Study design and participants
The present national cross-sectional study is a part of the fifth survey of Childhood and Adolescence Surveillance and Prevention of Adult Non-communicable disease (CASPIAN)-V study in Iran (2014–2015). Details of the study protocol have been published earlier (17). Briefly, in the CASPIAN-V study, 14,400 students from both urban and rural regions (30 provinces) were selected using a multistage, stratified cluster sampling method in winter. In each province, considering the eligible criteria, sampling was conducted proportionally based on gender (boy, girl), the level of education (primary and secondary), and the place of residence (urban or rural).
Only children and adolescents who met the following criteria were included: (i) primary and secondary students, (ii) having Iranian nationality, and (iii) children and adolescents without any chronic diseases.
In each province, considering the eligible criteria sampling was conducted based on gender (boy, girl), the level of education (primary and secondary), and the place of residence (urban or rural).
For biochemical measurements, 14 clusters (each cluster included 10 subjects) in each province were randomly chosen for biochemical analyses, and for this study, 2594 blood samples were selected for the assessment of serum Vitamin D concentrations.
Notably, after explaining the study procedures and its objectives, verbal and written informed consent was obtained from students and their parents. The Ethics committee of Isfahan University of Medical Sciences approved the protocol of the current study (Project number: 194049).
Assessments
In the present study, the information about students' basal characteristics, socio-economic status, screen time, anthropometric indices, physical activity, as well as SHC were collected.
Students' basal characteristics
We used the questionnaire provided by the World Health Organization-Global School Student Health Survey (WHO-GSHS) to assess health behaviors and protective parameters related to the leading causes of morbidity and mortality of children and adolescents (18). The validity and reliability of the Farsi format of the questionnaire were acceptable, as was identified earlier (Cronbach’s alpha coefficient: 0.97, Pearson’s correlation coefficient: 0.94) (19).
Socio-economic status
To estimate the SES of each student, five main parameters including parent’s education and occupation, type of children’s school (public, private), home-ownership (yes, no), and family assets (having a personal computer, vehicle ownership) were considered. Finally, SES score was calculated as the weighted average of the mentioned items. Students were then classified into three groups (low, medium, and high SES).
Screen time (ST)
To examine ST, participants were asked regarding the average time (hour/day) dedicated to watching TV and playing video games. After the calculation of total ST, students were classified into two groups (low and high). Those with less or equal to 2 hours per day were placed in low; otherwise they were considered as individuals with high ST.
Physical activity
A 7-day self-administrative physical activity questionnaire (PAQ-A) was used to examine the levels of physical activity. The PAQ-A is a valid and reliable questionnaire based on the information obtained previously (Cronbach’s alpha coefficient: 0.97, Pearson’s correlation coefficient: 0.94) (20).
Anthropometric measurements
Body weight and the height of students were measured with standard methods by a trained health care member of the team as described before (17). Body mass index (BMI) was calculated by dividing weight (kg) to the square of height (m).
SHC
To assess SHC, a valid questionnaire designed in the Health Behavior in School-aged Children (HBSC) study was applied (21). Using a face-to-face interview with students, the questionnaire was fulfilled. Participants were asked if they had experienced either any psychological (feeling low, feeling nervous, difficulty in getting sleep, and irritability) or somatic symptoms (stomach ache, headache, backache, feeling dizzy) in the last six months before the study. The frequency of each sign was asked, as well.
Response options for each symptom were as follows: (i) about every day, (ii) more than once a week, (iii) about every week, (iv) about every month, (v) and (vi) rarely or never. Finally, the responses were categorized as “weekly or more” and “rarely or never.”
Biochemical assessments
To assess the serum levels of 1,25 (OH) D3, 6 mL of venous blood samples was taken from students. All samples were stored at -70°C until biochemical analysis.
To measure the serum concentration of 25-hydroxy vitamin D,a direct competitive immunoassay chemiluminescent method with LIASON 25-OH vitamin D assay TOTAL (DiaSorin, Inc.) was used(coefficient of variation: 9.8%).
Serum levels of vitamin D lower than 10 ng/mL (deficient), between 10 and 30 ng/mL (insufficient), and greater than 30 ng/mL (sufficient) were considered as deficient, insufficient, and sufficient, respectively.
Statistical analysis
Qualitative and quantitative variables were expressed as number (percent) and Mean± Standard Deviation, respectively. Characteristics of students (age, location, SES, physical activity, ST) and vitamin D levels in each gender were provided, separately. The frequency of each health compliant as well as multiple complaints considering genders was reported in number and percentage. To assess the associations between vitamin D status and SHC, students were classified into three groups based on vitamin D concentrations (sufficient, insufficient, and deficient). The link between vitamin D status and SHC (8 health complaints) were examined across the mentioned three categories. In addition, the links that experienced multiple complaints were reported as well. Subjects with the sufficient levels of vitamin D3 (30- 50 ng/mL) were considered as a reference group, and the two remaining groups (deficient, insufficient categories) were compared with this one for each SHC. Logistic regression models, both crude and adjusted models (3 models), were applied to examine any links. Covariates such as age, sex, region, SES, physical activity, ST, and BMI were adjusted in the mentioned adjusted models. Model 4 was controlled for all the variables above. Statistical analysis was performed using STATA software version 11. P-values less than 0.05 were considered as statistically significant.