Study sample and data collection
Oman is one of the six Middle Eastern countries in the Gulf Cooperation Council (GCC). While the region is known for economic prosperity and a high GDP per capita, Oman is marked by its unique financial and cultural background among its counterparts. This small desert nation has a unique cultural heritage combined with a large immigrant population . Arabic is the official language and the predominant religion is Islam (85.9%). Children between the age of 0 to 14 make up the largest portion of the Omani population at 30.1%. The country has 4.8 million inhabitants, 45% of which are immigrants. While Oman is a high-income country, its economic background makes it markedly different from neighboring gulf countries. Oman’s Gross Domestic Product (GDP) per capita is $46,000, as compared to $124,100, $68,600, $54,500 in neighbouring Qatar, UAE, and Saudi Arabia respectively . Primary school enrollment is reported to be 100% in Oman, with a literacy rate of 96.1% at 15 years of age .
This is a secondary data analysis using data from the Global School Health Survey (GSHS) for Oman, collected in 2015 . The GSHS is a school-based survey administered to students enrolled in secondary school, with most students between 13 and 17 years of age, used primarily for obtaining self-reported data on health behaviors that affect major causes of morbidity and mortality among adolescents worldwide. The survey used two-stage cluster sampling design and included any school containing 8th to 12th grade classes. Ethical approval was obtained from the Research, Ethical Review and Approve Committee at the Ministry of Health in Oman. Parents and adolescents were sent a letter briefing them about the study and explaining that participation is voluntary.
The questionnaire consisted of 59 self-administered questions covering a number of topics: diet and exercise habits, drug use and exposure, hygiene, mental well-being, tobacco exposure and use, physical harm, violence and unintentional injury. From a total of 63 schools, 3468 students completed the survey and the response rate from schools and students was 98% and 94%, respectively. Responses from parents were not available from this questionnaire.
The following demographic variables were collected: age, sex and ‘how often did you go hungry in the past 30 days’. Hunger as a food security measure was used as a proxy for socioeconomic status. Of note, this was not the original intended purpose of the hunger question.
Independent variable: parental involvement
Parental involvement score utilized the four available questions regarding this topic in the database:
During the past 30 days, how often did your parents or guardians check to see if your homework was done?
During the past 30 days, how often did your parents or guardians understand your problems and worries?
During the past 30 days, how often did your parents or guardians really know what you were doing with your free time?
During the past 30 days, how often did your parents or guardians go through your things without your approval?
Respondents answered each question using a 5-point Likert scale with the following options: Never, Rarely, Sometimes, Most of the Time, and Always. Each response was given a numerical value. For questions one to three, the coding was as follows: Never = 1, Rarely = 2, Sometimes = 3, Most of the Time = 4, and Always = 5. For question four, Never = 5, Rarely = 4 Sometimes = 3, Most of the Time = 2, and Always = 1. The parental involvement score was the sum of the answers on the 4 questions and could range from 4 to 20 with higher scores indicating increased parental involvement. The median score on the parental involvement was then used to dichotomize the scale into high versus low parental involvement.
Dependent variables: adolescent mental well-being and physical health
A number of outcome scales were created to assess the overall mental and physical well-being of adolescents in this study. Each scale was based on survey questions related to its domain. These scales were nutrition (4 questions), exercise (4 questions), hygiene (4 questions), physical harm (3 questions), bullying (1 question), substance use (2 questions), tobacco use (2 questions), and mental health well-being (5 questions). All questions had Likert scale responses as described for the parental questions above. Coding was adjusted to allow higher scores to represent more favorable behaviour or better health. As such, higher scores on nutrition and on bullying represent better nutrition and less bullying respectively. Similar to the parental involvement scale, the median score for each outcome scale was used as a cut-off point to divide the population into high and low categories.
As an example of the individual scales used for the dependent variables, nutrition was determined using the following 4 questions:
During the past 30 days, how many times per day did you usually eat fruit, such as dates, apples, oranges, or bananas?
During the past 30 days, how many times per day did you usually eat vegetables, such as tomatoes, cucumbers, carrots, or lettuce?
During the past 30 days, how many times per day did you usually drink carbonated soft drinks, such as Pepsi, Coca Cola, or Mountain Dew? (Do not include diet soft drinks.)
During the past 7 days, on how many days did you eat food from a fast food restaurant, such as burger places, pizza places, or shawarma places?
Questions 3–4 were reverse coded so that higher consumption of carbonated drinks and fast food would result in a lower overall nutrition score. Scores ran from 0 to 29 based on the frequency of consumption of various food groups, with 0 representing the least healthy diet and 29 representing the most healthy diet.
Participants’ demographics and their self-reported answers to the parents involvement questions were summarized using frequency distributions. The association between parental involvement and adolescent behavioural and well-being scales were assessed in several ways.
First, Spearman’s correlations were used to measure the potential association between the scores of the several outcomes and the that of the parental involvement. Univariate and multiple linear regression were used to assess the relationship between each of the outcomes and the parental involvement scale as an independent predictor while adjusting for age, sex and food insecurity. Unadjusted and adjusted slopes along with p-values were reported. Finally, bivariate and multiple logistic regressions were fit for each of the dichotomized outcomes to assess the potential association with parental involvement (high/low) and adjusting for age, sex, and food insecurity. Unadjusted and adjusted odds ratios along with their 95% confidence intervals were presented. All data analyses were done using IBM SPSS (version 26, Armonk NY). A p-value of 0.05 or less was considered significant.