Investigation of Factors Influencing Sleep Status in Beijing Children Aged 3 to 14 Years

Introduction

Obstructive sleep apnea syndrome (OSAS) is a common cause of sleep disordered breathing in children. ¹ The prevalence reported abroad and in Hong Kong is 5.7% and 4.8%, respectively.^2,3 The poor sleep quality can produce excessive daytime sleepiness, behavioral problems, learning disabilities, right-sided heart failure, growth retardation, or failure to thrive. Large number of children with sleep disorders visit to ENT department. We investigated the pre-sleep factors influencing on sleep quality in children. The current gold standard for the diagnosis of OSAS is polysomnography (PSG), but it is relatively expensive. It is mainly used in hospitals, PSG is not widely used at home and abroad, so it is necessary to develop a low-cost, high-sensitivity and specific screening method for OSAS of children. Chervin et al based on American OSAS children’s sleep habits developed the Pediatric Sleep Questionnaire (PSQ). ⁴ The PSQ is the only screening questionnaire recommended by the American Academy of Pediatrics (AAP) in the 2012 OSAS guidelines. Currently widely used to evaluate children with OSAS. Studies at home and abroad have shown high reliability and validity.^5,6 We used the PSQ questionnaire to investigate the sleep status of 11 420 children in Beijing and to analyze the relationship between PSQ scores and the children’s behavior before sleep.

Materials and Methods

Results

Descriptive Statistics

A total of 11 420 questionnaires were issued and 10 743 (94.07%) were returned. After data cleaning, 9198 (80.54%) were considered valid. The mean PSQ score was 3.60 ± 2.69 points. A total of 816 children (8.87%) scored more than 7. The PSQ scores of different genders, ages and other characteristics are shown in Table 1. These differences reach statistical significance (P < .05).

OPEN IN VIEWER

Table 1.

Pediatric Sleep Questionnaire (PSQ) scores under different characteristics.

	Cases(n)	PSQ scores	t Value	P value
Age (years)
<6	2447	3.94 ± 2.58	(33.015)	<.001
6-12	4512	3.58 ± 2.66
>12	2239	3.30 ± 2.84
Sex
Male	4736	3.89 ± 2.75	10.560	<.001
Female	4462	3.30 ± 2.60
Obese
Yes	1004	4.50 ± 3.06	11.278	<.001
No	8194	3.49 ± 2.62
Snoring frequency*
<1 time/week	7987	3.19 ± 2.34	(1042.950)	<.001
1-2 times/week	532	6.41 ± 2.83
≥3 times/week	369	7.86 ± 3.34
Watching TV before sleep*
Yes	5331	3.78 ± 2.71	7.333	<.001
No	3833	3.36 ± 2.66
Eating before sleep*
Yes	2683	4.08 ± 2.79	10.842	<.001
No	6483	3.41 ± 2.63
Playing video game before sleep*
Yes	1741	4.23 ± 2.91	10.787	<.001
No	7426	3.46 ± 2.62
Surfing the Internet before sleep*
Yes	1227	4.03 ± 2.94	5.950	<.001
No	7941	3.54 ± 2.65
Exercising before sleep*
Yes	1724	3.77 ± 2.64	2.826	.005
No	7443	3.57 ± 2.70
Father/mother snoring*
Yes	5703	3.93 ± 2.76	14.918	<.001
No	3477	3.07 ± 2.49

*

Some indicators are missing.

Multi-Level Model Analysis

An empty model was implemented, with schools as a random effect. The results showed an intraclass correlation coefficient (ICC) of 1.76 (P = .003), suggesting that the data in this study has a hierarchical structure, and there were large differences between different schools, hence a multi-level model was used.

Taking individual schools as level 2 and their students as level 1, the multi-level model was implemented by incorporating possible relevant experiences. The results showed that age, gender, obesity, snoring frequency, eating before sleep, playing videogames before sleep, surfing the internet before sleep, and parental snoring may be related to PSQ score. The parameter estimation results of each index are shown in Table 2. It can be seen that children’s snoring frequency has the greatest impact on PSQ. Children with snoring 1 to 2 times/week and snoring ≥3 times/week have a mean PSQ score of 2.972 points and scored 4.311 points higher than those who do not snore. Child obesity and parental snoring also have a great impact. Obese children have a mean PSQ score 0.871 points higher than children with normal BMI; and children whose parents snore have a mean PSQ score 0.553 points higher than children whose parents do not. In addition, children who eat, surf the internet, and play videogames before sleep had higher PSQ scores than those who do not. These differences reach statistical significance (P < .05).

OPEN IN VIEWER

Table 2.

Multi-Level Model Analysis Results of Factors Related to the PSQ Score.

	Parameter estimation	95% CI	Standard error	t Value	P value
Age [<6 years vs >12 years]	0.615	0.386-0.844	0.115	5.354	<.001
Age [6-12 years vs >12 years]	0.264	0.076-0.453	0.095	2.777	.006
Sex [male vs female]	0.412	0.312-0.511	0.051	8.127	<.001
Obesity [obese vs normal]	0.871	0.706-1.035	0.084	10.349	<.001
Snoring frequency [1-2 times/week vs none]	2.972	2.764-3.180	0.106	28.010	<.001
Snoring frequency [≥3 times/week vs none]	4.311	4.063-4.560	0.127	34.010	<.001
Watching TV before sleep [yes vs no]	0.033	−0.072-0.138	0.054	0.619	.536
Eating before sleep [yes vs no]	0.470	0.357-0.583	0.058	8.172	<.001
Playing video games before sleep [yes vs no]	0.348	0.216-0.479	0.067	5.179	<.001
Surfing the Internet before sleep [yes vs no]	0.404	0.251-0.557	0.078	5.166	<.001
Exercising before sleep [yes vs no]	−0.014	−0.142-0.113	0.065	−0.220	.825
Father/mother snoring [yes vs no]	0.553	0.451-0.655	0.052	10.623	<.001

Discussion

There are many types of scales used for children’s OSAS screening worldwide. Spruyt and Gozal found that a large number of children’s sleep-related questionnaires and scales only reported their reliability and validity within the study. ⁷ Chervin et al developed a children’s PSQ questionnaire based on American OSAS children’s sleep habits and risk factors. ⁴ This questionnaire is the only recommended screening questionnaire in the 2012 OSAS guidelines of the AAP. Its sensitivity and specificity in the United States are as high as 85% and 81% respectively, which is superior to other children’s sleep questionnaires. However, due to differences in region, race, environment, living habits, growth and development, it is necessary to evaluate its reliability, validity, and applicability before applying the questionnaire outside the United States. Li et al studied the data of children in Beijing and evaluated the reliability, validity and screening ability of a simplified Chinese version of PSQ. ⁶

Sleep is of great significance in the growth and development of children, especially of the central nervous system. ⁸ The secretion of growth hormone, which is necessary for children’s physical development, appears mostly in the first slow-wave sleep after sleep initiation. ⁹ Therefore, the quality of sleep may directly affect children’s physical and mental development. Poor sleep quality may even cause behavioral problems.^10,11 Therefore, good sleep quality plays a very important role in the development and maturity of children’s central nervous system.

Our results showed that the higher the frequency of snoring, the higher the PSQ score. Children’s snoring frequency had a positive relationship with PSQ score. We found that younger children scored higher in the PSQ, with preschool children having the highest and adolescent children the lowest scores. In addition, boys scored significantly higher than girls. This suggests that more attention should be paid to the sleeping conditions of boys and preschool children.

We also found that certain activities before sleep such as eating, surfing the internet, and playing videogames are related to a higher PSQ score. Harvard University research shows that playing on smartphones, watching TV shows, or playing videogames before bedtime, exposes individuals to blue light which can inhibit melatonin secretion and affect sleep quality and increase disease incidence. ¹² Children and adolescents are in a period of growth and development in which the harmful effects of poor sleep may be greater than in adults. Owens et al found that playing games and watching TV excessively can seriously affect sleep quality in children. ¹⁰

The family is the main environment in children’s early life. Children’s sleep will inevitably be affected by the family environment and parents’ education methods. Furthermore, the bedtime emotional communication can contribute to an array of positive developmental outcomes beyond improved sleep, inclusive of language development, literacy, child emotional and behavioral regulation, parent-child attachment, and family functioning. Findings from the national Sleep in America poll, for example, indicate that children who follow parent-child communication evidence significantly longer caregiver-reported sleep duration. Other cross-sectional research has also shown that having a bedtime communication is concurrently associated with better caregiver-reported sleep quality and longer nighttime sleep duration. ¹³ When parents don’t, the lack of this parent-child communication makes sleep disorders prone to developing. Bad bedtime habits are an important factor affecting sleep disorders in children.^14,15

This research is innovative. This is the first time that the Chinese version of the PSQ questionnaire has been used in China to investigate the sleep quality of children. Second, this study is based on a large-scale population survey and analysis which can more accurately reflect the sleep status of Beijing children. Limitations include the cross-sectional study design, meaning that relationships between possible influencing factors and PSQ scores were analyzed at one point in time, and thus cannot confirm causal relationships. Further studies are required to differentiate causality from association.

Conclusion

As a result of our study findings, we recommend that parents should pay attention to their children’s sleep status; especially preschool children. If parents find that children have a high snoring frequency, physician consultation is recommended. Children’s bedtime behaviors such as surfing the internet and playing videogames negatively affect sleep quality in children.