Predicting Preventive Behaviors against COVID-19: A Structural Equation Modeling Approach from Iran : WHO South-East Asia Journal of Public Health

Secondary Logo

Journal Logo

Original Article

Predicting Preventive Behaviors against COVID-19

A Structural Equation Modeling Approach from Iran

Bastami, Fatemeh; Motlagh, Soraya Nouraei; Rahimzadeh, Seyedeh Faezeh1; Almasian, Mohammad2; Zareban, Iraj3; Ebrahimzadeh, Farzad4,

Author Information
WHO South-East Asia Journal of Public Health 11(2):p 79-86, Jul–Dec 2022. | DOI: 10.4103/WHO-SEAJPH.WHO-SEAJPH_56_22
  • Open



The purpose of this study was to use the health belief model (HBM) to predict the adoption of preventive behaviors against COVID-19 using the structural equation modeling approach.


As a descriptive-analytical study, this research was conducted on 831 men and women who were under the coverage of comprehensive health service centers in the Lorestan province of Iran in 2021. A questionnaire based on HBM was used to collect data. Data were analyzed by the SPSS software version 22 and AMOS version 21.


The mean age of participants was 33.0 ± 8.5, with a range of 15–68 years. The constructs of the HBM explained about 31.7% of the variance in COVID-19-related preventive behaviors. The greatest total effect on preventive behaviors against the COVID-19 disease belonged to the constructs of perceived self-efficacy (0.370), perceived barriers (−0.294), and perceived benefits (0.270), in descending order of impact.


Educational interventions can be useful in promoting COVID-19 preventive behaviors by bringing about a correct understanding of self-efficacy, barriers, and benefits.


The COVID-19 pandemic led to 98,925,221 cases of human infection and 2,127,294 deaths by January 26, 2021. According to the figures published by the World Health Organization, in Iran, 1,379,286 cases of infection were reported by the same date, 57,481 of which ended in death.[1] The increasing morbidity and mortality rates indicate that people do not engage in the necessary preventive behaviors. Therefore, it is crucial to identify factors affecting behavior change to promote health-related behaviors.[2,3]

Previous studies conducted in Iran have shown that despite the guidelines recommended by international, national, and regional organizations, the level of adherence to preventive behaviors against the coronavirus disease is still not satisfactory.[4,5] Therefore, it seems necessary to identify the factors affecting compliance with the recommended guidelines.[6] Behavioral models have been designed to identify these factors. If selected correctly, behavior change models can help with the identification of factors affecting health-related behaviors. Such knowledge is of great importance in designing educational interventions.[7] Previous studies have confirmed the health belief model (HBM) as a useful model to explain and predict preventive behaviors against infectious diseases.[7,8] The HBM is considered a comprehensive model in health and behavioral sciences.[9] Based on this model, people adopt health-related preventive behaviors when the following factors affect them: perceived susceptibility: the understanding and belief that they are susceptible to a disease; perceived severity: the understanding and belief that the health problem can be serious and troublesome; the sum of these two constructs is understood as perceived threat; perceived barriers: physical, psychological, financial, and other barriers that an individual is likely to face in adopting a health-related behavior; perceived benefits: individual's perception about the benefits or effects of the target behavior or adherence to recommendations in preventing the disease or reducing its severity and side effects; cues to action: health notices and messages and mass communication by trusted individuals and groups, such as doctors, political or religious leaders, and mass media that influences behavior in the target group; self-efficacy: an individual's self-confidence and estimation of how much certain behaviors will lead to the desired results; and finally, action: the observance and implementation of the desired health behavior that is affected by the above constructs.[10,11,12,13,14]

Since health outcomes such as disease prevention behaviors often have multiple causes and contributing factors and their study requires the consideration of multiple variables and constructs and their complex interrelationships, the application of structural equation models can be of great value. Structural equation modeling (SEM) is a multivariate analysis technique that allows the researcher to simultaneously test a set of regression equations and examine the relationships among different variables concomitantly. The importance of this technique in medical science research is that, in this field of study, researchers often examine the links between different variables in the form of a model or network of relationships. Therefore, based on the hypotheses that emerge about the relationships between variables, a general schema of these relationships is designed in the form of a prefabricated model. In such situations, researchers are faced with the fundamental question of whether the structure of the prefabricated model is actually supported by the data. The distinguishing feature of SEM is the capacity to examine the relationship fit between the study variables as well as to classify and isolate measurement errors from other errors in the model. Furthermore, the correlation between errors, which is a limitation in many classical models, is allowed and considered in this model.[15,16] In the present study, direct and indirect approaches to SEM were used to predict the performance of preventive behaviors against the COVID-19 disease using the constructs of the HBM by men and women residing in the urban and rural areas of the Lorestan province of Iran.


Study design

Participants in this cross-sectional study were male and female residents of urban and rural areas in the Lorestan province located in western Iran in 2021. The inclusion criteria consisted of being at least 15 years old, willingness to participate in the study, and being a resident of the Lorestan province. The exclusion criteria included suffering from COVID-19 at the time of the study.

Sampling and sample size

The sampling method was a combination of stratified and multistage cluster sampling. The cities of the Lorestan province formed clusters, and in addition to the capital of the province (Khorramabad), four other cities (clusters) were selected by a systematic random sampling way on the map of the province such that they are scattered and not limited to one geographical point. Within each cluster, there were two strata: urban and rural. In each city, in turn, there were three substrata: the northern, the central, and the southern sectors. In each geographical substratum, there were several comprehensive health service centers providing health services to the whole population of the city, such that each urban household has health records in a certain health center.

A comprehensive health service center was randomly selected from the northern, southern, and central sectors of the city. In each rural area, there were several rural health service centers. In each rural area, two rural health service centers were randomly and systematically selected on the map. In each selected center, quota sampling proportional to the total sample size was used, such that larger sample sizes were allocated to comprehensive health service centers covering larger populations. Then, households were selected based on the file number of households using a systematic sampling method proportionate to the total sample size. In each household, two individuals (if possible, a man and a woman) were selected.

Given structural equation models were for data analysis and as in these studies, the sample size was considered between five and ten times the number of free parameters of the model.[17,18] The initial sample size was estimated at ten times the free parameters of the model, i.e., 540. However, since the cluster sampling method was used and because of the design effect, the final sample size was estimated as 1.5 times this value, namely 810 people. After factoring in dropout rates and irrelevant data, ultimately 932 individuals were selected to participate in the study.

Data collection

Data were collected using a questionnaire based on the HBM. Its validity and reliability have been confirmed in a previous study.[5]

The first part of the questionnaire contained items on demographic/background characteristics, including age, sex, education, spouse's education, marital status, employment status, place of residence, and history of underlying diseases.

The second part consisted of 10 questions on perceived susceptibility, 5 questions on perceived severity, 5 questions on perceived benefits, 8 questions on perceived barriers, 8 questions on perceived self-efficacy, 6 questions on internal cues to action, and 5 questions on external cues to action. All items were rated on a five-point Likert-type scale ranging from “strongly disagree” to “strongly agree.”

The third part included 7 questions on activities over the past 2 months in the prevention of COVID-19. Each question's score ranged from 0 to 4, such that “never” = 0, “rarely” = 1, “sometimes” = 2, “often” = 3, and “always” = 4.

Because of the pandemic restrictions, telephone interviews were conducted by health-care personnel complete the questionnaires.


After evaluating validity, the test–retest method as an indicator of the repeatability of the measure was used to assess the reliability of the questionnaire. For this purpose, the validated version of the questionnaire was given to 25 people in Lorestan and then a week later they were asked to complete the questionnaires again and finally the intraclass correlation coefficient was calculated for each item which turned out to range from 0.42 to 0.98. To evaluate the internal consistency of the instrument, the Cronbach's alpha coefficient was used on a sample size of 793 people, which was 0.863 for the whole questionnaire and 0.857 for the construct of perceived susceptibility, 0.846 for perceived severity, 0.703 for perceived barriers, 0.880 for perceived benefits, 0.924 for perceived self-efficacy, 0.839 for internal cues to action, 0.866 for external cues to action, and 0.846 for performance.

Statistical analysis

Descriptive statistical methods (frequency distribution tables, mean, and standard deviation) were used to analyze the data in line with the objectives of the research. For intergroup univariate comparisons, inferential statistics, such as the independent t-test, one-way analysis of variance, and Tukey's honestly significant difference post hoc test for pairwise comparisons, were used. To investigate the direct effects, indirect effects, and total effects of HBM components on the construct of behavior, a structural equation model was used. The data were analyzed by IBM SPSS Amos 21. Chicago, IL: IBM Software Group; 2012, and the significance level was considered at 0.05.

Ethical considerations

This study was approved by the Ethics Committee of the Lorestan University of Medical Sciences with the code IR.LUMS.REC.1399.008. Participants were informed about the objectives of the study. Informed oral consent was obtained from them to participate in this study.


Descriptive statistics

Of the 932 questionnaires distributed, 831 (89.2%) were completed by telephone interviews. The mean age of participants was 33.1 ± 8.5, ranging from 15 to 68 years. About 46.9% of the participants were residents of Khorramabad, 66.8% lived in urban areas, 43.9% belonged to the age group of 30–39 years, and 77.3% were female [Table 1].

Table 1:
The means and standard deviations of the scores of the components of the health belief model and preventive behaviors against COVID-19 by demographic and background variables

Univariate analysis

Based on one-way analysis of variance/independent t-test, there were statistically significant relationships among behaviour and gender (P < 0.001), city of residence (P < 0.001), and the presence of underlying disease (P = 0.007). Table 1 compares the scores of the various components of the HBM and behaviour by demographic and background variables, as well as the result of the Tukey's test for pairwise comparisons.

The most frequent preventive behaviors against COVID-19 included covering the mouth when sneezing or coughing (74.1%) with the highest frequency, refraining from shaking hands with or kissing others (70.6%), using masks in public places (64%), and washing the hands daily for at least 20 s (38%), which was the least common.

Recommendations of health-care professionals with the highest frequency (46.1%), recommendations broadcast on radio and television (40.2%), family and acquaintances (40.2%), and newspapers with the lowest frequency (33.3%) acted as external cues to action in adopting preventive behaviors against the coronavirus disease. The importance of staying healthy with the highest frequency (53.2%) and having a sense of responsibility toward others and society with the lowest frequency (29.6%) were the internal cues to action in adopting preventive behaviors against COVID-19.

Structural equation modeling

After excluding 38 outliers based on Mahalanobis distances, 793 people entered multivariate analysis. One item was excluded from the perceived barriers subscale due to multiple collinearities with other items (variance inflation factors >5), and three items with a standardized factor loading of <0.4 were excluded from the perceived susceptibility subscale, resulting in a reduction in the number of questionnaire items from 54 to 50 [].

Table 2 presents the fit indices of the structural equation model. In this model, the value of the Chi-squared statistic is 3190.139, which is divided by the degree of freedom (1080) giving 2.954 (P < 0.001), which is acceptable due to the high sample size and the large number of variables.[19]

Table 2:
Goodness of fit measures for assessing the proposed model in prediction of preventive behaviors against COVID-19

The value of the root mean square error of approximation index is 0.05, which is evaluated as excellent (95% confidence interval: 0.048–0.052, P = 0.604). Other fitness indices were good (between 0.800 and 0.899) or excellent (≥0.900).[17] Other details of the model fit indices are given in Table 2.

Figure 1 shows the standardized direct path coefficients associated with predicting preventive behaviors against COVID-19 based on the HBM. Based on this SEM, the constructs of the HBM explained about 31.7% of the variance in COVID-19 preventive behaviors.

Figure 1:
Standardized direct path coefficients in the proposed Structural Equation Modelling (SEM). β indicates the standardized direct path coefficient, R 2 is indicative of the coefficient of determination, r is the correlation coefficient, and P means the p value. Unbroken lines show significant relationships and broken lines point out insignificant relationships. “**” shows significance at the 0.01 level and the “*” sign shows significance at the 0.05 level

As shown in Figure 1, the direct effect of internal cues to action on COVID-19 preventive behaviors was statistically significant (β = 0.094, P = 0.009) and the direct effect of external cues to action on COVID-19 preventive behaviors was insignificant but considerable (β = 0.070, P = 0.063). In addition, the direct effect of the construct of perceived benefits on COVID-19 preventive behaviors was significant (β = 0.102, P = 0.017) and also the direct effect of the perceived barriers construct was significant on such behaviors (β = −0.163, P < 0.001). Finally, the direct effect of the construct of internal cues to action on preventive behaviors against COVID-19 was also significant (β = 0.370, P < 0.001).

Table 3 details the direct, indirect, and total effects of the HBM on preventive behaviors against COVID-19. According to this table, the greatest overall effect on COVID-19-related preventive behaviors belonged to perceived self-efficacy, perceived barriers, perceived benefits, and internal cues to action, in descending order of impact [Table 3].

Table 3:
Direct, indirect, and total effects derived from standardized regression coefficients of the proposed standard error of mean


This study aimed to determine the predictive power of the HBM regarding preventive behaviors against the COVID-19 disease using the SEM approach. Based on SEM, the constructs of the HBM explained about 31.7% of the variance in COVID-19-related preventive behaviors. Previous studies in Mazandaran and the Iranian adult population using the HBM predicted 26% and 29.3% of the variance of COVID-19 preventive behaviors, respectively.[20,21] The present study was performed using the SEM approach. In addition to reporting the overall variance of the model in predicting behaviors, the direct effects, indirect effects, and the effect of all variables were measured. However, these relationships have not been examined in previous studies in Iran.[20,21,22]

The present study showed that among the constructs of the HBM, the greatest overall effect on preventive behaviors against COVID-19 belongs to the construct of perceived self-efficacy. A previous study on influenza preventive behaviors among students showed a positive and significant correlation among perceived self-efficacy with preventive behaviors.[23] In a study conducted in Mazandaran based on the HBM, it was shown that perceived self-efficacy was the most powerful predictor.[20] These results suggest that people with high levels of self-efficacy understand the risks of COVID-19 infection and its complications and perform more preventative behaviors against COVID-19.

In the present study, perceived barriers had a direct effect on both the adoption of COVID-19 preventive behaviors and ranked second in terms of the overall effect on the adoption of these behaviors. Qualitative results of a mixed methods study showed that the barriers to adopting preventive behaviors against COVID-19 were divided into 6 categories including social, psychological, religious, political, health system, and informational barriers. Furthermore, the quantitative results of this study showed that managerial-policymaking factors and then psychological factors were the most important barriers to self-care from the perspective of people.[24]

In the present study, perceived benefits had a direct effect both on the adoption of COVID-19 preventive behaviors and in relation to the overall effect on the adoption of these behaviors. Previous studies inside Iran and abroad showed that perceived benefits were strongly associated with the observance of preventive behaviors in relation to the influenza pandemic.[25,26,27] The high effect of perceived benefits on the adoption of COVID-19 disease preventive behaviors reflects the high perception of the participants of the benefits of performing behaviors to prevent COVID-19.

In the present study, internal and external cues to action had a direct effect on the adoption of COVID-19 preventive behaviors. In a previous study, cues to action were important predictors of pursuing colorectal cancer tests in individuals.[28]

In the present study, internal and external cues to action together predicted the perceived susceptibility variable as an intrinsic variable. This means that these cues to action increase awareness and increase people's perceived susceptibility, thus making people consider themselves susceptible to the disease. In a study conducted in China following the outbreak of COVID-19, perceived susceptibility to COVID-19 was found to be the strongest predictor of behavior change.[29] The high perceived susceptibility of the individuals indicates that they believe that they have a higher chance of being exposed to the disease if they do not follow COVID-19 related preventive behaviors. Therefore, the presence of this high susceptibility will make them voluntarily take action in performing preventive behaviors.

In the present study, the constructs of internal and external cues to action along with perceived susceptibility predicted perceived severity as an intrinsic variable. In the present study, the mean perceived susceptibility and severity were low and had a nonsignificant relationship with preventive behaviors. A previous study using the HBM to predict COVID-19 preventive behaviors among a sample of the Iranian adult population showed that perceived susceptibility and severity were not significant in the regression model.[21] A previous study in Australia found that only 33% believed that their health was seriously affected during the pandemic period.[30] A previous study in Hong Kong also found that people did not take the risk of disease seriously.[31] Belief in susceptibility to the COVID-19 disease increases people's perception of the severity and seriousness of the complications of the disease. It also makes people more resilient in performing preventive behaviors.

The results of the present study show that women and men are significantly different in terms of adopting preventive behaviors against COVID-19 so that the score of this variable was higher in women than men. Previous studies conducted in Iran showed that there was a significant relationship between being a woman and adopting COVID-19 preventive health behaviors.[22,32] A previous study aimed at the epidemiological study of gender differences in hand health knowledge and behaviors showed that women had a significantly higher knowledge score than men after adjusting for the effect of age and level of education.[33] A previous study of the H1N1 pandemic in Hong Kong among men and women found that women performed better than men in preventing the disease.[34]

The present study showed that the adoption of preventive behaviors against COVID-19 in people living in the capital of the province (i.e., Khorramabad) and Boroujerd was better than in the cities of Kuhdasht, Selseleh and Azna. According to the latest census in Iran in the year 2016, the cities of Khorramabad and Boroujerd have a higher percentage of urbanization than other cities.[35] A study conducted in the Golestan province showed that urban residents performed better against COVID-19 than villagers, which is probably due to differences in their literacy levels.[22] A systematic review and meta-analysis[36] has confirmed that there is a strong relationship between literacy and health literacy. The better adoption of COVID-19 preventive behaviors in the present study population may be due to their higher health literacy.

The present study showed that people with underlying diseases are more likely to adopt behaviors to prevent COVID-19 than those who do not suffer from such diseases. The results of a meta-analytic study of patients with underlying diseases who were hospitalized for COVID-19 showed that they were more likely to adhere to COVID-19 preventive behaviors.[37] These people feel more at risk for COVID-19 due to their underlying medical conditions.

Our study had several strengths. Its strengths included the use of a relatively large sample size and the use of the SEM model for statistical data analysis. In addition, despite quarantine conditions and traffic restrictions during the COVID-19 pandemic, the questionnaires were administered through telephone interviews by health-care providers so that all literate and illiterate people, both urban and rural, were on equal footing in answering the questions, while other studies conducted in Iran during the pandemic were conducted online,[21,22] in which people living in villages in Iran due to lack of access to the Internet and also illiterate people due to the inability to work with smartphones were not able to participate. The most important limitation of the study was that a larger number of urban/rural comprehensive health centers should be used, but due to the problems related to the coronavirus pandemic and the need to increase the sampling speed, a smaller number of centers were used. Another limitation of the study was the self-reported nature of the questionnaire.

Consistent with previous studies,[21,22] the present study showed that perceived self-efficacy, perceived barriers, and perceived benefits were the main determinants of COVID-19 preventive behaviors in individuals. Designing HBM-based counseling programs can improve motivation, skills, and behaviors and lead to improved beliefs and adherence to behavior. Other health determinants such as underlying illness, place of residence, and being a woman should be considered by policymakers.

Financial support and sponsorship

This study was funded by the Lorestan University of Medical Sciences as a research project.

Conflicts of interest

There are no conflicts of interest.


The researchers would like to express their gratitude to the participants and the staff of the health centers of Lorestan, Iran.


1. Cucinotta D, Vanelli M. WHO declares COVID-19 a pandemic Acta Biomed. 2020;91:157–60
2. Alsahafi AJ, Cheng AC. Knowledge, attitudes and behaviours of healthcare workers in the Kingdom of Saudi Arabia to MERS coronavirus and other emerging infectious diseases Int J Environ Res Public Health. 2016;13:1214
3. Alfahan A, Alhabib S, Abdulmajeed I, Rahman S, Bamuhair S. In the era of corona virus: Health care professionals' knowledge, attitudes, and practice of hand hygiene in Saudi primary care centers: A cross-sectional study J Community Hosp Intern Med Perspect. 2016;6:32151
4. Araban M, Karimy M, Mesri M, Rouhani M, Armoon B, Koohestani HR, et al The COVID-19 pandemic: Public knowledge, attitudes and practices in a central of Iran J Educ Community Health. 2021;8:35–40
5. Karimy M, Bastami F, Sharifat R, Heydarabadi AB, Hatamzadeh N, Pakpour AH, et al Factors related to preventive COVID-19 behaviors using health belief model among general population: A cross-sectional study in Iran BMC Public Health. 2021;21:1934
6. Glanz K, Rimer BK, Viswanath K Health behavior and health education: theory, research, and practice. 2008 John Wiley & Sons
7. Tadesse T, Alemu T, Amogne G, Endazenaw G, Mamo E. Predictors of coronavirus disease 2019 (COVID-19) prevention practices using health belief model among employees in Addis Ababa, Ethiopia, 2020 Infect Drug Resist. 2020;13:3751–61
8. Shewasinad Yehualashet S, Asefa KK, Mekonnen AG, Gemeda BN, Shiferaw WS, Aynalem YA, et al Predictors of adherence to COVID-19 prevention measure among communities in North Shoa Zone, Ethiopia based on health belief model: A cross-sectional study PLoS One. 2021;16:e0246006
9. Champion V, Skinner CGlanz K, Rimer BK, Viswanath K. The health belief model Health Behavior and Health Education: Theory, Research, and Practice. 2008 San Francisco Jossey-Bass
10. Larki A, Tahmasebi R, Reisi M. Factors predicting self-care behaviors among low health literacy hypertensive patients based on health belief model in Bushehr District, South of Iran Int J Hypertens. 2018;2018:1–10 9752736
11. Diddana TZ, Kelkay GN, Dola AN, Sadore AA. Effect of nutrition education based on health belief model on nutritional knowledge and dietary practice of pregnant women in Dessie Town, Northeast Ethiopia: A cluster randomized control trial J Nutr Metab. 2018;2018:6731815
12. Gharouni K, Ardalan A, Araban M, Ebrahimzadeh F, Bakhtiar K, Almasian M, et al Application of Freire's adult education model in modifying the psychological constructs of health belief model in self-medication behaviors of older adults: A randomized controlled trial BMC Public Health. 2020;20:1350
13. Bakhtiar K, Gharouni K, Gharouni B, Alavijeh FZ, Almasian M, Bakhtiar M, et al The effect of training interventions on the psychological factors of oral health in pregnant women Electron Physician. 2017;9:5506–15
14. Bastami F, Mostafavi F, Hassanzadeh A. Effect of educational intervention on knowledge, perceived benefits, barriers and self-efficacy regarding AIDS preventive behaviors among drug addicts J Educ Health Promot. 2015;4:90
15. Barmar S, Alimohammadian M, Sadjadi SA, Poustchi H, Hosseini SM, Yasseri M. Generalized structural equation modeling (GSEM) and its application in health researches J Sch Public Health Inst Public Health Res. 2018;16:51–62
16. Alavi M. Structural equation modeling (SEM) in health sciences education researches: An overview of the method and its application Iran J Med Educ. 2013;13:519–30
17. Iacobucci D. Structural equations modeling: Fit indices, sample size, and advanced topics J Consum Psychol. 2010;20:90–8
18. Wolf EJ, Harrington KM, Clark SL, Miller MW. Sample size requirements for structural equation models: An evaluation of power, bias, and solution propriety Educ Psychol Meas. 2013;76:913–34
19. Sharma S Applied Multivariate Techniques. 1996 New York John Wiley and Sons Inc
20. Khazaee-Pool M, Shahrvsand S, Naghibi SA. Predicting COVID-19 preventive behaviors based on health belief model: An internet-based study in Mazandaran province, Iran J Mazandaran Univ Med Sci. 2020;30:56–66
21. Mirzaei A, Kazembeigi F, Kakaei H, Jalilian M, Mazloomi S, Nourmoradi H. Application of health belief model to predict COVID-19-preventive behaviors among a sample of Iranian adult population J Educ Health Promot. 2021;10:69
22. Shahnazi H, Ahmadi-Livani M, Pahlavanzadeh B, Rajabi A, Hamrah MS, Charkazi A. Assessing preventive health behaviors from COVID-19: A cross sectional study with health belief model in Golestan province, Northern of Iran Infect Dis Poverty. 2020;9:157
23. Sharifirad G, Yarmohammadi P, Morowati SM, Rahayi Z. The Status of Preventive Behaviors regarding Influenza (A) H1N1 Pandemic Based on Protection Motivation Theory among Female High School Students in Isfahan in Iran Journal of Health System Research. 2011;7:108–17
24. Keyvanara M, Shaarbafchizadeh N, Jangi M, Samouei R. Evaluating self-care barriers in prevention of COVID-19 according to healthcare experts and laypersons: A mixed study J Mazandaran Univ Med Sci. 2020;30:117–25
25. Rezaeipandari H, Mirkhalili SM, Morowati Sharifabad MA, Ayatollahi J, Fallahzadeh H. Investigation of predictors of preventive behaviors of influenza A (H1N1) based on health belief model among people of Jiroft City, (Iran) Qom University Of Medical Sciences Journal. 2018;12:76–86
26. Kanadiya MK, Sallar AM. Preventive behaviors, beliefs, and anxieties in relation to the swine flu outbreak among college students aged 18-24 years Z Gesundh Wiss. 2011;19:139–45
27. Najimi A, Golshiri P. Knowledge, beliefs and preventive behaviors regarding Influenza A in students: A test of the health belief model J Educ Health Promot. 2013;2:23
28. Cyr A, Dunnagan TA, Haynes G. Efficacy of the health belief model for predicting intention to pursue genetic testing for colorectal cancer J Genet Couns. 2010;19:174–86
29. Wang D, Hu B, Hu C, Zhu F, Liu X, Zhang J, et al Clinical characteristics of 138 hospitalized patients with 2019 novel coronavirus-infected pneumonia in Wuhan, China JAMA. 2020;323:1061–9
30. Seale H, McLaws ML, Heywood AE, Ward KF, Lowbridge CP, Van D, et al The community's attitude towards swine flu and pandemic influenza Med J Aust. 2009;191:267–9
31. Lau JT, Griffiths S, Choi KC, Tsui HY. Widespread public misconception in the early phase of the H1N1 influenza epidemic J Infect. 2009;59:122–7
32. Baghernezhad Hesary F, Salehiniya H, Miri M, Moodi M. Investigating preventive behaviors toward COVID-19 among Iranian people Front Public Health. 2021;9:590105
33. Suen LK, So ZY, Yeung SK, Lo KY, Lam SC. Epidemiological investigation on hand hygiene knowledge and behaviour: A cross-sectional study on gender disparity BMC Public Health. 2019;19:401
34. Lau JT, Griffiths S, Choi KC, Tsui HY. Avoidance behaviors and negative psychological responses in the general population in the initial stage of the H1N1 pandemic in Hong Kong BMC Infect Dis. 2010;10:139
35. SCo I National Population and Housing Census. 2016Last accessed on 2021 Aug 16 Available from:
36. Zheng M, Jin H, Shi N, Duan C, Wang D, Yu X, et al The relationship between health literacy and quality of life: A systematic review and meta-analysis Health Qual Life Outcomes. 2018;16:201
37. Gebreheat G, Paterson R, Mulugeta H, Teame H. Adherence to COVID-19 preventive measures and associated factors in Ethiopia: A systematic review and meta-analysis PLoS One. 2022;17:e0275320

COVID-19; health belief model; Iran; preventive behaviors; structural equation modeling

Supplemental Digital Content

© 2022 WHO South-East Asia Journal of Public Health | Published by Wolters Kluwer – Medknow