than men (3.6 % vs. 0.7%, p <0.001) also wearing gloves was higher in older people (p=0.033). Also, women used shields more than men (0.5 % vs. 0.3%, p =0.036) but no signi cant association was found between use of shields and age (p=0.34) ( Table  3).
The observations and data gathering were occurred between 9:00 o'clock to 23.00 o'clock. The lowest prevalence of facemask usage was seen at 13.00 (37.5%) while the highest was observed at 23.00 (67.9%). In total, the prevalence of facemask usage was signi cantly higher during a.m. (49.4%) compared to p.m. (43.9%), (p<0.001). The prevalence rates of facemask usage at different times of day are presented in graph 2.
Because age and sex could play confounder role in this assessment, unconditional logistic regression model was performed to control effects of these potential confounders. After controlling for age and sex, we observed a signi cant association between prevalence of facemask usage and time of observations so that the odds of facemask usage during a.m. was 26 percent higher than p.m. (Odds Ratio: OR=1.26; 95% CI, 1.16-1.38; p<0.001). The results are presented in Table 4.

Discussion
At present, COVID-19 has sparked a pandemic and is spreading rapidly in many countries [14]. Because there is no vaccine or effective treatment for this disease, conducting interventions such as use of facemask, social distancing and washing hands are urgently needed to limit the transmission. Despite the WHO primary recommendations against universal masking, emphasis on this strategy is increasing in the world. In Iran, the law on the use of masks in public places, government o ces and banks was implemented on June 4.
In this study the prevalence of facemask usage was low (45.6%). This rate has been ine cient to control the disease hence, Ahvaz was in a critical situation and in the red zone for several months.
In the present study the prevalence of facemask usage was much lower than the rates from Hong Kong study among pedestrians [15], Malaysia study among hospital visitors [16] and Malaysia study on general public at wet markets [17] 94.8%, 96.9% and 99.7% respectively. Similar to our nding observed in Bangkok airport (46%) [18]. The prevalence rates of face mask use in Lima, Paris, Boston and Atlanta airports (27% , 4% , 3% and 2% respectively) [18] were much lower than Ahvaz. The observed differences can be due to demographic and cultural characteristics of the assessed population, different methods of data gathering, policy of the governments about mass masking and the risk of COVID19 transmission in the countries.
In our study the highest prevalence of face mask usage was in the age group of 70 years and older (71.7%). Our result showed the prevalence of face mask usage increased with age. Similar ndings were reported among the elderly in Japan (aged 60-69) [19] and Australia (aged 65-74) [20], the percentages face mask usage were 43.6% and > 60%, respectively. This may be due to the perception of higher risk of morbidity and death due to COVID19 for higher age groups.
Besides, the prevalence of face mask use in women was signi cantly higher than men (60.2% vs. 38.7%, p <0.001). This could be due to the fact that women generally pay more attention to their health status and making healthy behaviors. Conversely, the prevalence rates observed in Malaysia studies showed no difference in both sex [16,17].
The observed differences in prevalence rates of face mask usage were impressive among the districts and neighborhoods of Ahvaz. This could be mostly due to the differences in socio-economic status. The low socioeconomic level usually leads to low health literacy and public awareness, lack of access to masks, as well as low purchasing power.
The most common type of mask in our study was surgical mask (63.8%). The same nding reported by Gunasegaram et al. [16,17] and Tam et al. [15].This can be questionable because the WHO and CDC did not recommend the use of surgical masks in general population [2,21]. In contrast, they recommended using cloth masks in public setting .This type of masks can be easily manufactured or made at home and reused after washing [22] and it is more affordable than other masks. Besides, we found that the types of face masks were differently used among the age groups. Filtered masks were used higher by older pedestrians while younger people used cloth face masks much higher than the older pedestrians.
People in higher age groups usually ll more risk of COVID-19 so they may be use more frequently ltered masks with the purpose of their higher protection.
Our ndings showed higher prevalence rates of face mask usage during am hours in relation to pm hours, that it may be due to weather conditions especially the higher temperatures in the afternoon. The similar nding reported by Cheng et al. [5].
Wearing properly a mask is necessary to get the maximum protection against COVID19 [23]. In this study, acceptable rate of using masks among the observed pedestrians was 75.6%. The percentage of acceptable face mask practice in our study was lower than some similar studies. This rates were reported from Malaysia about 95.63% and 88.75% [16,17] and from Hong Kong about 87 % [15] . Besides, we found that the correct practice of face mask use in women was higher than men. This can be due to better following the health protocols by women.
Our study had a number of limitations. Due to use of observation method for the data gathering, we could not assess some important factors like socioeconomic status and the reasons for not wearing masks.
Besides, we did not ask exact age of the subjects and approximate ages were recorded instead. Therefore, a non-differential misclassi cation can be occurred in the age grouping.
This investigation had some major strengths. Using observation method for data gathering in this study can leads to more valid data in comparison to use of questionnaires and self-reporting method. Furthermore, our large sample size guaranteed su cient statistical power and precise estimation of the rates so that the calculated con dence intervals are mostly narrow.

Conclusion
In summary, we found that the overall prevalence rate of face mask usage in Ahvaz was fairly low especially in men. Hence, the observed rates of mask usage probably cannot protect the community against COVID-19 spread. Therefore, it is important to plan and conduct educational programs to promote healthy behaviors in the community, especially for the high risk groups. Besides, to stablish laws and regulations governing the use of face masks in public places is necessary for increasing the rate of mask population coverage.