Predictors of COVID-19 preventive behaviors based on the health belief model among pregnant women


 Background

COVID-19 is an infectious disease caused by the new coronavirus, and its widespread epidemic has caused many deaths and health, social, and economic consequences in the world. The purpose of this study was to explain self-care behaviors against COVID-19 based on the health belief model (HBM) in pregnant women.
Method

 The present study was conducted using a descriptive-analytical approach with the participation of 230 pregnant women referred to health centers in Urmia in 2021. The participants were selected using multi-stage random sampling. The data were collected using a valid and reliable questionnaire including participants’ demographic characteristics, the HBM items, and items relating to self-care behaviors against COVID-19. The data were analyzed using descriptive and inferential statistical methods (frequency, mean, standard deviation, Pearson correlation coefficient, and linear regression) in SPSS software (version 25).
Results

The results of this study showed that the rate of self-care behaviors against COVID-19 in the pregnant women participating in the present study was not very favorable. It was also shown that among the constructs of the HBM, awareness, self-efficacy, perceived barriers, and perceived severity were the most important predictors of self-care behaviors with a variance of 24% change among the pregnant women.
Conclusion

Awareness, self-efficacy, perceived barriers, and perceived severity were found in this study as the strongest predictors of self-care behaviors among pregnant women. The results of the study can be useful in planning educational and behavioral interventions to increase the adoption of self-care behaviors against COVID-19 and institutionalize health-promoting behaviors in women.

Introduction COVID-19 is caused by a new type of coronavirus that was rst identi ed in December 2019 in Wuhan, China. It has become a pandemic with far-reaching economic, social, and health consequences (1). One of the most important aspects of Covid-19 is its rapid spread through airborne droplets and surfaces and objects contaminated with these droplets, which indicates the need for greater personal and social hygiene (2). pregnant women makes them more vulnerable to COVID-19. They experience immunological and physiological changes and may be more prone to viral respiratory infections (7). Caring for these pregnant women can be complicated and di cult because the infection may adversely affect the mother, fetus, and baby (8).
Some speci c COVID-19 management advice for pregnant women is offered based on their previous experiences with previous coronaviruses. To prevent COVID-19 in pregnant women, they are recommended to avoid unnecessary travel, attending crowded places, using public transportation, and contacting suspicious or sick people. They also need to take personal and social health care and inform their doctor immediately if they experience early symptoms such as fever, cough, fatigue, muscle aches, sore throat, and shortness of breath (9). Some researchers have reported the risk of preterm delivery, premature rupture of membranes, fetal tachycardia, and fetal distress during the third trimester of pregnancy among women infected with coronavirus (10).
Many doctors are also afraid of congenital infection with the virus and decide to terminate the pregnancy, which imposes many psychological and physical complications on pregnant women (9).
Currently, effective training and strategies have been adopted to prevent this disease. However, we are witnessing an increase in the number of patients with this disease every day. Therefore, it seems that pure information is not enough and there should be a fundamental behavior (11). Planning and preparedness to deal with the COVID-19 crisis is one of the national and international necessities, and taking preventive measures at the community level to control the COVID-19 epidemics should be highly considered by policymakers and health o cials (12). Preventive measures such as education, awareness promotion, and fostering prevention skills can be promoted through diseases prevention and control programs (13).
The use of health education model will help researchers recognize factors affecting health behaviors. The health belief model is one of the most effective health education models developed based on cognitive psychology and is used to determine health behaviors. It also shows the relationship between health beliefs and health behaviors and emphasizes the intrinsic factors of individuals, such as knowledge, attitudes, beliefs, and behavior (14).
This model encompasses four main concepts, including sensitivity, severity, bene ts, and perceived barriers. According to the constructs of this model, to adopt preventive behaviors, people must rst feel the danger (perceived sensitivity), then understand the depth and severity of the danger and the effects it may have on them and society. People should also believe in the effects and bene ts of preventative measures and accept that taking preventive behaviors will cost less than getting the disease (13)(14)(15)(16).
Due to the novelty and sudden prevalence of the COVID-19 disease and the lack of studies on predictors of preventive behaviors against this disease based on the HBM, this study aimed to determine the predictors of adopting preventive behaviors against COVID-19 based on the HBM constructs among the pregnant women.

Methods
This descriptive-analytical cross-sectional study was performed on 230 pregnant women who referred to Urmia health centers. The sample size was estimated using the following equation (17): Where, p is the number of variables in the regression model, R 2adj is the adjustment coe cient, and δ is the difference between the coe cient of determination and the adjusted coe cient of determination.
Due to the lack of a similar study in this eld, preventive measures against the Zika virus in Thailand were used to estimate the value of the adapted coe cient of determination, and δ showing the difference between the coe cient of determination and the coe cient of determination was estimated from a previous study by Siramanrate (18). Thus using the above equation (p = 11; R 2 = 0.307; δ = 0.048), the required sample size was estimated as 159 people. Given the concern of people to go to clinics during the COVID-19 outbreak and considering the 30% probability of not participating in the study, the sample size increased to 226 people.
The participants were selected using multi-stage sampling. First, all 66 health care centers in Urmia were listed based on the socioeconomic and cultural conditions of the regions hosting the centers. To this end, 26 health centers were classi ed at the high level, 20 health centers at the moderate level, and 20 health centers at a low level based on socioeconomic status and cultural conditions. In this study, 12 health centers in Urmia at three high, moderate, and low levels in terms of socioeconomic and cultural conditions were selected randomly. Then, according to the population covered by each of the socio-economic levels of those health centers, pregnant women were randomly selected as candidates with equal chances to enter the study.
The inclusion criteria were: 1) Being a resident of Urmia, 2) Pregnant mothers covered by health centers, 3) Literacy for reading and writing, 4) Willingness to participate in the study, and 5) Lack of a history of mental illness. The women who were unwilling to participate in the study were excluded. The rst section of the questionnaire assessed the participants' socio-demographic information including their age, their number of pregnancies, job, and insurance status.
The second section of the questionnaire contained 22 items that assessed the participants' knowledge towards COVID-19 (e.g. COVID-19 can be asymptomatic). The responses were measured using the nominal scale of "True" and "False". One point was given for a correct answer and zero for an incorrect or no answer. The total score for each respondent varied from 0 to 22.
The third section consisted of items about the various constructs of the health belief model to perform self-care behaviors to prevent COVID-19: Perceived sensitivity (7 items): "If I do not wear a mask outdoors, I may become infected with the Coronavirus", perceived severity (7 items): "Mortality rate due to covid-19 is high", perceived bene ts (7 items): "Washing my hands regularly prevents me from becoming infected with the coronavirus", Perceived barriers (6 items): "Masks are expensive for me", Self-e cacy (6 items): "I am con dent in my ability to disinfect my environment", and Cues to action (10 items): "Have you read about the importance and necessity of disinfecting food and the environment through virtual space?" All constructs of the health belief model, except the cuse to action, were scored on a ve-point Likert scale (ranging from 1='strongly disagree' to 5='strongly agree'). However, the items on the Cues to action were scored as either zero or one. The responses were measured using the nominal scale of "Yes" and "No". One point was given for a "Yes" answer and zero for a "No" answer. The total score for each respondent varied from 0 to 10.
The fourth section consisted of 9 items about self-care behavior: "I observe a distance of at least 1.5 meters in dealing with others". The items were scored on a 4-point Likert scale (ranging from 0= 'Never' to 4='Always').
The participants' answers to all items were categorized as weak practice (if the participant scored < the mean score), moderate practice (if the participant scored 50-75% of the mean score), and good practice (if the participant scored > 75% of the mean score).
It should be noted that at the end, each person's score for all constructs was estimated based on a 100 score. Then the distance between the lowest and highest scores was conventionally classi ed into three categories: weak (0-50), medium (51-75), and good (76-100).
The content and face validity of the questionnaire were reviewed and veri ed by 10 health education and reproductive health professionals. To this end, the experts were asked to review and assess the items in terms of simplicity, clarity, relevance, and necessity. The items were then revised based on the feedback provided by the experts and the validity of the tools was con rmed.
In the psychometric analysis of the instrument, the content validity ratio (CVR) and the content validity index (CVI) of the instruments were reported as 0.86 and 0.90, respectively. Cronbach's alpha coe cient was estimated to assess the internal consistency of the instrument in a pilot study of 30 pregnant women who were similar to the participants in the research sample in terms of demographic characteristics. Cronbach's alpha for perceived sensitivity, perceived severity, perceived bene ts, perceived barriers, self-e cacy, Cues to Action, and self-care behaviors were estimated as 0.87, 0.7, 0.76, 0.71, 0.80, and 0.92, con rming the reliability of the tools was at the acceptable level.
Data analysis was performed in SPSS (version 22) using Pearson's correlation coe cient and linear regression analysis. Moreover, P-values less than 0.05 were considered statistically signi cant.

Results
The participants in this study were 230 pregnant women with a mean age of 29.57 ± 5.82. Moreover, 214 women were housewives and 16 women were employed. Table 1 summarizes the participants' demographic characteristics.  Pearson correlation coe cient showed a positive and signi cant relationship between the constructs of the health belief model and self-care behaviors related to prevention of COVID-19 except perceived bene ts. A negative correlation was found between perceived barriers and self-care behaviors related to prevention of COVID-19. Other correlation coe cients of the HBM constructs and individual factors in relation to self-care behaviors are listed in Table 3. To determine the predictive power of adopting self-care behaviors in confronting COVID-19, regression analysis and stepwise method were used. The regression analysis assessed the constructs of the HBM and some individual factors.
Based on the results, awareness, self-e cacy, perceived barriers, and perceived severity were determined as the nal predictors of changes in self-care behaviors. In total, these variables were able to explain about 24% (R2 = 0.235) of the variances in self-care behaviors related to COVID-19 prevention (Table 4).

Discussion
This study aimed to investigate the predictive factors associated with the adoption of self-care behaviors for preventing COVID-19 disease in pregnant women using the HBM. The results of this study con rmed a statistical model consisting of four variables including awareness, self-e cacy, perceived barriers, and perceived severity, which were able to account for about 24% of self-care behaviors associated with COVID-19 disease. Of course, these four variables varied in terms of their power to explain the adoption of self-care behaviors. Similarly, Khazaei Poor (13) reported that the constructs of the health belief model could explain 26% of self-care behaviors effective in the prevention of COVID-19.
The results of this study showed that the average score of adopting self-care behaviors in preventing COVID-19 was about 64, showing that self-care behaviors taken by the pregnant women to prevent COVID-19 disease were not desirable due to the high prevalence of the disease in Iran.
However, some studies (19)(20)(21) estimated preventive function of different study groups as desirable. Given the speci c physiological conditions of pregnant women, they need to pay attention to the plans and strategies adopted to care for COVID-19 disease. Considering the possible bene cial effects of adopting self-care behaviors in health-promoting and reducing the risk of the COVID-19 epidemic, it is necessary to adopt an appropriate strategy to remove possible obstacles to managing this global epidemic.
To increase the effectiveness of interventions to promote the adoption of self-care behaviors against COVID-19 in pregnant women, it is necessary to identify the determinants of preventive and self-care behaviors in this population.
Accordingly, the present study showed awareness, self-e cacy, perceived sensitivity, perceived severity, and the Cues to Action were positively correlated with the adoption of self-care behaviors. However, this correlation was reversed between perceived barriers and self-care behaviors, indicating that those with high perceived barriers are less likely to engage in self-care behaviors to prevent COVID-19. This nding is consistent with the results of previous studies (3,22). There was no correlation between individual factors such as age, employment, and education of pregnant women with self-care behaviors.
In the present study, knowledge about the nature of coronavirus and self-care in confronting COVID-19 disease was the strongest and most effective variable, which could predict the relationship between self-care in pregnant women alone and with a variance of about 12%. It seems that due to the high prevalence of COVID-19 disease and also due to the widespread information by the mass media, radio and television channels, the Ministry of Health and the distribution of warning banners throughout the areas affected by the disease and public awareness, people are more likely to do optimal self-care behaviors to prevent COVID-19 disease (13).
The self-e cacy as the second construct of the HBM could explain the adoption of self-care behaviors in confronting COVID-19. Perceived self-e cacy had a positive and direct effect on preventive behaviors against COVID-19. The more people feel motivated, capable, and hoping to succeed in ghting the coronavirus, they show more willingness to do individual health behaviors (23). In the present study, self-e cacy was conceptualized to the extent to which a person feels that they can use self-care and preventive strategies to combat the coronavirus.
Perceived barriers construct as the third construct of the HBM was able to account for the adoption of self-care behavior. Perceived barriers represent the most important construct of the HBM with the high predictive power of behavior (24). It should be noted that the low perceived barriers are a privilege because the individuals believe that they face fewer obstacles and fewer problems in adopting self-care and preventive behaviors. Besides, the objective and psychological costs of the recommended activities are low or they are preferable due to the bene ts of the behavior. Therefore, it can be suggested that by performing a series of interventions and adopting effective policies to reduce barriers as much as possible, the possibility of adopting self-care behaviors for preventing COVID-19 increases.
Perceived severity as the fourth construct of the HBM could predict the adoption of self-care behaviors in confronting COVID-19. Perceived severity is ultimately related to the perceived threat and refers to the extent to which individuals perceive the danger and seriousness of the spread of coronavirus. This nding is consistent with the results of previous studies by Bates et al. (25), Khazaee-Pool et al. (13), and Didarloo et (15).
Although the present study showed a positive and signi cant correlation between perceived sensitivity, perceived bene ts, and Cues to Action in the adoption of self-care, these constructs in the model extracted from regression analysis were not con rmed as predictive constructs. In fact, if people are aware of the bene ts of taking preventative behavior, they are better prepared to take activities and they more likely accept the mentioned behaviors. In other words, if a person believes that home quarantine and the use of personal protective equipment can reduce the risk of developing the disease or transmitting it to others, or have social bene ts such as reducing treatment costs or potential costs to the health system of the country, they will be more likely to do self-care behaviors (26, 27).
Cues to Action can act as a stimulus received by people from the COVID-19 outbreak. This stimulus can be in the form of a clip, news of the death of people due to COVID-19, the effects and consequences of this disease on people's lives posted on mass media and social networks, or the advice and warnings of the staff of health centers (27,28). The majority of the women in this study reported that the most important sources of information about COVID-19 were health workers and warnings from the Ministry of Health through the mass media.
Given the importance of mass media, especially in the period of growth of new technologies and virtual networks, and given the importance of non-aggregation to reduce the transfer of COVID-19, the high potentials of social media can be used for educational, awareness-raising, and behavior change purposes.
Of course, it should be noted that pregnant women's engagement in self-care can be affected by some non-behavioral environmental factors such as the availability of facilities and preventive devices, high cost of protective equipment and disinfectants, and strategies adopted nationwide, etc. By informing the community and o cials and adopting useful policies and effective interventions, it is possible to increase the adoption of self-care and preventive behaviors against COVID-19. One of the limitations of this study was the use of a self-report questionnaire. Thus, by stating the objectives of the study and assuring the pregnant women that their information is con dential, an attempt was made to reduce this limitation.

Conclusion
Awareness, self-e cacy, perceived barriers, and perceived severity were found in this study as the strongest predictors of self-care behaviors among pregnant women. The results of the study can be useful in planning educational and behavioral interventions to increase the adoption of self-care behaviors against COVID-19 and institutionalize healthpromoting behaviors in women.

Declarations
Funding Urmia University of Medical Sciences Availability of data and materials The datasets generated during and/or analyzed during the current study arenot publicly available due to con dentiality of data and subsequent research,but are available from the corresponding author on reasonable request.

Ethics approval and consent to participate
Research has been presented in the ethics committee of Urmia University of Medical Sciences and has received the code of ethics (IR.UMSU.REC.1399.142). informed consent was obtained from all participants in this study, and all provisions of the Helsinki Statement on Research Ethics were considered.

Consent for publication
Not applicable.