Eating and sleep problems, related factors, and effects on the mental health of the parents in children with autism spectrum disorder

Abstract

This study aimed to evaluate the frequency of sleep and eating problems in children with autism spectrum disorder (ASD), the factors associated with these problems, and the relationship with the risk of developing depression and anxiety in their parents. In the study, 156 children with ASD and their parents were included. The Brief Autism Mealtime Behavior Inventory (BAMBI), the Children's Sleep Habits Questionnaire (CSHQ), the Hospital Anxiety Depression Scale (HADS) were completed by the parents. The Childhood Autism Rating Scale (CARS) was administered by the researchers to evaluate the severity of ASD in children. The mean age of children was 8.08 ± 2.84, and 87.2% of the children were male. The mean score of CARS was 41.25 ± 6.16, the mean score of BAMBI was 45.67 ± 11.87, and the mean score of CSHQ was 51.67 ± 10.17. According to the CSHQ, 83.3% of the children had sleep problems. The response rate above the cut-off point in the HAD-A subscale was 59.6%, whereas it was 67.3% in the HAD-D subscale. There was a positive correlation between CARS and BAMBI, CSHQ, HAD-A, and HAD-D subscales. There was a positive correlation between BAMBI and CHSQ, HAD-A, and HAD-D subscales. There was a positive correlation between the CSHQ, HAD-A, and HAD-D subscales. There was a negative correlation between the age of the child with ASD, food refusal, and resistance to bedtime. As sleeping and eating problems affect a large portion of children diagnosed with ASD and their caregivers, large population-based studies evaluating both these problems must be designed and carried out to understand factors affecting the prevalence, development, and persistence of sleeping and eating problems and to determine interventions to reduce these issues.

Introduction

Autism Spectrum Disorder (ASD) is characterized by persistent deficiencies in interpersonal interaction and communication, presence of repetitive, restricted, stereotyped behaviors and interests. It is classified under the section of neurodevelopmental disorders in the DSM-5 (APA 2013). Current data report that the diagnosis of ASD is present in one of 44 children. ASD is seen four times more commonly in boys than in girls (Maenner et al. 2018). Restricted, repetitive behavior patterns, which are the diagnostic features of ASD, may lead to being preoccupied with interests with abnormal intensity or adhering strictly to certain routines (Gabriels et al. 2008).

In children with the diagnosis of ASD, besides the core symptoms, compulsive behaviors, eating and sleep problems, which significantly affect the quality of life of the child and the family, are common (Matson et al.2010, Williams et al. 2010). For example, children with ASD experience sleep problems more commonly when compared to normally developing children. While the rate of sleep problems in school-aged children in the general population has been reported as 10.8%, the rate of sleep problems has been shown to vary between 44% and 86% in the studies conducted with children with ASD (Stein et al. 2001, Liu et al. 2006, Richdale and Prior 1995). A relationship between sleep problems and compulsive behaviors in children with ASD is also known (Goldman et al. 2011). In addition, a review evaluating the relationship between autism severity and sleep problems showed a correlation between autism severity and sleep disturbance in 77% of the studies (Hollway and Aman 2011a). On the other hand, sleep-related problems affect the child and the whole family. For example, in a study examining the relationship between a child's sleep and the mother's mental health, it was detected that the presence of sleep problems in a child with ASD was related to the deterioration of the mother's mental health and the increase in the stress level (Hodge et al. 2013).

Eating problems in children with ASD are reported at rates between 46% and 89%. Refusing to eat, not eating home-prepared meals, inappropriate amounts of eating, refusal of new foods, and improper mealtime routine are among the problems frequently expressed by the parents (Ledford and Gast 2006). It has been reported that the prevalence of food selectivity increases as the severity of ASD increases from mild to moderate and severe (Pham et al. 2020). In another study, when parent reports and direct observation were evaluated together, it was found that as the severity of ASD increased, children exhibited more destructive behaviors when unpreferred food was offered (Aponte and Romanczyk 2016). Eating problems in children with ASD may cause the whole family to experience difficulties in accepting appropriate food, maintaining eating routines, taking diet, and interacting with proper mealtimes (Ledford and Gast 2006). Also, a higher stress level was detected in parents who reported eating problems in their children (Rogers et al. 2012, Suarez et al. 2014).

Considering the effects of sleep and eating problems on children diagnosed with ASD and their parents, it is essential to recognize and intervene in the early period. Although there is a high number of studies evaluating eating and sleep problems separately in children with ASD in the literature, the number of studies investigating both together and examining their effects on parental mental health is limited. This study aimed to evaluate the frequency of sleep and eating problems in children with ASD, the factors associated with these problems, and their relationship with the development risk of depression and anxiety in parents.

Methods

Procedure

A total of 156 children, who were diagnosed with ASD in the psychiatric evaluation in the child and adolescent psychiatry outpatient clinic, and their parents were included in this study. Children who had difficulties in oral motor skills, a neurological deficit that may affect nutrition, severe brain damage, a severe medical disorder affecting physical development, and who were under diet due to a metabolic or other medical disease were excluded. Also, children with a diagnosis of an organic disease that may cause sleep disorder (such as obstructive sleep apnea) and who have a specific drug use for a sleep disorder (like melatonin) were excluded from the study. Furthermore, among the children included in the study, nobody was supported because of sleep or eating disorder before or during the study.

Upon their admission to the outpatient clinic, parents filled out the sociodemographic data form, the Brief Autism Mealtime Behavior Inventory (BAMBI) to assess children's eating problems, the Children's Sleep Habits Questionnaire (CSHQ) to assess children's sleep problems, and the Hospital Anxiety Depression Scale (HADS) to determine the risk of anxiety and depression in parents. The Childhood Autism Rating Scale (CARS) was applied by the researchers during the interview to evaluate the severity of ASD in children. Data collection for the study started in June 2021 and was completed in December 2021. The patients who presented to the hospital where the study was performed mainly came from urban rather than rural areas and did not have significant cultural differences.

Evaluation tools

Sociodemographic data form

In the form prepared by the researchers, the information such as the monthly income level of the family, the education level of the parents, the speech level of the child (whether they could form a sentence), drug use, the eating characteristics of the child, and the history of mental diseases in the family were evaluated.

Childhood autism rating scale (CARS)

This scale, developed by Schopler et al. allows for determining the severity of autism. Each item is graded between 1-4 with a half-degree point. While the total score of the scale varies between 15-60, a score of 30-36.5 indicates mild-moderate autism, and a score of 37-60 indicates severe autism (Schopler et al. 2007)

The brief autism mealtime behavior inventory (BAMBI)

This scale is developed by Lukens to evaluate mealtime and eating behaviors of children with ASD. It has three subscales: food refusal, limited food variety, and behavioral characteristics related to autism. It consists of 18 items evaluating the nutritional problems observed in individuals diagnosed with ASD and mental disability. A high total score indicates the severity of nutritional problems (Lukens 2005).

Children's sleep habits questionnaire (CSHQ)

The scale consists of 33 items, which aims to investigate children's sleep habits and sleep-related problems. It has eight subscales that screen children's sleep disorders such as resistance to bedtime, difficulty in falling asleep, sleep duration, sleep anxiety, night waking, parasomnia, sleep problems related to respiration, and daytime sleepiness. The total test score is calculated by summing the scores obtained from the items. Values ​​above the cut-off point of 41 are considered clinically significant (Owens et al. 2000).

Hospital anxiety and depression scale (HADS)

It was developed to determine the risk of anxiety and depression in adults and to follow the level and change in the severity of symptoms. It consists of two subscales: Anxiety (HAD-A) and Depression (HAD-D) (Zigmond and Snaith 1983). The cut-off points were determined as 10 for the anxiety subscale and 7 for the depression subscale. Scores above the cut-off point are considered at risk for anxiety and depression (Aydemir et al. 1997).

Data analysis

The data were recorded and analyzed with the SPSS 25.0. In statistical analyzes, data obtained by measurement were presented as arithmetic mean ± standard deviation, and data obtained by counting were shown as a percentage (%). To evaluate whether the numerical variables fit the normal distribution, the Kolmogorov Smirnov test was performed. Whether the variables showed a statistically significant difference between the groups was evaluated with the Mann-Whitney U test. Categorical variables were compared with the Pearson chi-square test. The relationship between continuous variables was evaluated with the Spearman correlation test. For the results, p < 0.05 was considered statistically significant.

Results

Of the children diagnosed with ASD, 87.2% (n = 136) were male, 12.8% (n = 20) were female. The mean age of all children was 8.08 ± 2.84 years, and the age range was 2-12. During the study, 84% (n = 131) of the participants' mothers filled out the forms. The mean age of the mothers was 36.57 ± 5.09 years, and the mean age of the fathers was 40.91 ± 5.59 years. Regarding education level, 55.1% (n:86) of the mothers did not complete high school. The income level of 38.5% of the families was equal to or below the minimum wage. There was no sentence formation in 60.3% (n: 94) of the children, and 82.1% (n:128) had at least one sibling (Table 1).

Table 1.

Sociodemographic features.

		n (%)
Child age (mean ± sd)		8.08 (±2.84)
Child gender	Female	20 (12.8)
	Male	136 (87.2)
Parent’s age (mean ± sd)	Mother	36.57 (±5.09)
	Father	40.91(±5.59)
Mother education status	Did not complete high school	86 (55.1)
	Completed high school	70 (44.9)
Which parent filled the form?	Mother	131 (84)
	Father	25 (16.0)
Family’s monthly income	Minimum wage or lower	60(38.5)
	Higher than minimum wage	96(61.5)
Does the child have a sibling?	Yes	128 (82.1)
	No	28 (17.9)
Speech level of the child	No sentence formation	94 (60.3)
	Forms sentences	62 (39.8)
Concomitant medical illness in the child	Yes	30 (19.2)
	No	126 (80.8)
Psychiatric drug use of the child	Yes	65 (41.7)
	No	91 (58.3)
Sensitivity for food (Color, smell,or texture)?	Yes	88 (56.4)
	No	68 (43.6)
How does the child eat food?	By him/herself	71 (45.5)
	With the parents’ help	85 (54.5)
Does the food of the child have a process (like a blender )?	Yes	15 (9.6)
	No	141 (90.4)
Is any thing used while eating?	Yes (television, tablet, telephone, etc.)	109 (69.9)
	No	47 (30.1)

The mean score of CARS was 41.25 ± 6.16. According to CARS, 25.6% (n:40) of the children showed mild-moderate, and 74.4% (n:116) showed severe ASD symptoms. The mean of all children's BAMBI total scores was 45.67 ± 11.87. The mean score of CSHQ was 51.67 ± 10.17 in all children, and when the children above the cut-off point of 41 were considered, sleep problems were found in 83.3% (n: 130) of the children (Table 2).

Table 2.

CARS, BAMBI total scale and subscale scores, CSHQ total scale and subscale scores, HAD-A and HAD-D subscale scores.

Scales		Scale scores Mean ± SD
CARS		41.25 (±6.16)
BAMBI total score		45.67 (±11.87)
BAMBI limited variability		23.51 (±6.34)
BAMBI food refusal		10.60 (±4.37)
BAMBI autism-specific behavior features		11.54 (±3.69)
CSHQ total score		51.67 (±10.17)
CSHQ resistance to bedtime		12.03 (±3.22)
CSHQ difficulty in falling asleep		1.92 (±0.82)
CSHQ sleep duration		4.98 (±1.86)
CSHQ sleep anxiety		7.97 (±2.57)
CSHQ night waking		5.03 (±1.62)
CSHQ parasomnia		10.18 (±2.75)
CSHQ respiration related problems		4.07 (±1.32)
CSHQ daytime sleepiness		9.74 (±2.87)
HAD-A		10.83 (±4.50)
HAD-D		9.16 (±4.04)
HAD-A parental anxiety risk according to cut-off level n (%)	Yes	93 (59.6)
	No	63 (40.4)
HAD-D parental depression risk according to cut-off level n (%)	Yes	105 (67.3)
	No	51 (32.7)

In the study, we also examined the relationships of BAMBI, CSHQ, HAD-A, and HAD-D subscales with parameters such as the mother's education level, the monthly income level of the family, the presence of siblings, and sentence formation in the child. The children of families with a monthly income equal to or below the minimum wage had significantly higher mean scores on BAMBI (p:<0.001) and CSHQ (p:0.027). No significant differences were found for parameters such as the mother's education level, the presence of siblings, and sentence formation in the child. In addition, to examine how the risk of sleep problems, eating problems, anxiety, and depression in children progressed according to the developmental periods of the children, the children were divided into two groups as early childhood (2-6 years of age) and late childhood (6-12 years of age) and analyzed. There was no statistically significant difference between the two groups for CSHQ (p:0.166), BAMBI (p:0.352), HAD-A (p:0.724), and HAD-D (p:0.846) levels (Table 3)

Table 3.

Comparison of CSHQ, BAMBI, HAD-A and HAD-D subscales regarding the sociodemographic features.

	Mother’s education level			Families’ monthly income level			Sentence formation of the child			Does the child have a sibling?			Developmental periods of the children
	Did not complete high school	Completed high school	p*	Minimum wage or lower	Higher than minimum wage	p*	Yes	No	p*	Yes	No	p*	early childhood	late childhood	p*
BAMBI	46.25(±11.46)	44.95 (±12.39)	0.388	48.28 (±11.24)	46.63 (±11.53)	0.000	45.17 (±12.77)	46.00 (±11.29)	0.461	46.14 (±12.11)	43.53 (±10.64)	0.302	47.12 (±12.47)	44.88 (±11.51)	0.352
CSHQ	52.13 (±9.96)	51.11 (±10.47)	0.425	53.56 (±11.02)	51.79 (±9.23)	0.027	50.82 (±12.41)	52.24 (±10.02)	0.452	51.77 (±10.49)	51.25 (±8.69)	0.826	53.29 (±10.06)	50.80 (±10.17)	0.166
HAD-A	10.91 (±4.38)	10.74 (±4.68)	0.885	11.50 (±4.64)	10.83 (±4.11)	0.163	10.64 (±4.07)	10.96 (±4.78)	0.658	10.93 (±4.65)	10.39 (±3.81)	0.440	11.01 (±4.84)	10.74 (±4.33)	0.724
HAD-D	9.60 (±3.91)	9.62 (±4.17)	0.162	10.26 (±3.93)	8.77 (±3.95)	0.021	9.01 (±3.97)	9.26 (±4.11)	0.623	9.09 (±4.21)	9.50 (±3.21)	0.649	9.00 (±3.70)	9.25 (±4.24)	0.846

^*Mann Whitney U Test.

The mean score of the HAD-A subscale was 10.83 ± 4.50, and the mean score of the HAD-D subscale was 9.16 ± 4.04, which was filled by the parents. The response rate above the cut-off point in the HAD-A subscale was 59.6% (n:93), and in the HAD-D subscale was 67.3% (n:105). Although the HAD-A subscale mean score was higher in mothers who did not completed high school, in families with a monthly income equal to or below the minimum wage, in the presence of siblings, and the absence of sentence formation, these differences were not statistically significant. The HAD-D subscale was significantly higher in families with an income below minimum wage (p:0.021). In the HAD-D subscale, there were no significant differences in other variables except the income level (Table 3). Also, in parents with scores higher than the cut-off levels in HAD-A and HAD-D subscales, we examined whether maternal education level, monthly income level, some sociodemographic variables, and sleep and eating problems in children differed. Parents with a mean score above the cut-off value in the HAD-A subscale had significantly higher rates of sleep problems in children (p:0.001), CARS severity (p:0.042), and BAMBI mean scores (p:0.001). While in parents with a mean score above the cut-off value in the HAD-D subscale, the severity of CARS (p:0.004) and the presence of sleep problems (p:0.003) were higher. In addition, the risk of depression was lower in parents who reported income levels above the minimum wage (p:0.008) (Table 4).

Table 4.

Comparison of the risk of depression and anxiety in the HAD-A and HAD-D subscales in parents according to sociodemographic variables, CARS, BAMBI and CSHQ.

		HAD-A parental anxiety risk according to cut-off level		P	HAD-D parental depression risk according to cut-off level		P
		Yes n(%)	No n(%)		Yes n(%)	No n(%)
Developmental periods of the children	Early childhood	35 (37.6)	20 (31.7)	0.450*	36 (34.3)	19 (37.3)	0.716*
	Late childhood	58 (62.4)	43 (68.3)		69 (65.7)	32 (62.7)
Does the child have a sibling?	Yes	75 (80.6)	53 (84.1)	0.578*	83 (79.0)	45 (88.2)	0.161*
	No	18 (19.4)	10(15.9)		22 (21.0)	6 (11.8)
Mother’s education status	Did not complete high school	51 (54.8)	35 (55.6)	0.930*	63 (60)	23 (45.1)	0.079*
	Completed high school	42 (45.2)	28 (44.4)		42 (40)	28 (54.9)
Family’s monthly income	Minimum wage or lower	37 (39.8)	23 (36.5)	0.680*	48 (45.7)	12 (23.5)	0.008*
	Higher than minimum wage	56 (60.2)	40 (63.5)		57 (54.3)	39 (76.5)
Sentence formation of the child	No	56 (60.2)	38 (60.3)	0.990*	65 (61.9)	29 (56.9)	0.546*
	Yes	37 (39.8)	25 (39.7)		40 (38.1)	22 (43.1)
Sensitivity for food (Color, smell,or texture)?	Yes	54 (58.1)	34 (54.0)	0.613*	58 (55.2)	30 (58.8)	0.672*
	No	39 (41.9)	29 (46.0)		47 (44.8)	21 (41.2)
Does the child have sleep problems?	Yes	85 (91.4)	45 (71.4)	0.001	94 (89.5)	36 (70.6)	0.003*
	No	8 (8.6)	18 (28.6)		11 (10.5)	15 (29.4)
CARS		42.18 (±5.74)	39.88 (±6.53)	0.042**	42.22 (±6.00)	39.26 (±6.06)	0.004**
Mother’s age		36.04 (5.15)	37.36 (±4.93)	0.106**	36.53 (±5.28)	36.66 (±4.71)	0.883**
Father’s age		40.33 (±5.74)	41.77 (±5.29)	0.093**	41.01 (±5.94)	40.70 (±4.84)	0.813**
BAMBI total score		48.22 (±11.93)	41.90 (±10.80)	0.001**	46.80 (±11.79)	43.33 (±11.78)	0.071**

^*Chi Square Tests.

^**Mann Whitney U Test.

The relationships between CARS, BAMBI and its subscales, CSHQ and its subscales, HAD-A, HAD-D subscales, and the child's age were examined. There was a statistically significant positive correlation between CARS and BAMBI total score (r:0.306, p < 0.01), food refusal (r:0.342, p < 0.01) and autism-specific behavioral characteristics (r:0.450, p < 0.01) subscales, CSHQ total score (r: 0.307, p < 0.01), resistance to bedtime (r:0.293, p < 0.01), difficulty in falling asleep (r:0.213, p < 0.01), sleep duration (r:0.230, p < 0.01), and night waking (r:0.187, p < 0.05) subscales, HAD-A (r:0.181, p < 0.05) and HAD-D subscales (r:0.188, p < 0.05) (Table 5).

Table 5.

Correlations between CARS, BAMBI, CSHQ, HAD-A, HAD-D and child’sage.

	CARS	Child’s age	BAMBI total score	Limited variability	Food refusal	Autism-specific behavior characteristics	CSHQ total score	Resistance to bedtime	Difficulty in fallingasleep	Sleep duration	Night waking	Daytime sleepiness	HAD-A	HAD-D
CARS
Child’sage	−0,050 (0.534)
BAMBI total score	0.306**	0.128 (0.110)
Limited variability	0.128 (0.110)	−0.062 (0.442)	0.848**
Food refusal	0.342**	−0.218**	0.821**	0.489**
Autism-specific behavior characteristics	0.450**	−0.008 (0.924)	0.739**	0.404**	0.611**
CSHQ total score	0.307**	−0.109 (0.175)	0.573**	0.440**	0.508**	0.450**
Resistance to bedtime	0.293**	−0.205*	0.358**	0.277**	0.345**	0.232**	0.669**
Difficulty in falling asleep	0.213**	0.073 (0.362)	0.228**	0.164*	0.202*	0.191*	0.471**	0.392**
Sleep duration	0.230**	−0.018 (0.828)	0.207**	0.099 (0.218)	0.223**	0.235**	0.597**	0.334**	0.415**
Nightwaking	0.187*	−0.157 (0.050)	0.341**	0.195*	0.264**	0.308**	0.624**	0.384**	0.165*	0.436**
Daytime sleepiness	0.135 (0.094)	−0.089 (0.270)	0.364**	0.271**	0.370**	0.236**	0.591**	0.304**	0.208**	0.174*	0.216**
HAD-A	0.181*	−0.050 (0.538)	0.327**	0.247**	0.266**	0.301**	0.407**	0.216**	0.209**	0.204*	0.289**	0.264**
HAD-D	0.188*	−0.034 (0.677)	0.235**	0.119 (0.140)	0.212**	0.276**	0.237**	0.168*	0.113 (0.159)	0.159*	0.208**	0.145 (0.072)	0.647**

Spearman's correlationanalysis, *p < 0.05 statistically significant, **p < 0.01 statistically significant.

There was a statistically significant positive correlation between the BAMBI total score and CSHQ total score (r:0.573, p < 0.01), resistance to bedtime (r:0.358, p < 0.01), difficulty in falling asleep (r:0.228, p < 0.01), sleep duration (r:0.207, p < 0.01) p < 0.01), night waking (r:0.341, p < 0.05) and daytime sleepiness (r:0.364, p < 0.05) subscales, as well as HAD-A (r:0.327, p < 0.01) and HAD-D (r:0.235, p < 0.01) subscales (Table 5).

A statistically significant positive correlation was also observed between the CSHQ total score and food refusal (r:0.508, p < 0.01), autism-specific behavioral characteristics (r:0.450, p < 0.01), limited variability (r:0.440, p < 0.01) subscales, as well as HAD-A (r:0.407, p < 0.01) and HAD-D subscales (r:0.237, p < 0.01) (Table 5).

There was a negative correlation between the age of the child with ASD, food refusal (r:-0.218, p < 0.01), and resistance to bedtime (r:-0.205, p < 0.05) (Table 5).

Discussion

In the current study, we investigated the frequency of sleep and eating problems, the relationship of these problems with the severity of the ASD, anxiety and depression symptom signs in parents, and other variables in children with ASD without any other diseases that may affect sleep and nutrition. Raising a child with ASD causes different challenges to the parents. These children may present with a wide variety of symptoms such as abnormal reactions to sensory stimuli, hyperactivity, tantrums, self-harming behaviors, abnormal eating patterns, and sleep abnormalities (APA 2000).

In children diagnosed with ASD, sleep problems not only impair the child's quality of life but also cause stress to other family members. In our study, sleep problems were detected in 83.3% of the children, according to the CSHQ. Parental reports indicate that 15% to 35% of typically developing children have sleep problems (Mindell 1993). This rate has been shown to vary between 44% and 83% in children with ASD (Richdale 1999). In a study in which sleep problems and related factors were evaluated in children with ASD, it was determined that 86% had sleep problems, 54% had sleep resistance, 56% had insomnia, and 31% had daytime sleepiness (Liu et al. 2006). Regarding the literature, it is seen that the rate of sleep problems in our study is in parallel with the previously reported studies. In our study, there was also a significant positive correlation between CSHQ and CARS. In a study investigating sleep problems in children with ASD, a negative correlation was found between the decrease in sleep duration and the severity of CARS (Devnani and Hegde 2015). Mayes and Calhoun (2009) reported that the severity of autism was the strongest predictor of sleep problems in children, and it explained 20% of the variance. In our study, the mean total score of CSHQ was significantly higher in the children of families with a monthly income below the minimum wage. Also, in another study investigating the risk factors associated with sleep problems in children with ASD, a relationship was reported between the poor quality of life and sleep problems (Hollway et al. 2013). The decrease in family income may cause a decrease in the quality of life, increased stress for all family members, and sleep problems in the child.

Another finding in our study was that a significant positive correlation was found between CSHQ and HAD-A and HAD-D subscales. The presence of sleep problems in the child can also be a source of anxiety for the parents. The relationship between sleep problems and parenting stress has been shown in typically developing children and children with mental disabilities (Meltzer and Mindell 2007, Quine 2008). In a study in 2011, actigraphy was used to investigate the factors associated with depressive symptoms in parents of a child diagnosed with ASD. A relationship was found between depressive symptoms in the mother and shorter sleep duration and sleep disorders in the child. Also, poor sleep quality in the child predicted the presence of depressive symptoms in the father (Meltzer 2011). It has been reported that the relationship between children's sleep problems and parents' attitudes and behaviors is bidirectional. Negative attitudes of the parents toward the child may lead the sleep problems in the child to consolidate (Sadeh and Anders 1993). Studies in the literature report that the parent's education effectively reduces sleep problems in children with ASD (Johnson et al. 2015, Johnson et al. 2013). Considering the prevalence of sleep problems and their effects on children and parents, it seems that the evaluation for ASD should also include questions about sleep problems. When sleep problems are detected in the child, it is essential to plan appropriate psychopharmacological treatment or behavioral interventions and develop parent education programs that involve sleep problems in the child.

Our study showed a significant positive correlation between BAMBI total score and CARS or CSHQ. In a study in 2019, atypical eating behaviors such as limited food preferences and brand-specific preferences in a study sample of 2102 children were found to be 4.8% in the group without any diagnosis, 13.1% in the group with non-ASD developmental problems, and 70.4% in children diagnosed with ASD (Mayes and Zickgraf 2019). Parents who state that their children have eating problems also report that they show higher levels of autism-related symptoms, sleep problems, internalization, and externalization problems (Allen et al. 2015). Another study showed that there was a relationship between eating problems and other destructive behaviors, and a decrease in problems with eating and mealtime contributes to a reduction in destructive behaviors (Johnson et al.2014). Yang et al. (2018) found that selective eating in ASD was associated with sleep disorders. Considering our study findings and literature data, it is seen that sleep and eating problems affect each other in children diagnosed with ASD. Besides, in our study, BAMBI was significantly higher in the children of families with a monthly income below the minimum wage. Previous research has shown that the low-income level may be associated with various eating problems in children (Janssen et al. 2006). Nevertheless, another study reported that income level was not influential in eating problems in children with ASD (Seiverling et al. 2018). Considering that a significant number of children with ASD are selective for food, it can be commented that the increase in the family's income level provides an increase in the variety of food that parents can offer to their children.

Another finding in our study is a significant positive correlation between BAMBI and the HAD-A and HAD-D subscales. It has been reported that eating problems in children diagnosed with ASD cause stress in parents and negatively affect the family's social life (Johnson et al. 2014). Teaching parents strategies to reduce eating problems and destructive mealtime behaviors has been shown to minimize parenting stress (Thullen and Bonsall 2017). Effective results can be obtained with early behavioral interventions in treating eating problems, like other behavioral problems. Applied Behavior Analysis, a commonly used intervention method frequently used in the management of behavioral problems in children diagnosed with ASD, has been shown to be an effective method that can also be used in the treatment of eating problems (Matson et al. 2009). When our findings and the literature are considered, there is a strong possibility that early intervention for eating problems could have a positive effect on the well-being of both ASD children and their parents.

In our study, a significant negative correlation was found between the age of the child diagnosed with ASD and food refusal (r:-0.218, p < 0.01) and resistance to bedtime (r:-0.205, p < 0.05). In a study in 2013, it was found that as children with ASD grew up, there was a general decrease in food selectivity (Beighley et al. 2013). Another study investigated the change in food selectivity in 6-year intervals in children with ASD. While a decrease was detected in food refusal over time, it was observed that the variety of food eaten did not increase. The decrease in food selectivity has been attributed to the fact that, over time, parents less often offer food their children do not prefer (Bandini et al. 2017). Also, in our study, it is thought that the decrease in food refusal in older children may be related to the change in parents' attitudes. In a study in which 1856 children between the ages of 3-18 diagnosed with ASD were evaluated using CSHQ, the rate of bedtime resistance was higher in the smaller age group than in adolescents (Goldman et al. 2012). This condition suggests that compliance with sleep-related routines such as bedtime may increase with age, as well as with the contribution of education.

In our study, the response rate above the cut-off point in HADS was 59.6% for HAD-A and 67.3% for HAD-D. There was a significant positive correlation between HAD-A and HAD-D subscales and CARS. Also, the mean CARS score was higher in the parents with a mean score above the cut-off value in the HAD-A and HAD-D subscales compared to the parents with a mean score below the cut-off value. Studies are showing that emotional and behavioral problems in children are more closely related to family routines and parents’ mental health rather than the severity of ASD symptoms (McStay et al. 2014b, Pozo et al. 2014, Vasilopoulou and Nisbet 2016) In a study conducted with parents of children diagnosed with ASD, it was found that more than 80% felt stressed (Bitsika et al. 2013). Hastings (2002) stated that problematic child behaviors cause parenting stress and that parents under stress present behaviors that can reinforce the child's behavioral problems. This negatively affects their ability to parent effectively and manage the special needs of their children (Ludlow et al. 2012, Myers et al. 2009). Also, our study detected a higher risk of depression in parents with low income. It was found that social isolation, financial concerns, and care demands could be among the causes of stress experienced by families with a child diagnosed with ASD (Myers et al. 2009). It was stated that social support, mainly provided by friends and family, could reduce the stress that can be caused by raising a child diagnosed with ASD (Ekas et al. 2010). The parents with a child diagnosed with ASD cope not only with their child's sleep and eating problems but also with many difficulties that can increase the risk of anxiety and depression in parents. Considering the impact of parental well-being on the child diagnosed with ASD, evaluation of parental mental health and, when necessary, providing a treatment plan during the follow-up process has critical importance.

The study has several limitations. First, parental reports were considered the only source of information about the child's sleep and eating problems and related factors. The possibility that parents may underestimate or exaggerate their children's sleep or eating problems should not be ignored. Also, the parent's recall bias may influence the association between eating or sleeping problems. Another limitation is that the study did not include comparison groups such as typically developing children and/or children with other special needs. Third, our study is cross-sectional and does not provide information about the change in sleep or eating problems over time.

In conclusion, there appears to be significant evidence that sleep and eating problems are common in children with ASD. Research on sleep and eating problems in ASD is constantly expanding, however, the number of studies on the coexistence of sleep and eating problems is limited. Our study findings show that the rate of sleep and eating problems is high in children diagnosed with ASD. Besides, a statistically significant relationship was found between the presence of sleep and eating disorders in children and parents' risk of anxiety and depression. As sleeping and eating problems affect a large portion of children diagnosed with ASD, extensive population-based studies evaluating both these problems must be designed and carried out to understand factors affecting the prevalence, development, and persistence of sleeping and eating problems and to determine interventions to reduce these issues.

Disclosure statement

No potential conflict of interest was reported by the authors.

Funding

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