Sleep and social–emotional problems in preschool-age children with developmental delay

Abstract

Sleep problems and social–emotional (SE) problems are common in preschooler children and may be affected by culture. However, little is known about the link between sleep and social–emotional problems in non-Western countries. This study aims to compare sleep problems and their association with SE problems in Turkish children with either typical development (TD) or neurodevelopmental delay (NDD). In this case–control study, children with NDD (n = 126) and children with TD (n = 102) were included. Parents completed the Children’s Sleep Habits Questionnaire (CSHQ), and the Ages and Stages Questionnaire: Social–Emotional (ASQ:SE). More than half of the TD and NDD groups had sleep problems when assessed by CSHQ (54.4%, 72.2%, respectively). The correlation between CSHQ and ASQ:SE scores in the NDD and TD groups was r = 0.44 and r = 0.352, respectively (both p < 0.001). Children who slept less than 9 h had lower ASQ:SE scores in the NDD (p = 0.003) and TD group (p = 0.023). In the TD group, those who slept after 23:01 h had lower ASQ:SE scores compared to early sleepers (p = 0.04). Multivariate analysis revealed associations between SE problems and male gender, lower family income, bedtime resistance, and shorter sleep duration in the NDD group. In the TD group, male gender, sleep onset delay, sleep duration of CSHQ subscale were found to be associated with SE problems. Sleep problems were identified in more than half of all preschooler children, regardless of developmental delay, and were associated with increased SE problems. Interventions to correct sleep problems may have a positive impact on children’s emotional development.

Introduction

Sleep plays a crucial role in a child's physical and mental health and the quality of life of children and their families [1]. Children are prone to sleep problems, yet it is more common to encounter such problems in children with a developmental delay [2]. It has been reported that 20–60% of children with typical development (TD) have sleep problems [3, 4]. In children with neurodevelopmental delay (NDD) this proportion rises to 50–80% [2, 5, 6].

Sleep onset problems and sleep maintenance problems are frequently encountered in children with TD and children with NDD [7]. Pre-frontal cortical dysfunction in children with TD, likely resulting from poor sleep, has been associated with a variety of daytime consequences including emotional and behavioral issues [8]. Domains such as executive functioning, social behavior, and emotional regulation are impaired in children with NDD, and are further affected by sleep problems [9].

The co-occurrence of sleep problems and social–emotional (SE) problems in the pediatric population has been a focus of research over the past decade. The first five years of life are crucial for the evolution of sleep duration, quality, patterns, and architecture although these factors continue to develop throughout childhood [10]. There is a strong association between sleep and mental health issues, as indicated by several studies across all age groups [11, 12]. Yet some studies have shown that the strength and nature of the link between sleep and emotional functioning may vary depending on age through childhood [13]. The association between sleep and emotional functioning in later childhood has been investigated much more compared with this association in younger children [14], even though younger children have been included in some study samples [15, 16].

The cultural milieu and environment have a major impact on children’s sleep and play an important role in the social and emotional development of children. This study was conducted in Turkey, a Eurasian country. As mentioned above, the sleep problems of children who are older than 2 years, especially preschool age predict behavior problems in later childhood and adolescence [13]. Our objective in this study was to compare sleep problems and their association with SE problems in both TD and NDD preschool-age children (3–6 years) in Turkey.

Methods

This is a case–control study conducted in Ihsan Doğramacı Children’s Hospital Developmental Pediatrics Clinic of Hacettepe University. The study was approved by the local Ethics Committee. The study conformed to the Provisions of the Declaration of Helsinki in 1995.

Participants

The study included children aged 3–6 years with a neurodevelopmental disorder (ASD, GDD, LDD) or typical development followed up at the Developmental Pediatrics Clinic. Exclusion criteria for all children were using medication for sleep problems and having ADHD comorbidity diagnosis. At the first visit, a detailed developmental history, physical examination, and developmental screening assessment using the Ages and Stages Questionnaires, Third Edition (ASQ-3) [17], or Bayley Scales of Infant and Toddler Development III [18] were conducted by a developmental pediatrician. As a result of these evaluations, a diagnosis of global development delay (GDD) or language development delay (LDD) was made. Children who were diagnosed as having autism spectrum disorder (ASD) during a clinical evaluation by a child psychiatrist and a developmental pediatrician were also enrolled to the study. Diagnoses were made using the Diagnostic and Statistical Manual for Mental Disorders—Fifth Edition (DSM-5) criteria, taking all assessment information gathered into consideration. Further formal psychological or speech and language and/or occupational therapy assessments were conducted for children with ASD, GDD or LDD, as recommended. Children, also aged 3–6 years, who were admitted to the clinic because of parental concern about their development but who were considered to be developing typically, after holistic evaluation, constituted the TD control group. The initial assessment lasts for 30–45 min. Observation is a standard procedure for assessment of all cases. Cases are than followed up with 3-month intervals. The sessions with therapist are 45 min.

Procedure

All parents who had a follow-up visit between September 2018 and November 2018 were invited to participate the study and were given written information explaining the purpose of the study and informed that participation was voluntary. During the follow-up visit, the importance of sleep on child’s well-being was explained by the developmental pediatrician. All parents whose children took part gave written informed consent for the study. Parents filled out the questionnaire during the face to face interview in the clinic. The interview took approximately 20–30 min per parent and was conducted by the researcher (developmental pediatrician fellow).

The Children's Sleep Habits Questionnaire (CSHQ) is a sleep-screening instrument designed for children (4–10 aged) that uses parental reporting [19]. It was used for screening sleep problems of toddler and preschooler children [20]. The CSHQ is a 33-item, parent-completed questionnaire with eight subscales: Bedtime Resistance, Sleep-Onset Delay, Sleep Duration, Sleep Anxiety, Night Waking, Parasomnias, Sleep-Disordered Breathing, and Daytime Sleepiness. Parents were instructed to answer questions regarding their child’s sleep during a typical recent week. Each question had three possible answers: 1 for rarely (0–1 time per week), 2 for sometimes (2–4 times per week), and 3 for usually (5–7 times per week). Higher scores indicate more sleep problems. A total CSHQ score of 41 has been reported to be a sensitive clinical cut-off for the identification of probable sleep problems [19]. The CSHQ has also been used to assess the presence of sleep disorders in children with NDD [21]. The Turkish validation and reliability study of the CSHQ was conducted by Fiş et al. [22].

According to National Sleep Foundation's sleep quality recommendations, first report, a wake time after sleep onset of ≤ 20 min indicates good sleep quality [23]. Therefore, we used the cut-off 20 min for classifying wake time after sleep onset.

The Ages and Stages Questionnaire: Social–Emotional (ASQ:SE) is a short, parent-reported questionnaire, used to identify developmental status in the areas of social and emotional functioning among children aged between 3 and 72 months. Questions were included in seven different fields of social–emotional development: self-regulation, compliance, communication, adaptive functioning, autonomy, affect, and interaction with people. The Turkish validation study has been conducted by Kucuker et al. [24]. The Turkish translation has eight forms of questionnaire according to age. We used the 36 months (33–41 months), 48 months (42–53 months) and 60 months (54–72 months) questionnaires in the present study. Higher scores indicate higher levels of maladaptive symptomatology and are indicative of risk. A high score on the ASQ:SE indicates social–emotional deficits or delays.

Statistical analysis

To detect a significant difference of sleep problems among TD children (20%) and NDD children (40%) with a power of 80% and two-sided 'a' of 0.05, the sample size needed was estimated to be 64 children per group.

Statistical analysis was performed using the IBM SPSS 22.0 software package (IBM Inc., Armonk, NY, USA). Descriptive statistics were used to describe children with TD and NDD. The total CSHQ score and scores of the ASQ:SE for each child were used as dependent variables in statistical analyses. Scores with higher values reflect greater problems. Data analysis consisted of two different strategies: (1) comparison of ASQ:SE scores and the CSHQ total score for children with TD and NDD; and (2) evaluation of the diagnoses of children and sleep parameter effects on all ASQ:SE scores for all children. Normality of the continuous variables was evaluated by the Kolmogorov–Smirnov test. Numerical variables are given as mean ± standard deviation or median (interquartile ranges). Categorical variables are presented as frequencies and percentages. Categorical variables were compared using the Chi-square test. Differences between the TD and NDD groups according to sleep parameters and ASQ:SE scores were determined by the Mann–Whitney U test. Differences between the groups of total sleep duration and sleeping time in both TD and NDD groups were determined using the Kruskal–Wallis test. Spearman correlation was used for the relation of subscales of CSHQ and ASQ:SE score. Forward logistic regression was performed to investigate if any specific sleep variable could predict SE problems in the preschool period when confounding variables were controlled for.

Results

Three hundred and thirty caregivers were interviewed. Of these 70 did not complete the Children's Sleep habits Questionnaire (CSHQ) and 32 did not adequately complete the social and emotional component of the ASQ, so they were excluded from the study. As a result, 126 children diagnosed with NDD [ASD (n = 31), LDD (n = 43), GDD (n = 52)] and 102 preschool-age children with TD were included in the study (Fig. 1). In the NDD group, 6% (n = 8) of children had gastroesophageal reflux, 5% (n = 7) of children had epilepsy, 4% of children had a psychiatric disorder (like hyperactivity, self-injury), 11% (n = 14) of children had constipation and 5.5% (n = 7) had dental problems according to parental reports. The power of the study, which included 228 children, was 87.5% for sleep problems in NDD and TD with a 5% error rate. The demographic characteristics of children in the TD and NDD groups are shown in Table 1. There was no difference in age between the two groups. There was a statistically significant difference between the two groups in terms of sex, education level of the parent, the employment status of the mother, and income of the family (see Table 1).

Fig. 1.

Selected of study sample

Table 1.

Demographic characteristics of children with typical development (TD) and neurodevelopment delay (NDD)

	TD (102)	NDD (126)	Mean difference (95% CI)	p value
Age, month, mean ± SD	51.8 ± 11.1	52.0 ± 10.2	− 0.2 (− 3.0 to 2.6)	0.875
Sex (female/male)	52/50	38/88		0.001
Maternal age, mean ± SD	33.1 ± 5.4	34.0 ± 5.8	− 0.9 (− 2.5 to 0.6)	0.234
Paternal age, mean ± SD	36.3 ± 5.3	37.3 ± 7.5	− 1.0 (− 2.8 to 0.8)	0.274
Maternal education, n (%)				< 0.001
$\le$ 8 years	19 (18.6)	58 (46.8)
9–12 years	31 (30.4)	29 (23.4)
> 12 years	52 (51.0)	37 (29.8)
Paternal education, n (%)				0.016
$\le$ 8 years	14 (13.7)	35 (28.2)
9–12 years	31 (30.4)	39 (31.5)
> 12 years	57 (55.9)	50 (40.3)
Maternal employment, n (%)				0.032
Employed	34 (33.7)	26 (21)
Unemployed	67 (66.3)	98 (79)
Income monthly, n (%)				0.015
0–5000 TL	54 (61.4)	82 (75.2)
> 5000 TL	34 (38.6)	27 (24.8)

The statistical significance (p < 0.05) was presented in bold

Most children in the TD (54.4%) and NDD (72.2%) groups had sleep problems according to the CSHQ. In the NDD group, the percentages of sleep problems in children with GDD, LDD, and ASD were 67.3%, 69.8%, and 83.9%, respectively (p = 0.240). In the NDD group, bedtime resistance (sharing bed with parent/sibling, needs a person in the room to fall asleep, afraid of sleeping alone, struggles at bedtime, unable to fall asleep in own bed) was significantly more frequent than in the TD group (see Table 2).

Table 2.

Sleep parameters of children with typical development (TD) and neurodevelopmental delay (NDD)

	TD	NDD	Mean difference (95% CI)	p value
Total sleep duration, hours, mean ± SD	10.5 ± 1.4	10.5 ± 1.5	0.0 (− 0.4 to 0.3)	0.863
Total sleep duration, n (%)				0.283
$\le$ 9 h	23 (22.3)	20 (16.4)
9–11 h	51 (49.5)	73 (59.8)
$\ge$ 11 h	29 (28.2)	29 (23.8)
Wake after sleep onset, n (%)				0.001
< 20 min	88 (95.7)	97 (80.2)
> 20 min	4 (4.3)	24 (19.8)
Sleeping time, n (%)				0.385
At or before 21:00 h	29 (28.4)	26 (21.1)
Between 22:00 and 23:00 h	61 (59.8)	84 (68.3)
After 23:00 h	12 (11.8)	13 (10.6)
CSHQ total score, mean ± SD	45.03 ± 7.7	45.98 ± 7.4	− 0.95 (− 2.93 to 1.04)	0.348
CSHQ > 41 n (%)	56 (54.4)	91 (72.2)		0.005
CSHQ subscale scores, median (IQR)
Bedtime resistance	10 (8–12)	11 (9–13)		0.002
Sleep onset delay	1 (1–1)	1 (1–2)		0.184
Sleep duration	3 (3–4)	3 (3–4)		0.945
Sleep anxiety	7 (5–8)	7 (6–9)		0.240
Night waking	4 (3–5)	4 (3–5)		0.998
Parasomnia	9 (7–10)	9 (7–10)		0.356
Sleep disordered breathing	3 (3–4)	3 (3–4)		0.775
Daytime sleepiness	9 (8–12)	9 (8–11)		0.187

The statistical significance (p < 0.05) was presented in bold

CSHQ Children Sleep Habits Questionnaire

The mean of ASQ:SE scores of children with TD and NDD were 29.9 ± 22.9 and 59.98 ± 43.09, respectively (p < 0.001). In the TD group, nine (8.8%) children had an ASQ:SE score above the cut-off whereas this proportion was 40% (n = 50) in children with NDD (p < 0.001).

In the TD group, all children with a high ASQ:SE score also had sleep problems. The score of ASQ:SE was found to be higher in children with TD, GDD, and LDD with sleep problems compared with children without sleep problems (see Table 3). In the TD group, the median ASQ:SE score of children who sleep before 21:00 h, between 21:01 and 23.00 h, and after 23:01 h were 15, 25, and 30, respectively (p = 0.040). The difference between those who slept before 21:00 h and after 23:01 h was statistically significant (p = 0.021). In the NDD group, the median ASQ:SE scores of children who sleep before 21:00 h., between 21:01 and 23.00 h., and after 23:01 h. were 40, 50, and 65, respectively (p = 0.343). The relationship between total sleep duration and the ASQ:SE score in the TD and NDD groups was examined by Kruskal–Wallis analysis. In the TD group, the median ASQ:SE score of children whose sleep duration was less and equal 9 h, between 9 and 11 h, and more than 11 h were 30, 25, and 15, respectively (p = 0.023). There was a significant difference between the median score of ASQ:SE of children whose sleep duration was less than 9 h compared to those longer than 11 h in TD group (p = 0.009), the other significant statistical difference was found between the median score of ASQ:SE of sleep duration between 9 and 11 h. and the median score of ASQ:SE of sleep duration for more than 11 h. (p = 0.033). In the NDD group the median ASQ:SE score of children whose sleep duration was less and equal 9 h, between 9 and 11 h, and more than 11 h were 82.5, 50, and 30, respectively (p = 0.003). In the NDD group, the significant statistical differences were found between the median score of ASQ:SE of children whose sleep duration was less than 9 h and more than 11 h (p = 0.003), and between the median score of ASQ:SE of sleep duration between 9 and 11 h. and the median score of ASQ:SE of sleep duration for more than 11 h. (p = 0.009).

Table 3.

The relationship between ASQ:SE, sleep problems and developmental diagnosis

	ASQ:SE score median (IQR)	p value
Typical development (n = 102)		0.001
CSHQ ≤ 41 (47)	20 (10–30)
CSHQ $>$ 41 (55)	30 (20–45)
Neurodevelopmental delay (n = 126)		< 0.001
CSHQ ≤ 41 (35)	30 (15–50)
CSHQ $>$ 41 (90)	55 (35–95)
Global developmental delay (n = 52)		< 0.001
CSHQ ≤ 41 (17)	35 (15–50)
CSHQ $>$ 41 (35)	60 (35–90)
Language development delay (n = 43)		0.04
CSHQ ≤ 41 (13)	20 (15–30)
CSHQ $>$ 41 (30)	38 (25–45)
Autism spectrum disorder (n = 30)		0.290
CSHQ ≤ 41(5)	80 (40–85)
CSHQ $>$ 41(25)	105 (80–135)

The statistical significance (p < 0.05) was presented in bold

CSHQ Children Sleep Habits Questionnaire, ASQ:SE Ages and Stages Questionnaire: Social–Emotional

In the TD group, 16.5% (n = 17) of children usually woke up after sleep onset once at night, 7.8% (n = 8) of children usually woke up after sleep onset more than once at night according to parental reports. In the NDD group, these rates were 18.3% (n = 23) and 6.3% (n = 8), respectively. There was not any statistical difference between the two groups (p = 0.433, p = 0.435). In the whole group, only 28 (28/228; 12.3%) children woke up for more than 20 min after sleep onset during the night and only four were in the TD group (4/28; 14.3%). In the NDD group, 34.4% (33/96) of children whose wake time after sleep onset was less than 20 min, and 58.3% (14/24) of children whose wake time after sleep onset was more than 20 min had ASQ:SE scores above cut-off (p = 0.032).

The CSHQ scores were positively correlated with ASQ:SE scores (r = 0.43, p < 0.01) in the whole cohort. In the NDD and TD groups, the correlations between the scores of CSHQ and ASQ:SE were r = 0.441 and r = 0.352, respectively (both p < 0.001). The correlations between the scores of CSHQ subscales and ASQ:SE are shown in Table 4.

Table 4.

Bivariate correlation between subscales of CSHQ and ASQ:SE scores in children with typical development and neurodevelopmental delay

	Typical development ASQ:SE		Neurodevelopmental delay ASQ:SE
	r (95% CI)	p	r (95% CI)	p
Bedtime resistance	0.231 (0.038 to 0.407)	0.019	0.429 (0.275 to 0.562)	< 0.001
Sleep onset delay	0.05 (− 0.146 to 0.242)	0.579	0.267 (0.097 to 0.422)	0.002
Sleep duration	0.263 (0.072 to 0.435)	0.007	0.265 (0.094 to 0.42)	0.003
Sleep anxiety	0.277 (0.087 to 0.447)	0.005	0.315 (0.148 to 0.464)	< 0.001
Night waking	0.275 (0.085 to 0.446)	0.005	0.198 (0.024 to 0.36)	0.026
Parasomnia	0.292 (0.103 to 0.46)	0.003	0.323 (0.157 to 0.471)	< 0.001
Sleep disordered breathing	0.076 (− 0.12 to 0.267)	0.446	0.085 (− 0.091 to 0.256)	0.342
Daytime sleepiness	0.197 (0.003 to 0.377)	0.046	0.088 (− 0.088 to 0.259)	0.329
Total CSHQ score	0.352 (0.169 to 0.511)	< 0.001	0.441 (0.288 to 0.572)	< 0.001

The statistical significance (p < 0.05) was presented in bold

CSHQ Children Sleep Habits Questionnaire, ASQ:SE Ages and Stages Questionnaire: Social–Emotional

Logistic regression analysis of ASQ:SE score risk factors models included the following variables: sex; monthly income; developmental diagnosis; educational status of the parents; sleeping time; total sleep duration; wake time after sleep onset; and subscales of CSHQ. In the TD group male gender, CSHQ subscales: sleep onset delay and sleep duration affected the ASQ:SE score. In the NDD group male gender again, low monthly income, sleep duration more than 11 h reported by parents and CSHQ subscale bedtime resistance had a significant effect on ASQ:SE score (Table 5).

Table 5.

Logistic regression analysis of factors associated with ASQ:SE

Neurodevelopmental delay	OR	95% CI		p
		Lower	Upper
Total sleep duration more than 11 h.*	0.140	0.029	0.672	0.014*
Bedtime resistance (CSHQ subscale)	1.462	1.195	1.788	< 0.001
Family income < 5000TL**	7.123	1.874	27.079	0.004**
Sex (male)	3.339	1.094	10.189	0.034
Constant	0.002			< 0.001
Typical development
Sleep onset delay (CSHQ subscale)	19.815	2.342	167.626	0.006
Sleep duration (CSHQ subscale)	2.499	1.280	4.880	0.007
Sex (male)	18.434	1.151	295.111	0.039
Constant	< 0.001			< 0.001

Models included the following variables: sex; monthly income; developmental diagnosis; educational status of the parents; sleeping time; total sleep duration; waking at night; and subscales of CSHQ

*Compared to those with a sleep duration less and equal to 9 h

**Compared to those with a family income more than 5000TL

Discussion

Sleep and social–emotional development are influenced by child development, environmental, and cultural factors. Several studies have evaluated the effect of sleep parameters on SE problems in children without neurodevelopmental delay [25–29]. But a few studies investigating sleep problems and SE behavior of preschool children with developmental delay in non-Western countries [6, 30]. And also, the previous studies focused on these topics in children with ASD and ADHD [6, 30, 31]. This study investigated sleep problems in children with NDD and age-matched TD, and the relationship between sleep and SE problems in preschool children both with and without developmental delay.

The majority of the NDD group (72.2%) had sleep problems, but there was also a considerable number of sleep problems in children with TD (54.4%) as assessed by the CSHQ. In the literature, the prevalence rates of sleep disturbances in children with NDD and TD children were previously reported to be between 40 and 80% [2, 3, 5, 6, 32] and 20 and 60% [2–4, 6, 32], respectively. The wide ranges of reported prevalence of sleep disturbance in both NDD and TD children may be due to differences in sleep measurements, samples, culture, and environmental factors which have changed the prevalence of sleep problems. The findings in the current study were in the upper range for children with NDD as from some studies in Western and non-Western countries.

Shorter sleep duration was related to higher social–emotional problems than longer sleep duration according to parental reports in TD and NDD groups, consistently with the literature. Several studies have shown that shorter sleep duration was related to social–emotional problems [25, 29, 33] and believed that the prefrontal cortex might play a role in the associations between sleep and social–emotional development [8]. In multivariate analysis, the effect of sleep duration on social–emotional problems was seen only in NDD group.

In the current study, the frequency of night waking did not differ between the groups but a longer duration (≥ 20 min) of waking after sleep onset was more commonly reported in children with NDD, which is again in agreement with some studies [34, 35]. Also, it was shown that children who had longer duration of waking after sleep onset (> 20 min) had higher ASQ-SE scores compared to those who had shorter duration of waking after sleep onset in the NDD group. Night waking is more commonly reported by parents in preschoolers regardless of developmental delay [36, 37]. The longer duration of waking after sleep onset may be associated with having difficulty of self-regulation, or abnormal melatonin in children with NDD [38]. Bedtime resistance was significantly more frequent in the NDD than the TD group in this study. This result was consistent with the literature [3, 6, 30, 37, 39]. Abnormal melatonin synthesis, dysregulation of neurotransmitter systems, comorbidities of NDD, medication use and hypersensitivity to environmental stimuli, having difficulty of self-regulation, restricted and repetitive behaviors may be reasons for this situation [1]. Iron deficiency, D vitamin deficiency due to narrow food preferences [40, 41] and co-sleeping could contribute to the development of bedtime resistance. Although co-sleeping is a cultural habit in our country [42], the parent’s anxiety about the disability of their child and concerns that their child will need care at night may have increased co-sleeping in children with NDD [43]. Parents of children who had bedtime resistance stated that “He doesn’t sleep without me, after he sleeps, I go to my bed.” “She doesn’t sleep alone, she is afraid, “We sleep together in my bed”.

The other sleep problems such as parasomnia [6, 30, 39], sleep anxiety [39], sleep disordered breathing [6], and sleep onset delay [30] were reported more often by parents of children with NDD in some studies. However, in the present study, no significant difference was identified between the two groups concerning these problems and total CSHQ score.

An association between sleep problems and SE problems was observed in the TD and NDD groups with the exception of the ASD subgroup. Social–emotional problems are widely reported in children with ASD. The majority (83.3%) of the children with ASD in our cohort had CSHQ scores > 41 indicating sleep disturbances. So, although ASQ:SE scores were higher in ASD with sleep problems, the difference was not statistically important, possibly because of the small number of children with ASD whose parents did not report sleep disturbances.

Multivariate analysis showed that male gender, low family income, low socio-demographic features, sleep duration less than 11 h, and bedtime resistance were associated with SE problems in children with NDD. In the literature, some studies have also shown that bedtime resistance was associated with emotional problems in children with NDD [5, 6]. In the literature, a recent Chinese study demonstrated that sleep onset delay was associated with SE problems in children with ASD [30], while another Canadian study in ASD children reported that, in addition to sleep onset delay, bedtime resistance, sleep duration, daytime sleepiness, and night waking were all associated with SE problems in the pre-school age group [5]. In multivariate analysis in this study, while no association was identified between the full sleep duration CSHQ subscale and the ASQ:SE score, a total sleep duration more than 11 h as reported by parents was associated with the score of ASQ:SE in the NDD group. We believe this might be due to the structure of the questions, and the knowledge of the parents. One question being categorical and mostly depending on the perception of the family and the other being open-ended and may be more objective. Lower family income in the NDD group was a predictor for SE problems, consistent with the literature [44] although, because of a generally higher family income, this association was not found in the TD group.

This study demonstrated that sleep duration and sleep onset delay were found to be associated with SE problems in the TD group. The previous studies from different countries supported that the association between sleep and emotional problems in children seems to be general. The variation in the link between sleep problems and emotional problems change with study population age, study sample, and culture [45]. Wang et al. demonstrated that pre-school children from two Asian cultures had different sleep problems consequently associated with emotional problems [24]. The previous studies showed that sleep duration, parasomnias, daytime sleepiness were associated with emotional problems in Chinese preschoolers [25]; later sleep time [26, 28], sleep onset delay, sleep anxiety, night waking, parasomnias were associated emotional problems in Japanese preschoolers [25]. The studies made in the Southeastern United States and Finland showed that sleep duration was associated with emotional problems [27, 29]. Our study findings were consistent with some other studies made in either Eastern or Western countries. This can be attributed to the bridge role of our country between both cultures.

There are some limitations of this study that we are aware of. First, as mentioned above, we employed a case–control design for this study. Thus, we were not able to test for the causal directions of effects. Long-term studies would provide further information about the time-course of relations between childhood sleep problems, social–emotional problems, and child development. Second, all measures used were parent-rated questionnaires. Caregiver reports are subjective but provide useful information about many aspects of a child’s sleep. In addition, Souders et al. found that the CSHQ was consistent with actigraphy in children with and without ASD, supporting the reliability of parental reporting [21]. In the age-matched TD group, parent education and percentage of maternal employment were higher than in the NDD group. When adjusting for a range of potential confounders, parent's age, education, and maternal employment did not attenuate the associations between any of the sleep variables and social–emotional problems.

Despite these limitations, the results indicate that behavioral sleep problems, especially bedtime resistance, total sleep duration, and sleep onset delay make an important contribution to emotional dysregulation among pre-school children with and without developmental delay.

Conclusion

The present study suggests that more than half of preschool children with and without developmental delay have sleep problems, which were found to be associated with social–emotional problems according to their caregiver's reports. Pediatricians should screen sleep problems of both children with TD and NDD during health visits and treatment of sleep problems could decrease social–emotional problems of children.

Author contributions

EIB conceived and designed the study; contributed to drafting the article and revising it critically for important intellectual content, PZA, ANC contributed acquisition of data, drafting the article, EK contributed to analysis and interpretation of data; drafted the article, ENO contributed to analysis and interpretation of data; revised it critically for important intellectual content. All authors approved the final manuscript as submitted and agreed to be accountable for all aspects of the work.

Funding

This research did not receive any specific grant from funding agencies in the public, commercial, or not-for-profit sectors.

Declarations

Conflict of interest

No financial or non-financial benefits have been received or will be received from any party related directly or indirectly to the subject of this article.

Ethical approval

The study protocol was approved by Local Ethics Committee of The Hacettepe University Faculty of Medicine.