Effect of Social Beliefs on Consumption of Dairy Products and Its Predicting Factors Based on the Transtheoretical Model: A Population-Based Study

Background Social beliefs on the consumption of dairy products are associated with health conditions, and the aim of this study is to investigate associated factors with the rate of dairy product intake, in accordance with social health-related beliefs and the elements predicting dairy consumption, based on the transtheoretical Model (TTM). Methods 981 subjects (chosen from Mashhad citizens, Iran) were surveyed in random public places in 2014, using demographic surveys and questionnaires based on TTM and advantage/disadvantage by trained interviewers. Results 981 Subjects with a mean age of 30.39 ± 14.83 were surveyed in dairy nonconsumer and dairy consumer groups. There was a significant relationship between dairy consumption and gender (P < 0.001). Factors such as age, educational level, job status, and opium addiction were found to be significantly associated with dairy consumption status. Young and female subjects consume more dairy products than their older and male counterparts, respectively. People with a diploma degree and lower levels of education consumed substantially more dairy products than their educated equals. Unemployed participants consumed considerably more dairy products than their fellow employed participants. Opium-addicted subjects were more likely to avoid dairy products. Conclusions Despite the general belief of dairy consumption being beneficial, subjects in the precontemplation stage as nonconsumers described dairy products as of poor taste having low diversity in markets. Also, among the reasons, dairies' short shelf-life and behaviours under the influence of society and family were the mains. The termination stage's subjects as consumers consumed dairy products mostly for losing weight.


Introduction
1.1. Background. Dairy products, including milk, are considered as rich nutrients. Despite their low-level calories, these products supply the elements required for the normal function of the organs [1][2][3], such as water, fat, necessary proteins, calcium, phosphorus, magnesium, potassium, zinc, selenium, and group of vitamins (A, E, K, B2, B5, and B12). As a preventive factor, these products lower the risk of diabetes, stroke, hypertension [4], bone fracture [5], cardiovascular diseases, and obesity [6]. According to the annual reports of the International Dairy Federation (IDF), the average global consumption of milk has risen gradually in the last decade (from 101.4 kg/capita/year in 2005 to 113 kg/capita/year in 2017) [7,8]. Due to the latest reports of the Food and Agriculture Organization (FAO), low-fat and whole milk consumption in Iran is about 64.38 kg/capita/ year [9], which is far lower than the global average.
Demographic and socioeconomic factors, such as price, availability, awareness, and convenience could afect dietary behavior [10]. Although, dairy industry in Iran is advance and also, During 1996-2001, dairy products involved just 2.5 percent of total Iranian family expenditure [11] but recently due to economic sanctions the dairy price increased as the family income decrease . In addition to, poor awareness of food values among people and lack of promotions of companies in Iranian society could lead to low dairy consumption [12]. In addition, if dairy products are not prepared under hygienic conditions, then it will lead to infection disease [13].
As the consumption of dairy products in Iran is lower than the world average, it is necessary to fnd predictors and related factors associated with this issue. Unlike Iran, there is promising evidence on attraction factors for dairy consumption [14].
Te transtheoretical model (TTM) is an intentional method emerged from comparative analyses of approximately 300 psychotherapy theories [15]. Tis model is used for categorizing people into diferent stages towards specifc health behaviors [16] and developing correct behaviors in each person individually according to his/her conditions [17].
According to controversies and lack of efcient studies on dairy products consumption, we decided to assess the related factors of dairy consumption and beliefs associated with the population's behavioral conditions based on TTM in Mashhad, Iran.

Study Design, Participants, and Data Collection.
In 2014, 981 people were surveyed in our cross-sectional study in order to assess the local population's dairy product consumption and its predictors in Mashhad (second largest city of the country), Iran [18]. Te multistage selection method was administered to select the participants. At the frst step, we made a list of public sites such as parking lots, banks, hospitals, and universities belonging to municipal categorizations and urban maps all around the city; then, we assigned numbers to each site, and after that, we selected random numbers from the list. Ten, at the second step, in each place, subjects were chosen by the random convenience method and interviewed by a trained interviewer. Te criteria for selecting the subjects were literacy, being socially active, consent to participate in this study, and citizenship of Mashhad. Subjects consuming more than 3 units of dairy products were defned as consumers. Each unit of dairy food was defned as a glass of milk, a cup of yogurt, a cup of curd, a glass of ice cream, a matchbox of cheese or two glasses of "dough," and a local salty beverage similar to soured milk. All the measures were drawn from the previous studies [19,20].

Questionnaires.
We designed a questionnaire consisting of two main sections. In the frst section, due to the 6 stages of TTM, 6 yes/no questions were designed. Someone in the precontemplation stage had no tendency to consume dairy products within the last 6 months. In addition, subjects in the contemplation stage have not yet consumed dairy products but are planning to consume them in the following 6 months. Furthermore, participants in the preparation stage have taken some measures to modify their lifestyle and are planning to start dairy consumption within the next month. Participants starting dairy consumption in the last 6 months were classifed in the action stage. Te maintenance stage comprised people beginning consumption for 6 months or more. Finally, subjects in the termination stage do not have any intention to return to their previous unhealthy habits. Te second section consisted of questions evaluating the subjects' beliefs on the advantages, disadvantages, and barriers to consuming recommended dairy products on a 5-point Likert scale. Validating the questionnaire was performed with expert council which included 6 experienced researchers [ Table 1]. Te internal consistency for total and subscales was calculated by Cronbach's alpha coefcient, and reliability between the test and retest was performed using the intraclass correlation coefcient (ICC). Te total ICC was 0.90, and Cronbach's alpha was 0.65-0.74. We also added a sociodemographic checklist including age, gender, height, weight, job status, family income, educational level, marital status, household size, and exercise habits.

Ethics.
Interviewers informed all people participating in this study of the aim of the research and ensured them of the confdentiality of their personal data. After oral consent, interviewers flled the questionnaire during a face-to-face conversation. Te study was approved by the Ethics Committee of the Mashhad University of Medical Sciences (ethical approval code: 930054).

Statistical Analysis.
Te 16th version of the SPSS software (SPSS Inc., Chicago, Illinois, USA) was used for all statistical analyses. Data normalizations were checked by the Kolmogorov-Smirnov test. Te chi-square test was used to examine the signifcance between categorical data. Te Mann-Whitney test was applied for nonparametric data analyses. In order to eliminate the efects of confounding variables, logistic regression analysis was used in assessing pros/cons of items among the subjects of precontemplation and termination stages. All tests were 2tailed, and probability values below 0.05 were deemed signifcant.

Dairy Consumption Status.
Te transtheoretical Model for dairy product consumption indicated that the largest groups of men and women were at the termination stage.
Moreover, the results demonstrated that in every 4 female, about 3 subjects (70.00%) consumed dairy products, whereas in males, some two thirds (61.06%) of the subjects consumed dairy products (Table 3).

Comparison between Dairy Consumers versus
Nonconsumers. As discussed before, the frst three stages (precontemplation, contemplation, and preparing) and the second three stages (action, maintenance, and termination) were categorized into dairy nonconsumer and dairy consumer groups, respectively. As shown in Table 4, the chi-square test showed a signifcant relationship between dairy consumption and gender (P < 0.001). Furthermore, we found factors such as age, educational level, job status, and opium addiction to be signifcantly associated with dairy consumption status. Young and female subjects consume more dairy products than their  Journal of Environmental and Public Health 3 older and male counterparts, respectively. People with a diploma degree and lower levels of education consumed substantially more dairy products than their educated equals. Unemployed participants consumed considerably more dairy products than their fellow employed participants. Last but not least, opium-addicted subjects were more likely to avoid dairy products. No meaningful relationship was found between the subjects' marital status, family size, monthly income, and BMI in the consumer and nonconsumer groups (Table 4).

Subjects' Beliefs of Dairy Foods.
Overall, our participants agreed with the benefts of consuming dairy products for their health. Tey also frmly believed in the consumption of dairy foods as a benefcial substitute for unhealthy diets. Te most important factors leading to the lack of motivation were, from highest to lowest, the presence of preservatives in dairy products, limited sustainability, and high prices. Table 1 indicates the mean scores given by people for each advantage/disadvantage. Te multiple regression test was used to recognize most related beliefs in precontemplation and termination stages. So, we defned two extremes of consuming stages precontemplation and termination as outcome and participants believes as predictors.
As shown in Table 5, in the precontemplation stage, 6 items and their related beliefs are signifcantly associated with participants. Also, the benefts of consuming dairy products for general health and weight loss persuaded several subjects to take dairy products permanently in the termination stage.

Discussion
Milk and dairy products contain various necessary nutrients, such as calcium, phosphorus, magnesium, zinc, vitamins [21], and proteins, which are important for immunity, prevention of tumour development, and iron absorption. A recent study on the Iranian society has demonstrated that the consumption of skimmed milk as a common food has decreased considerably, from 65.39 kg/capita/year in 2007 to 46.69 kg/capita/year in 2013 [9].

Journal of Environmental and Public Health
We aimed to evaluate the factors afecting dairy consumption and assess social beliefs of citizens of the second biggest city in Iran about dairy products.
Te participants' mean age was 30.39 ± 14.83, close to the Iranian population's mean age (31.10 years) which was unofcially declared by the Statistical Centre of Iran in 2016, indicating that in terms of age, this study could be a proper sample of Iranian people. Among participants, males are 3 times more employed than females, due to patriarchal circumstances in Iran. In addition, more than a half of male subjects have academic degree, even though 38.73% are graduated from college. Leaving college before graduation were because of males' tendency to acquire more comfortable, professional, and high-income jobs and form sustainable families in fnancial, behavioural, and social perspectives.
In a recent study on 56233 people with an age of 50 years and over, among 16 countries of Europe, confrms our results that women's higher knowledge about advantages of dairy products for health could lead to more dairy consumption [22]. In our study, females are signifcantly more motivated to consume dairy products than males, and overall, their percentage in the consumer group is higher than males. A study conducted by Green et al. found no signifcant relationship between gender and dairy consumption, even though they found a specifc relationship between gender and milk consumption [23]. Mirmiran et al. indicated that among 462 subjects of 16 years and over, men consume signifcantly more dairy products than women [24]. Other studies based on children and adolescents demonstrated similar results to those reported by Larson et al. [25,26]. We believe that the subjects' number, age range, and social conditions are responsible for these controversies.
Our results, similar to those of the study by Gopinath et al., imply an indirect relationship between subjects' age and dairy consumption [26]. According to educational level, in our study, subjects with diplomas and lower degrees were more likely to consume dairy foods. Among the Iranian populace, people of higher education, due to their higher ages, deal more difculties in their daily lives and spend a lot of time in working environments. Tus, they cannot pay enough attention to their lifestyles and adjust them correctly.
In this study, opium-addicted people signifcantly utilize lower dairy foods than nonabusers. Despite the low percentage of addicted subjects in this study, this outcome confrms previous results that unhealthy behaviors are associated with a clustering pattern, performed in another Iranian metropolis, Shiraz, Iran [27].
Our study found no signifcant relationship between the BMI and dairy consumption. Similarly, in the prospective cohort study by Gopinath et al., the participants' (school children) BMI was not signifcantly related to high dairy consumption during 5 years of subjects' adolescence [26]. In contrast to our results, a cross-sectional research study by Mirmiran et al. showed an indirect moderate correlation (r � −.36, P value<0.05) between dairy servings and BMI [24]. Furthermore, a prospective cohort study found a signifcant inverse association between primary school children's BMI and frequency of dairy consumption. In this topic, our study subjects' BMI was self-reported, and due to the diferences between study designs and subjects' age, the results could have also varied.
For evaluating the behavioral conditions of Mashhad citizens, we used multiple regression analysis, a very general system in the feld of behavior assessment [28]. Due to the subjects scoring each item, we used multiple regression tests to analyse data in order to fnd signifcant pros and cons of dairy product consumption. According to the results of this test, even though nonconsumer subjects without the intention of consuming dairy foods in the precontemplation stage believed in benefts of dairy products for their health and diet, they argued that common and available dairy products are of poor taste and have low diversity in markets. Also, they believe that dairy products' relatively short expiration date could be an issue. Last but not least, peer pressure and social and family impacts in the precontemplation stage are proven as important factors leading to an unhealthy diet. We found that their knowledge of dairy products' health advantages, such as efcacy to weight control, was signifcant factors for the participants consuming dairy products and satisfed with their dairy diet habit..
Tis study was administered on a large population randomly selected from public places. Tus, it could be highly good representative of our study population. Furthermore, we used the multiple regression test as a very general and useful test for analysing people's behavioral situations. To the best of our knowledge, this is the frst pro/ con study performed in the feld of dairy consumption which was analysed by the logistic regression test.

Journal of Environmental and Public Health
Te most important limitation of the current study was its study design since we did not follow up the subjects participating in this research. Moreover, all data collected by written surveys were self-reported, and hence, their measurable characteristics such as height and weight were not verifed by using standard instruments and methods.
As mentioned before, we found no previous studies assessing the efects of population beliefs on diet behaviors in consuming dairy foods. Terefore, we recommend future studies to evaluate this issue in order to fnd out factors alienating people from dairy products and solve them by purposeful advertisements and conscious decisions in dairy industries to convince people to include dairy products in their diet. In addition, we advise researchers to design large longitudinal studies determining predictors of dairy consumption, particularly due to controversies between previous researches.

Conclusions
Tis study demonstrated that young, female, unmarried, and unemployed people were signifcantly more dispersed in consumer groups and that opium-addicted subjects were more in the nonconsumer group. Moreover, despite the general belief of dairy consumption being benefcial, subjects in the pre-contemplation stage, as non-consumer, described dairy products as of poor taste and diversity in the markets. Also, among the reasons, dairies' short shelf-life and behaviours under the infuence of society and family were the mains. Te termination stage's subjects as consumers consumed dairy products mostly for losing weight.

Data Availability
Data are available on reasonable request from the corresponding author.

Conflicts of Interest
Te authors declare that they have no conficts of interest.