Sleeping Habits Amongst Riyadh Children: A Cross-Sectional and Validation Study

Background: To validate the Arabic translation of the Children’s Sleep Habit Questionnaire and determine the behavioral prevalence of sleep habits amongst school aged children in Riyadh, Saudi Arabia. Methods: This prospective cross sectional study involved a Linguistic validation involving 50 bilingual parent participants in each of the Arabic and English Surveys, with a cross sectional pilot involving 1546 parent participants regarding their children’s Sleeping Habits. The Children’s Sleep Habit Questionnaire was distributed via Survey Monkey and subjected to IBM-SPSS statistics. Parents of 1546 school aged children ranging from 3 years to 12 years of age completed the survey. Results: The Total score of all domains ranged between 11 and 70. Mean Score of 35-37 in 11 % of this population. The Children’s Sleep Habit Questionnaire scores were High in comparison to the Owen’s article; greater research is needed to determine the underlying causes of the increased likelihood of sleep disorders. Conclusion: Our ndings, especially in sleep anxiety are alarming and warrant for more research to determine underlying causes of the more prevalent sleep disorders (Sleep Anxiety) and any other apparent or possible causes for sleep disturbances amongst children.


Background
In children, sleep is of major signi cance for brain development, growth, and information processing. 1 Sleeping plays an integral role in the bio-psycho-social aspects of individuals. 2 The modern general populace recognizes the importance of good quality sleep more than they ever have in past decades thanks to unlimited resources freely provided over the Internet. 3 Citizens of rst world countries pay great interest in all matters affecting their child's well-being equally with scienti c communities; the latter of course examine topics through an analytical, and objective lens. Rigorous and consistent ow of studies ood research databases from wide demographics of diverse geopolitical backgrounds studied the matter thoroughly, with recommendations and renovations to sleeping guidelines published every where from governmental organizations such as the Centre's for Disease Control to academic institutions such as the American Academy for Sleep. 4 Without sleep, children can't process information and fatigue takes a toll on physical ability and mental acuity. 5 Signi cant evidence taken from multiple cross-sectional and cohort studies demonstrated the negative impact sleep deprivation has on cognitive and behavioral domains 6 ; children might also suffer from even more problematic sleep disorders when they grow older, Dr. Al-Mamun, a professor of School of Population Health in University of Queensland, Australia "During childhood development, children who had Trouble Sleeping Behavior were 1.49 times more likely to experience di culties sleeping at 14 years compared to those children without TSB" 7 . Al Mamun et al's main focus was to prove individuals who suffer in childhood of sleeping problems carry on suffering in their adult life, with greater risks of exacerbated sleeping disorders. 7 With the barrage of portable gaming computers, smart phones and electronics making them accessible in the car, in bed and even at school there are concerning detrimental health risks on so young an age group including and not limited to sleep quality. A great example of which can be found in a paper published by Dr. Li of the Otolaryngology-Head and Neck Surgery department in Beijing's Children's hospital; he noted "Increased incidences of sleep loss in children are due to the accessibility of electronics. Habitual use of electronics before bed-time decreases sleep duration amongst children by 2 times in comparison with other children. 2 Recently a study by Prado and colleagues evaluated factors in adolescents probable sleep bruxism and found that those who snored, as well as those that wore xed orthodontic appliances were more prone to Sleep Bruxism than those that didn't. 8 In several cases, determining duration and quality of sleep amongst children, depends on several factors. 9 Researchers who attempted to qualify sleep noted multiple sociodemographic markers such as gender, age group, residential region, parental education, personal activities before bedtime, as well as biometrics such as BMI grouping, in addition to familial history and genetics such as Sleep Disordered Breathing, and other sleeping disorders, to be of relevant use during data collection. 10 Sleep can be evaluated via multiple tools. 11,12,13 The most popular of which is Polysomnography (PSG) which is an overnight multi parametric diagnostic tool assessing brain waves, blood oxygen levels, heart and pulse rate, along with eye and lower limb movements. 14 Actigraphy, on the other hand, is a portable sleep tool and has an actimetry sensor that evaluates rest and activity sleep cycles. 15 Although PSG is the gold standard in reliability and validity for assessing sleep behavior, followed closely by Actigraphy, their use is costly, both monetarily and time wise, also challenging when used on a wide scale. Sleep diaries are runners up as the preferred sleep evaluators, however, require the parents and/or guardians scheduled compliance in logging for longer durations of time.
A study conducted by Owens and Colleagues in 2000, designed a questionnaire that psychometrically analyses the personality and attitude of children whilst asleep. It is a valid and reliable tool to assess both behaviorally based and medical sleep disorders. 16 This tool is based on 35 questions and these are entailed to one of 8 domains, to be completed by the parent and or guardian of the child. Initially this questionnaire was designed for children aged 4 to 10 years of age. The older age group wasn't recommended to be included to avoid hormonal pubertal changes affecting the domains. However, in our study, we included children from both the Phallic and Latent stage. That is, according to Freudian Psychoanalysis, children between the ages of 3 and 12 whom are categorized into Phallic and Latent stage. The former encompasses the ages 3 to 6 which is the third stage of psychosexual development, subsequently the latter known as the Latent stage consists of age 6 to puberty and is the fourth stage of psychosexual development. 10 During our search we concluded there isn't su cient literature to measure sleeping habits amongst Riyadh's children nor has the Owens Children's Sleep Habit Questionnaire (CSHQ) been translated into the Arabic language. Hence, this study was conducted to assess the prevalence of sleep disorders and translate the CSHQ 16  A total of 50 parents from Riyadh region participated in the survey.
Each parent lled the survey once in Arabic and another time in English (total of English and Arabic surveys were 100). Potential participants were briefed about the purposes and procedures of the study and a verbal consent was taken as well as an electronic consent before completing the electronic survey. Inclusion criteria were as follows, parents of male and female medically t children from 3-12 years of age. Children more than the age of 12 and less than the age of 3, children with hearing or visual impairment, children with learning disabilities, major psychiatric illnesses, or with maxillofacial deformities were excluded from the survey. 2 Exclusion and inclusion criteria's were mentioned with the consent.
The correlation coe cient was computed to examine the correlation between responses of English and Arabic questionnaires. For continuous variables, Pearson correlation coe cient (r)was used. The Pearson coe cient of correlation (r) measure the degree of association between quantitative variables. The values of r lie between -1 and +1, a coe cient of +1 indicates that the two variables are perfectly positively correlated, so as one variable increases, the other increases by a proportionate amount. Conversely, a coe cient of -1 indicates a perfect negative relationship: if one variable increases the other decreases by a proportionate amount. A coe cient of zero indicates no linear relationship at all and so if one variable changes, the other stays the same. Pearson coe cient of correlation calculated using the following formula: Where Xs are observed values of the independent variable, Ys are observed values of the dependent variable, and N is the sample size.
Spearman rank correlation was used for categorical variables.
A value of 0.70 and above correlation was considered as indication of consistency in responses.
The Saudi Arabic version of Owens (CSHQ) 16 was adapted in the following six stages. Fig.1

Stage I Forward translation
Three bilingual translators, with Arabic as their mother tongue independently completed three forward translations of the original English version into the Saudi Arabic version. The translators aimed for a literal and conceptual translation. A written report with their comments on any di culties and the rationale for their choices for problematic questions was completed.

Stage II Synthesis
A meeting between the forward translators was arranged, and the three translators compared both translated documents. The translators synthesized both translated documents into one Arabic version by resolving any discrepancies on their reports.

Stage III Backward translation
Three bilingual translators with English as their mother tongue, back-translated the Arabic version into English. The three back-translation versions were then compared. The discrepancies were discussed and resolved by agreement.

Stage IV Expert review
An expert forward translator, a specialist in Orofacial pain from the Oral Maxillofacial Surgery department at Riyadh Elm University, reviewed all translated versions. The result was Pre-nalized Saudi Arabic version of the Owens Children's Sleep Habits Questionnaire. The Pre-nal version was compared to the original English version to ensure semantic equivalences.

Stage V Pilot Study
The Pre-nalized version was tested on a sample of 50 Saudi parents with medically t children from age 3-12 years of age. Parents were asked to complete the questionnaire twice, once in the original English version and secondly in the pre-nal Arabic version. An expert reviewed the answers of both completed questionnaires of each individual parent and no changes were needed.

Stage VI Prevalence of affected sleep domains
Both English and Arabic versions of the Owens CSHQ were distributed separately on a different sample of 1546 parents. The choice was up to each participant whether to solve the English or the Arabic version of the standard Questionnaire.

Prevalence participants
A total of 1546 parents from Riyadh region participated in the survey. Participants were made aware regarding the purposes and procedures of the study and an electronic consent was attained before conducting the electronic survey. Inclusion criteria were as follows, parents of male and female medically t children from ages 3-12 years old. Children more than the age of 12 and less than the age of 3, children with hearing or visual impairment, children with learning disabilities, major psychiatric illnesses, or with maxillofacial deformities were excluded from the survey. 2 Exclusion and inclusion criteria's were mentioned with the consent. These participants had the choice of completing the questionnaire in their preferred language (Arabic or English).

Demographics
Demographics containing the following questions, how old is the mother, how old is the father, what is the education level of the mother, what is the education level of the father, average household income per month, which of the parents is currently working, which parent spends more time with the children, the marital status of the parents, how many children do you have, how old is your child, what gender is your child, which grade is your child in, does your child go to a private or public school, when does your child sleep during the weekdays, and when does your child sleep during the weekends (specifying that the participant should only choose one child to ll the survey).

Owens Children's Sleep Habits Questionnaire
The Owens Children's Sleep Habits Questionnaire 16 consists of 8 sub-scales including Bedtime Resistance (6 items) with two reversed scores, Sleep Onset Delay (1 item) has a reversed score, Sleep Duration (3 items) with two reversed scores, Sleep Anxiety (4 items), Night Wakings (3 items), Parasomnias (7 items), and Sleep Disordered breathing (3 items) all with no reversed scores, and Daytime Sleepiness (8 items) with one reversed score. Total Sleep Disturbance score (33 items), scoring from 1-3 where 1 is Never/Rarely, 2 is sometimes, and 3 is usually. Missing data were treated according to developer guidelines.
All Questionnaires were distributed through survey monkey and subjected to IBM-SPSS statistical analysis.

Psychometric measurements and statistical analysis
Sample size estimation Estimation of sample size was established within Riyadh with a population size of 621272 and a con dence level of 95% with 5% margin of error, suggesting a sample size of 384 in total.
Floor and sealing effect When 15% or more of the participants score the lowest or highest possible sub-scale scores and this is called "Floor and sealing effect". The gures in the table above show the correlation results for consistency between English and Arabic versions of socioeconomic questions in the questionnaire. The value of correlation coe cient equal or above 0.70 is considered as indication of consistency in responses between two versions of questionnaire. The results above, con rm consistency for all characteristics except for three variables namely, the parent who spends more time with the children, marital status and type of school. The responses on the three aforementioned characteristics were different between two collection time points.  The maximum age of children was 12 years with mean ages 5.0 years. In the rst domain, Bedtime Resistance (BR) score with its scale between − 2 to + 10 we found 122 individuals (7.9%) scored a minimum score of -2, with 3 individuals (0.2%) having a maximum score of 10, with a mean score of 4 in 151 (9.8%) individuals.

Discussion
The current study aimed to translate and evaluate the properties of the Saudi Arabic version of Children's Sleep Habits Questionnaire in children from 3-12 years old. This paper reports the development and preliminary validation of Owens Children's Sleep Habits Questionnaire. A cognitive straight forward structure has been established for validating the Questionnaire as in Fig. 1 The reported psychometric properties of the sleep screening questionnaire was designed primarily for survey the sleep habits and sleep disturbances in community populations. 16 Owens States that the higher the score the worse the result is.
To relatively determine the most prevalent domain in our community, we've taken into consideration two items: the rst is the number of respondents in each category and the second, the frequency of their The total score of all domains ranges with their respective individual questions of CSHQ was calculated, with results between 11 and 70 where 2% of our population scored a minimum of 11-12 and 1% with a maximum of 63. Between 0% and 1% of the population scored from 64-70 with a mean score of 35-37 in 11% of our population.
copious amounts of literature supporting both of these especially in Youth which in included in our age sample. 19 Parasomnias Domain with a score ranging from 7-20 with 2 (0.1%) individual scoring the maximum at 20 and the minimum with 262 (16.9%) individuals scoring 7. Sleep Disordered Breathing with a score ranging from 3-9 with 6 (0.4%) individual scoring the maximum at 9 and the minimum with 914 (59.1%) individuals scoring 3. Night Wakings with a score ranging from 3-9 with 16 (1%) individual scoring the maximum at 9 and the minimum with 438 (28.3%) individuals scoring 3.
Bedtime Resistance domain with a score ranging from − 2 to 10 with 3 (0.2%) individuals scoring the maximum at 10 and the minimum with 112 ( Early period of adolescence (10-12 years of age) as in our age included group plays a critical role in determining emotional health. That is poor sleep might trigger negatively with resultant anxiety,in turn, paving the road to depression. 22 This is seen in older age group where those aged 12-18 had depression rather than anxiety when their circadian rhythm was affected. 23 Our study corroborates the risk of poor sleep in increased likelihood to anxiety in early adolescence.
The tell tale signs of sleep disorders is concerning for they de ne a trajectory of developmental impact and poor emotional well being. Awareness of parents on the impact of poor sleep on health might be the simplest yet most e cient of measures. This study,didn't only assess the prevalence of sleep quality and /or disorders in children but unintentionally raised the alarm for parents/guardians to take their children's sleep more seriously. In the Kingdom of Saudi Arabia, the young aged 20 and below account for over half of the Kingdoms Population. If this generations sleep disorders not addressed, it will indeed be ubiquitous that this generation will be riddled with concerning health problems both developmentally and most importantly psychologically. This will impact not only the economy, but be a burden on the healthcare system.

Limitations
The study was conducted during an academic semester and the researchers had a strict timeline within which to complete survey distribution. Better reliability was examined through test retest and not limited to the internal consistency tests. 24,25 Validation limitations The limitations we have encountered included a low response rate, missing entries, recall bias, reporting bias, and selection bias.

Prevalence limitations
The limitations we have encountered were, missing demographics for the prevalence sample only due to a technical issue saving it on survey monkey, some missing entries, recall bias, reporting bias, and selection bias.

Conclusion
CSHQ scores were high in comparison with Owen's article, even though we are missing key demographic data, our ndings, especially in sleep anxiety are alarming and warrant for more research to determine underlying causes of the more prevalent sleep disorders (Sleep Anxiety) and any other apparent or possible causes for sleep disturbances amongst children. All further material is available and can be acquired upon request.

Competing Interests
The Authors declare no con ict of interest.  Flowchart of translation and linguistic adaptation Figure 2