Autism in Turkey: demographics, behavior problems, and accompanying medical conditions in a sample of Turkish youth with autism spectrum disorder

Abstract

Autism spectrum disorder (ASD) is an etiologically heterogeneous neurodevelopmental condition that eludes a single explanation or cure. Epidemiological studies reveal risk factors, relevant comorbidities, and behavioral correlates to reach a better understanding of ASD. To contribute such data from an understudied ASD population, this paper presents epidemiological data from a Turkish sample of individuals with ASD (n = 911, 748 boys (82.1%) and 163 girls (17.9%) between 1 and 18 years of age). Average age at diagnosis was 31.06 ± 11.88 months, and the male-to-female ratio was 4.6:1. Three in 4 individuals with ASD had obsessive behaviors, and 1 in 4 had allergic conditions, inappropriate sexual behaviors, self-harming behaviors, and harmful behaviors towards others. One in 3 received a dietary treatment for at least 3 months; almost half received vitamin supplements; the majority (70%) did not experience constipation; and 2 in 3 were picky eaters. This paper presents data on the age of diagnosis, gender ratios, accompanying behaviors, and dietary interventions in Turkish individuals with ASD, which are topics of current research interest about ASD. Such data from non-Western populations may supplement epidemiological knowledge gained from Western populations to help reach a more comprehensive understanding of this condition with many unknowns.

Introduction

Autism spectrum disorder (ASD), a condition associated with social, communication, and behavioral challenges, affects 1 in 54 individuals according to the most current estimates from the US (Maenner et al., 2020). Though both genetic and environmental factors are implicated, neither a clear etiology nor a cure is available for ASD. Therefore, information on how this condition manifests and its associations with other medical and behavioral indices are important to unravel the many unknowns of ASD, which can be obtained via epidemiological investigations.

Epidemiological studies on ASD have started in the UK in the mid-1960s (Lotter 1966) and most epidemiological and prevalence studies come from Western countries. Therefore, findings from non-Western countries constitute a valuable contribution to the field and to the purpose of reaching a global, inclusive, and comprehensive understanding of ASD. Though the most up-to-date ASD prevalence data usually comes from the US with their multi-site investigations across the country (Baio et al., 2018, Maenner et al., 2020), studies from other countries provide information about the prevalence of ASD and in some cases demographic, clinical, and medical characterizations of individuals with ASD. Some examples are the French epidemiological study (Fombonne et al. 1997), prevalence data from Israel (Davidovitch et al. 2013, Gal et al., 2012, Raz et al. 2015), prevalence and age at diagnosis data from Denmark (Parner et al. 2008) and comparisons with Western Australia (Parner et al. 2011), prevalence and characteristics of adults with ASD in England (Brugha et al. 2011), epidemiological data of a paediatric population in Portugal (Oliveira et al. 2007), and the more recent epidemiological information about the ASD cases in Mali (Traoré et al. 2018). Italians made a call for an epidemiological surveillance system for ASD cases in Italy, whose first step they consider should be defining a reliable estimate of ASD occurrence in Italy (Mazzucco et al. 2016). As an initial step to understanding such information about ASD in Turkey, this study presents epidemiological data from Turkish youth with ASD.

We have gathered topics of research curiosity about the ASD population from different domains that are usually investigated separately and explored them collectively in a sample from the Turkish population with ASD. We investigated childbirth-related factors, male-to-female ratio, age of diagnosis, accompanying medical and behavioral conditions; and nutritional and dietary status, constipation, and selective eating, where the former are traditional items of interest, and the latter receive traditional as well as a renewed attention due to the recent focus on the gut-brain axis in psychiatry and neurology.

Method

Participants

Parents of 911 Turkish individuals with ASD aged between 1 and 18 years participated in this study; therefore, the study included infants, toddlers, children, and adolescents with ASD. All demographic characteristics of the participants are presented in Table 1 in detail.

Table 1.

Sociodemographic characteristics of participants (n = 911).

Variable	Category	Frequency (Percent)
Gender	Male	748 (82.1)
	Female	163 (17.9)
Age	1–3 years	151 (16.6)
	4–6 years	328 (36.0)
	7–9 years	226 (24.8)
	10–12 years	94 (10.3)
	13–15 years	39 (4.3)
	16–18 years	73 (8.0)
Age at diagnosis	0–18 months	120 (13.2)
	19–24 months	218 (23.9)
	25–36 months	390 (42.8)
	> 36 months	183 (20.1)
Birth week	<28 weeks	8 (0.9)
	28–32 weeks	41 (4.5)
	32–36 weeks	123 (13.5)
	36–40> weeks	739 (81.1)
Birth type	Normal	295 (32.4)
	Cesarean	616 (67.6)
Birth order	First birth	530 (58.2)
	Second birth	259 (28.4)
	Third birth	93 (10.2)
	Others	29 (3.2)
Staying at NICU	Yes	173 (19.0)
	No	738 (81.0)

NICU, neonatal intensive care unit

The inclusion criteria of the study were to be between the ages of 1 and 18 years and to be diagnosed with ASD by a child psychiatrist. The exclusion criterion was being above 18 years of age, and 8 participants whose parents completed the questionnaire were excluded from the sample because they were older than 18. The satisfaction of the criterion of having ASD was confirmed by only recruiting parents from the ‘Autism Brotherhood’ social network whose members are families of individuals with ASD, announcing the title of survey as an investigation of the general characteristics of children with ASD living in Turkey, and specifying at the beginning of the survey that if parents have more than one child with ASD, they should fill a separate form for each one. Moreover, all the questions in the survey started as ‘Does your child with autism spectrum disorder…’ Therefore, all parents responded to the questions about their child with an ASD diagnosis. Additionally, 93.9% of the individuals with ASD in this sample had a report from Turkish Guidance and Research Center (RAM), provided by state officials connected to the Ministry of Education who complete developmental evaluations to determine developmental delay and confirm the presence of a disability. The rest of the participants presumably did not receive this official report to avoid recording a disorder in the child's personal record. Therefore, not having this report did not constitute an exclusion criterion, since this report shows disability severity and we have confirmed the inclusion criterion of having an ASD diagnosis with the aforementioned measures. In sum, all individuals whose parents completed the survey had an ASD diagnosis and the majority additionally had an official disability report.

Procedure

Survey method was used to collect data. An internet survey was shared with parents of individuals with ASD via an autism network, the Facebook group called ‘Autism Brotherhood’, which includes parents of individuals with ASD from all over Turkey. Before starting the survey, their consent was obtained. It was conveyed at the onset of the survey that only parents of individuals with ASD, who were diagnosed with ASD by a child psychiatrist, could fill the survey. Ethical approval was received from the Institutional Review Board of Istinye University (meeting number 2020/8, decision number 03). Descriptive statistical methods, independent samples t-test, Pearson chi-square test, point biserial correlation, and tetrachoric correlation were used to describe and analyze the data.

Materials

The online survey used in this study was developed by the authors based on their experiences in the clinic with individuals with ASD. The total of 27 questions in this survey can be considered under five main groups.

The first eight questions were related to the sociodemographic characteristics of individuals with ASD. The age of the individual with ASD was entered manually and other questions were answered as multiple-choice responses. There were two response choices for gender (male or female). The question that probed how they would define the area that their child with ASD lived had four choices of metropolitan city, city, town, and village. There were two response options for the question about birth type that were vaginal and caesarean, four choices for birth week that were being born before 28 weeks of gestation, born between 28 and 32 weeks, born after 32 weeks, and born between 36 and 40 weeks of pregnancy, and four choices for birth order which were first, second, third, and other. The sixth question was about whether their child had stayed at NICU as a newborn and the question had two choices of yes or no. Then, the parents were asked if their child had a RAM disability report, and if so, in the next question, they were asked to manually enter the disability percentage on that report. The RAM report from the government documents the disability level of children from all disability groups including ASD. Therefore, in this study, it was used as a formal report of ASD severity of participants.

The subsequent questions were about the accompanying medical conditions in individuals with ASD. There were three questions about seizures. First, they were asked if their child with ASD ever experienced any feverish or feverless seizure. Then, they were asked if their child with ASD had an electroencephalogram (EEG) taken. After that, they were asked if their child with ASD had an EEG, were any problems detected in their brain waves. To this question, they could have responded as yes, no, or no EEG was taken. The question about allergies only probed whether their child with ASD had any allergic conditions, to which parents responded as yes or no.

The following questions investigated behavioral problems of individuals with ASD. All questions were replied as multiple-choice responses, and the question asked ‘Does your child with autism spectrum disorder have (problem behavior/condition)?’ and the question was answered as yes or no. The first question of this part was about the obsessive behaviors of the children with ASD. The question asked if their child with ASD has obsessive behaviors. They had only two options for the parents to reply to this question, which were yes or no. In the next part of this section, there was one question about the presence of inappropriate sexual behaviors of children. Its response choices were similarly yes or no. The two next questions were also in the aforementioned format and it asked about the harmful behavior of the children to themselves and the others. One asked if their child with ASD has self-harming behaviors, and the other asked if their child with ASD has harming behaviors towards others, where both questions were responded to as yes or no. Finally, there was a question s about the presence of inappropriate sexual behaviors in their children with ASD. This question asked if their child with ASD had inappropriate sexual behaviors such as masturbation and was replied as yes or no.

Next, psychiatric medication use was investigated. First, parents were asked if their child with ASD ever used a psychiatric medicine, to which they responded yes or no. Then, they were asked of the psychiatric medications their child with ASD used, from which one they benefited the most. When inquiring about the psychiatric medicines, six options were provided to choose from based on the authors’ clinical experience of mainly utilized medications in the Turkish ASD population, which were Risperdal (Risperidone), Abilify (Aripiprazole), Vayarin (Lipirinen), Cortexin, none, and other. Risperidone, Aripiprazole, Lipirinen, and Cortexin are some of the most commonly prescribed medications by psychiatrists and neurologists for conditions where individuals experience attention and regulation problems such as attention deficit and hyperactivity disorder, attention problems, anxiety disorders, and ASD. These medications are utilized to alleviate the detrimental effects of severe symptoms on the lives of individuals with ASD and are used to treat the irritability, explosive, and aggressive behaviors that can accompany ASD. Specifically, Risperidone and Aripiprazole are prescribed for behavior problems such as biting, hitting, throwing that harms oneself or others, masturbation, anxiety, and stereotypical behaviors. Lipirinen is given for attention deficit and hyperactivity problems. Cortexin is a cognitive supplement used in cases where there are accompanying cognitive delays.

The final part of the survey had five questions and they were related to nutritional and dietary status, constipation, and selective eating of children with ASD. The first question asked if their child with ASD received a special dietary treatment at any time for at least three months, which was answered as yes or not, and the next question probed if their child with ASD benefited from this dietary treatment to which they answered as yes or no. Subsequent questions probed whether their child with ASD received any vitamin supplement and its choices were yes and no. If the parents replied yes to this question, they proceeded to the next question. Second question was asked among the vitamin supplements their child with ASD used, from which of the below they experienced the most benefits. Again, according to the authors’ clinical experiences, various options for replying to this question were provided, which were vitamin D, B6, B12, TMG, none (this option was provided since all participants saw this question and not only those who responded yes to the previous question), and other. Four other choices were also provided that are commonly used as supplements in the Turkish ASD population though they are not vitamins, which are fish oil, probiotics, TMG (a vitamin-like substance used to support neurological development of children with ASD), and quercetin. The other similar question about the feeding process was about whether their child with ASD used a special nutritional supplement such as coconut oil and garlic cure, to which they responded as yes or no. The last two questions were about constipation and their selective eating. The parents were asked if their child with ASD experienced a constipation problem, and a problem in selecting foods. These questions were responded to as yes or no by the participants.

Data analysis

NCSS (Number Cruncher Statistical System) 2007 (Kaysville, Utah, USA) program was used for the statistical analyses evaluating the findings obtained in the study. Mean, standard deviation, median, frequency, percentage, minimum, maximum) were used to describe the study data. The relationships between variables were investigated using independent samples t-test, Pearson chi-square test, point biserial correlation, and tetrachoric correlation. The significance of all statistical procedures was accepted as p < 0.05.

To detail, descriptive statistics were used for determining the frequency (frequencies) as well as average, min-max, and standard deviation (descriptives) of the ages of males and females in this sample. Independent samples t-test was used to assess if the average age of boys and girls significantly differed. Pearson chi-square test was used to investigate the relationship between living places and gender. Gender ratios were computed by taking the ratio of males to females. For the total sample, there were 748 males and 163 females, making the ratio 748/163 = 4.59, which was reported as 4.6:1. For the villages, there were 27 males and 4 females, making the ratio 27/4 = 6.75, which was reported as 6.8:1.

Descriptive statistics were used to report the responses of parents to questions. Specifically, descriptive statistics (frequencies) were used to report the frequency and percent of parents selecting each response to the asked question.

Pearson chi-square test was used to test the relationships between psychiatric medication use and ASD severity, and psychiatric medication use and masturbation behavior. Point biserial correlation was used for analyzing the relationship between ASD severity and receiving vitamin supplements, and between ASD severity and vitamins reported by parents to provide the most benefit. The descriptive statistic of frequency was used to determine the most beneficial vitamin reported by parents.

The relationship between receiving dietary treatment and constipation, and that between receiving dietary treatment and being a picky eater was explored using Pearson chi-square test. The relationship between constipation and probiotic use was investigated using tetrachoric correlation.

Results

Participants comprised 748 boys (82.1%) and 163 girls (17.9%), with an average age of 7.22 ± 4.25 that was similar in boys (M_{age_boys} = 7.26 ± 4.31) and girls (M_{age_girls} =7.02 ± 3.99) (t (909) = .67, p = .50) (see Table 1 for age distribution). In this sample, 569 (62.5%) parents reported to be living in metropolitan cities of Turkey such as Istanbul or Ankara, 147 (16.1%) in cities, 164 (18.0%) in towns, and 31 (3.4%) in villages. A statistically significant difference was found in the distribution of the living places according to gender (χ² (3, n = 911) = 7.96, p = 0.047) using Pearson chi-square test, where the rate of girls living in big cities was significantly higher compared to boys and their rate of living in towns was significantly lower than that of the boys. Of the 853 (93.9%) participants who had a RAM report that indicates the severity of disability, 32 (3.8%) had an ASD severity level below 30% corresponding to a mild level of ASD, 452 (53.0%) had a severity level between 31% and 69%, denoting a moderate level of ASD, and 369 (43.2%) had a severity level of 70% and above, reflecting a severe level of ASD. Birth-related information; accompanying medical and behavioral conditions; and nutritional status, constipation, and selective eating data are presented in this section; gender and birth-related information is depicted in Table 1.

Birth-related factors: birth week, delivery mode, NICU, and birth order

In this sample, 0.9% (n = 8) of the cases were born before 28 weeks of gestation, while 4.5% (n = 41) were between 28 and 32 weeks and 13.5% (n = 123) were born after 32 weeks. It was also observed that 81.1% (n = 739) were born between 36 and 40 weeks. It was determined that 32.4% (n = 295) of the cases had vaginal delivery and 67.6% (n = 616) had a caesarean section. In addition to this, it was observed that 19% (n = 173) of the cases had stayed in the neonatal intensive care unit (NICU). It was found that 58.2% (n = 530) of the cases were the first child, 28.4% (n = 259) were the second child, 10.2% (n = 93) were the third child (Table 1).

Age at diagnosis

The mean age at diagnosis was 31.05 months ± 11.88, with a minimum age of 8 months and maximum of 96 months. In this sample, 120 (13.2%) participants were diagnosed between 0 and 18 months of age, 218 (23.9%) when the child was between 19 and 24 months of age, 390 (42.8%) was diagnosed between 25 and 36 months of age, and 183 (20.1%) when they were 36 months or older (Table 1).

Male-to-female ratio

The male-to-female ratio in this sample of Turkish individuals with ASD is 4.6:1. While this is the ratio for the overall sample, the ratio in the villages was 6.8:1, with 27 boys (87.1%) and 4 girls (12.9%) in the subgroup of Turkish individuals with ASD in the villages.

Accompanying conditions in individuals with ASD: seizures, obsessions, allergy, inappropriate sexual behavior, harming behavior, and psychiatric medicine use

Seizures are a common comorbid condition observed in ASD. In this sample, 18.6% (n = 169) of the participants had at least one feverish or feverless seizure. As the child psychiatrists diagnosing and following up ASD cases in Turkey usually ask for EEG scans, we asked if their child had an EEG scan taken, and 69.3% (n = 632) parents reported their child to have EEG results and 30.6% (n = 279) of the participants did not have an EEG scan taken. We also asked parents if their psychiatrist detected a problem in their EEG results. While 15.1% (n = 138) of the participants were said to have problems in their brain waves as seen in their EEG results, the parents reported no problems to be observed in the EEG scans of 54.2% (n = 494) participants by neurologists.

When asked if their child had obsessive behaviors, 680 (74.6%) of parents said yes, and 231 (25.4%) said no. Parents were asked if their child had any allergic conditions. In this sample, 268 (29.4%) said yes, 643 (70.6%) said no. When asked whether their child had inappropriate sexual behaviors such as masturbation, 222 (24.4%) said yes and 689 (75.6%) said no. When asked if their child displayed self-harming behaviors, 231 (25.4%) said yes, 680 (74.6%) said no. When asked if their child displayed harming behaviors towards others, 206 (22.6%) said yes, 705 (77.4%) said no. So, 3 out of 4 participants in this sample exhibit obsessive behaviors, whereas approximately 1 out of 4 is reported to have allergic conditions, inappropriate sexual behaviors, self-harming behaviors, and harmful behaviors towards others.

In this sample, 437 (48.0%) individuals with ASD were reported by their parents to use psychiatric medication and 474 (52.0%) individuals with ASD have never used psychiatric medicines. When asked about which medicine their child benefited from the most, 113 (12.4%) participants said Risperdal (Risperidone), 46 (5.0%) said Abilify (Aripiprazole), 18 (2.0%) said Vayarin (Lipirinen), 12 (1.3%) said Cortexin, 109 (12.0%) said other medicines, and 130 (14.3%) reported that they benefited from none of the medications. In this sample, whereas there was no significant difference (t (851) = 1.68, p = .094) between the ASD severity levels of the participants who used psychiatric medication (M_{disability percentage} = 60.60 ± 21.67) and those who did not (M_{disability percentage} = 58.14 ± 21.18), the rate of using psychiatric medicine in individuals with masturbation behavior was found to be significantly higher than the individuals without masturbation behavior (χ² (1, n = 911) = 19.39, p = 0.001) using Pearson chi-square test.

Nutritional and dietary status, constipation, and selective eating

When asked if their child received a dietary treatment for a duration of at least 3 months, 301 parents (33.0%) said yes, and 610 (67.0%) said no. Out of the 301 parents who said yes, 215 (23.6%) reported their child to have benefited from the dietary treatment, and 86 (9.4%) said that their child did not benefit from such treatment. Parents were asked if they used vitamin supplements for their child. To this, 434 (47.6%) said yes, 477 (52.4%) said no. When asked which vitamin they benefited the most out of the ones they used, 162 (17.8%) said fish oil, 42 (4.6%) said B12, 38 (4.2%) said vitamin D, 35 (3.8%) said B6, 20 (2.2%) said probiotics, 20 (2.2%) Quercetin, 3 (0.3%) said TMG, 84 (9.2%) said other, 28 (3.1%) said none of the above. To this question, a total of 432 participants responded, suggesting that they are likely to also have said yes to using vitamin supplements in the previous question, and possibly 2 participants who said yes to receiving vitamins did not respond to this question. To inquire about special nutritional supports they may have used, parents were asked if they have ever administered special nutritional support for their child such as coconut oil and garlic cure, 146 (16%) said yes, 765 (84%) said no. A point biserial correlation was used to determine the relationship between the severity of ASD and the status of receiving vitamin supplements and the most beneficial vitamin supplements (Vitamin B12, Vitamin B6, Fish oil, vitamin D, Quercetin, Probiotic, other, none). No statistically significant relationship emerged between ASD severity and the status of receiving vitamin supplements (p = 0.895) and the vitamin supplements that were reported by parents to provide the most benefit (p values ranging between 0.245 and 0.954). In other words, there was no significant relationship between the severity of ASD and receiving any type of vitamin or nutritional supplements. Parents considered the most beneficial vitamin supplement for their children to be fish oil. In this sample, 37.5% (n = 162) of the parents of children with ASD who had received vitamin supplements stated that they benefited the most from fish oil.

The parents were asked if their child with ASD has constipation problems. To this, 271 (29.7%) said yes, 640 (70.3%) said no. Pearson chi-square test revealed no statistically significant relationship between receiving dietary treatment and constipation (χ² (1, n = 911) = 0.14, p = 0.7050. When asked if their child chose foods i.e. were picky eaters, 625 (68.2%) said yes, 286 (31.4%) said no. The rate of being a picky eater was found to be significantly lower in participants who received dietary treatment compared to those who did not receive dietary treatment (χ² (1, n = 911) = 25.85, p = 0.001) using Pearson chi-square test. Tetrachoric correlation showed no statistically significant relationship between constipation and probiotic use of individuals with ASD (p > 0.05).

Discussion

This study presents epidemiological data about youth with ASD from Turkey in topics of research interest for the ASD population. Specifically, it focuses on sociodemographic information, behavior problems, accompanying medical conditions, and dietary conditions of individuals with ASD. A recent study showed children with ASD from developed English-speaking countries to improve less compared to children with ASD from non-English-speaking countries over time (Mahapatra et al., 2020). This finding highlights the importance of disseminating knowledge about epidemiological profiles of individuals with ASD in other countries beyond developed English-speaking countries like the US and the UK. Such efforts can help understand why children with ASD from non-English speaking countries improved more in terms of symptoms than children with ASD from developed, English-speaking countries and thereby inform and guide global interventions for ASD. Therefore, the present study aimed to contribute valuable information about the personal, physiological, and behavioral characteristics of Turkish individuals with ASD to the literature about ASD from non-English speaking countries and across the world.

When two primary elements of epidemiology, age at diagnosis and gender distribution are considered, the findings from this sample are within the range reported globally, though with a somewhat earlier age of diagnosis and with certain variations in male-to-female ratio in terms of location. Though clinical ASD diagnoses made at 18 and 24 months show high stability, many children with ASD do not receive an ASD diagnosis before 36 months of age, which was around 5 years of age between 2012 and 2015 and decreased to 4 years in 2015–2016, with the mean age of diagnosis usually found to be around 3.1 years or 38 ± 15 months of age (Mandell et al. 2005, May and Williams 2018, Ozonoff et al., 2015, Valicenti-McDermott et al. 2012). In our study, the mean age of diagnosis was 31.06 ± 11.88 months. This may be related to the Autism Screening Program that was developed by the Ministry of Health of Turkey. Also, campaigns by autism foundations are often made for increasing awareness about autism in Turkey. These may have induced parents to notice children on the risk of autism early, and then refer them to physicians. ASD manifests more commonly in boys than girls. Whereas the traditional gender ratio estimate in ASD has been 4:1 with more males than females affected, more current analyses reveal ratios such as 6:1, 3.5:1, and 3:1, with an analysis of data between 2012 and 2016 showing this ratio to change between 4:1 and 2:1 through the years (Loomes et al. 2017, May and Williams, 2018, Rutherford et al. 2016, Whiteley et al. 2010). Loomes et al. (2017) suggested a diagnostic gender bias, where girls meeting ASD criteria face a considerable risk of not receiving a clinical diagnosis of ASD, and said that the true male-to-female ratio is not the traditionally assumed four to one, but is closer to three to one. The present findings of an overall male-to-female ratio of 4.6:1 and 6.8:1 in the villages aligns with Loomes et al. (2017)’s predictions, as difficulties in diagnosing girls are likely to be present in Turkey, especially in villages.

In the latest revision of DSM in 2013 (DSM-5), it was conveyed that high-functioning girls with ASD may go unrecognized ‘perhaps because of subtler manifestation of social and communication problems’ (American Psychiatric Association 2013). According to a study, although the age of onset of clinically important symptoms occurred at similar ages for males and females, girls were diagnosed about 0.2 years after boys, which was a statistically significant difference (May and Williams 2018). In screening for ASD, questionnaire items related to social communication and interaction problems were found to be most diagnostically efficient overall and especially for girls, and girls with ASD were found to show fewer restricted-repetitive symptoms than boys (Evans et al. 2018). Therefore, despite future research being required, striking social communication difficulties around 4–5 years of age may be one of the most suitable ways for assessing and diagnosing ASD in girls (Evans et al. 2018).

These findings suggest that diagnosis may be more difficult for girls, as they are likely to manifest social communication difficulties that are more subtle and expected of girls in patriarchal cultures like the Turkish culture, compared to the more easily discernible behavioral manifestations in boys. Such difficulties for noticing ASD in girls are likely to be more prevalent in towns compared to metropolitan cities or cities, which can explain the finding in this sample that the rate of girls living in towns was significantly lower than that of the boys. These diagnostic challenges may have potentially resulted in girls with ASD being overlooked, misdiagnosed or identified late in Turkey, especially for those girls who do not live in big cities.

Birth-related factors

Prematurity is an important risk factor for neurodevelopmental conditions. According to the definition of the World Health Organization (WHO), babies born before completing the 37th week or 259th day from the first day of the mother's last normal menstrual period are classified as preterm or premature (Dbstet 1977). While the premature birth rate in the world varies between 5 and 18% (WHO, 2012), according to the most up-to-date statistics that report 2018 data, the premature birth rate in Turkey is 15.6% (Turkish Ministry of Health 2018a). In our sample of Turkish individuals with ASD, the rate of premature birth was 19%. Also, according to the Turkish Statistical Institute (2018) and the statistics provided by the Turkish Ministry of Health (2018b), the prevalence of cesarean section births is 53% in Turkey. Our study showed that the prevalence of c-section births in Turkish individuals with ASD is 67.6%. Since both premature birth and c-section rates seem to be higher in individuals with ASD in this sample compared to the general Turkish population, the history of prematurity and cesarean birth may be considered as risk factors for autism.

Accompanying conditions

Seizures

The incidence of epilepsy accompanying autism in the literature has been reported to vary between 5% and 46% (Spence and Schneider 2009). As a convenient and relatively budget-friendly method to show abnormal cortical excitability underlying epilepsy and seizures (Smith 2005), EEG scans are often required from individuals with ASD in Turkey. In our study, 18.6% of the individuals with ASD were reported by their parents to have experienced at least one feverish or feverless seizure. In this study, 69.3% of individuals with ASD had an EEG scan taken, and the psychiatrists or clinicians noted abnormal EEG waves in 15.1% of the individuals in this sample. As EEG abnormalities have been reported to correlate with ASD severity, motor stereotypies, intellectual disability, language impairment, aggression, and self-harm (Nicotera et al. 2019), they may be important in understanding the phenotypic presentation of ASD in patients and may be one measurable component to be used in creating ASD subgroups.

Allergic conditions

In a nationally representative sample of American children, a significant positive association was found between ASD and common allergic conditions (Xu et al. 2018). In this Turkish sample, 29.4% of the parents reported their child to have an allergic condition. Even in studies where prevalence of allergies did not differ between children with ASD and control participants, allergies in children with ASD were associated with higher stereotypy scores (Lyall et al. 2015). So, in the future, not merely the presence or absence of allergies but also to which ASD behaviors allergies relate to those individuals with ASD who experience them should be investigated. Additionally, studies find a particular association between ASD and food allergies (Xu et al. 2018) and food allergies were found to be significantly associated with ASD even when overall allergies did not (Lyall et al. 2015), indicating food allergy as a potential target for future studies in this and other samples, which may further be explored for its relationship with certain ASD behaviors.

Behavior problems

Obsessive behavior

Obsessive compulsive disorder (OCD) is diagnosed in 37% of children with ASD (Leyfer et al. 2006) and autism traits are prevalent in OCD patients (Ivarsson and Melin 2008). Such a connection likely relates to the repetitive behavior tendency of these two groups, where both children with ASD and OCD have more compulsions and obsessions than control groups and parents report similar repetitive movement and sameness behaviors in children with ASD and OCD (Zandt et al. 2007). In the present sample, 74.6% of the Turkish individuals with ASD were reported to have obsessive behaviors by their parents, paralleling the previous findings in the literature.

Inappropriate sexual behavior

Although previous studies suggest that the sexual development and maturation processes of children with ASD are different from typically developing (TD) children, more recent studies show the opposite (Hayward and Saunders 2010, Stokes and Kaur 2005). The social immaturity of children with ASD may make it more difficult for them to cope with adolescence and act according to these urges in socially appropriate ways, and these confusing hormonal changes and inability to understand others’ behaviors may create frustration and anxiety that may manifest as inappropriate sexual behaviors, which are common in adolescents with ASD (Beddows and Brooks 2016). Similar to the definition in the literature, in the Turkish culture, all kinds of unsuitable sexual behavior displayed in public areas and excessive sexual behaviors performed in private are defined as inappropriate sexual behaviors. Inappropriate sexual behaviors of individuals with ASD are much more intense compared to their TD peers. TD individuals have an awareness about their social behaviors and their effects on other people. For instance, they do not show any intense and unsuitable sexual behavior in front of family members and public areas. On the other hand, individuals with ASD usually act in a way that is unacceptable to the requirements of public places such as touching their genital area, lying on the floor for rubbing their genital area to the surfaces, or squeezing their legs to feel sexual gratification. So, these behaviors are considered as inappropriate sexual behaviors, one of the most common such behaviors is masturbation. Though the literature shows most individuals with ASD, approximately 75%, to have inappropriate sexual behavior and most of them to masturbate (Sullivan and Caterino 2008), in our study, only 25% of participants were reported by their parents to display masturbation behavior. This difference may be related to the unwillingness of Turkish parents to report inappropriate sexual behaviors for their children given the conservative cultural norms in Turkey, further discussed under medication use below.

Harming behavior

Children with ASD exhibit more self-injurious behaviors than TD children or children with other neurodevelopmental disabilities, where self-injurious behaviors refer to behaviors that may lead to physical harm such as self-biting or head banging (Minshawi et al. 2014). In the US, a population-based study across the years 2000, 2006, and 2008 found self-injurious behavior prevalence among children with ASD to be 27.7% (Soke et al. 2016). In the present Turkish sample, 25.4% of parents reported their child to display self-harming behaviors, which aligns with the US statistics. Besides self-harm, children with ASD also engage in harmful behaviors towards other individuals. One out of four children with ASD experience aggressive behavior problems (Hill et al. 2014) and children with ASD show higher levels of physical and reactive aggression compared to children with other intellectual and developmental disabilities (Farmer and Aman 2011). In the present Turkish sample, 22.6% of parents reported their child to display harming behavior towards others, which is similar to the previous finding of 25% of children with ASD experiencing aggressive behavior problems (Hill et al. 2014). The findings of harmful behaviors toward self and others in this sample of Turkish individuals with ASD parallels other findings in the literature, showing around 1 in 4 individuals with ASD to display harmful behaviors toward themselves and others.

Psychiatric medication use

Though no medication exists to treat the core symptoms of ASD, psychiatric or psychotropic medications are often prescribed to individuals with ASD to address accompanying symptoms such as aggression, self-injurious behaviors, hyperactivity, and stereotypies, and different studies show 35%, 56%, 70% of children with ASD to use at least one psychotropic medication (Mandell et al. 2008, Oswald and Sonenklar 2007, Rosenberg et al., 2010). In this Turkish sample, 48.0% of the parents reported their child to use psychiatric medicines, and no significant difference was found in the ASD severity levels between the participants who used psychiatric medications and those who did not, supporting the notion that medication is usually used in ASD to treat accompanying conditions since there is no medication for the core symptoms of ASD. In this sample, there was a significant difference in medication use between participants who displayed masturbation behaviors and those who did not. There are different solutions in order to prevent masturbation behavior such as taking medicine or behavioral interventions. While taking medicine is a faster solution, behavioral interventions may take a longer time to remedy inappropriate behaviors. The authors’ clinical experience indicates a preference for medications by parents when their child exhibits inappropriate sexual behaviors with the hopes of getting quicker results, whereas parents do not prefer medicine use otherwise. This was supported by the study findings that found psychiatric medication used to be significantly higher in individuals with ASD with masturbation behavior. This tendency can be explained by the conservative cultural structure of Turkey.

One study with Turkish parents found that even though all participating parents reported that they consider sexual education necessary, 65.6% said that they received no information or training about how to provide sexual education to their children, and 25% of the parents said that they were not able to comfortably talk with their children about sexuality because they felt inadequate, they did not know how much information should be given at what age or considered talking about sexuality to be sinful or shameful (Ceylan and Çetin 2015). In our sample, parents might have avoided giving sexual education to their children because of their concerns that may be similar to those reported above by Turkish parents of TD children, and based on this, they might have preferred a drug-based solution. There is a possibility that parents may prefer the medicine because they consider it a less costly and short-term solution, and more appropriate for the Turkish culture than talking and getting training about sexuality. In addition, parents might prefer the drug instead of the behavioral approach because of the lack of professionals who are trained in sexual education for children with ASD and their families in Turkey.

Selective eating, constipation, and dietary interventions

In children with ASD, food selectivity or picky eating is a common problem, where 60–67% of parents surveyed report strong food preferences and picky eating in their children (Twachtman-Reilly et al. 2008), and patient records and evaluations revealed 62% of children with ASD to display food selectivity (Field et al. 2003). Of the parents of the present study, 68.6% declared their child to be a picky eater, matching the previous literature findings, and 33.0% reported their child to receive a dietary treatment for at least three months. There were fewer picky eaters in the group who received dietary treatments, which makes sense considering the difficulty experienced by parents to administer dietary plans to picky eaters who are already selective about what they eat. In this sample, 47.6% reported to use vitamin supplementation for their child, and 29.7% reported their child to experience constipation problems.

The non-significant relationship between probiotic use and constipation found in this study may be explained by the way the questions were asked. They were asked whether their child with ASD ever received a nutritional supplement and in the next question, they were asked from which nutritional supplement they received the most benefit where probiotics were listed as an option. Yet, the question about constipation was asked in the present tense, as ‘Does your child with ASD have a constipation problem?’ So, it is possible that the individual with ASD received probiotics some time in their lives and are not using them currently but are experiencing or not experiencing constipation now. Therefore, future studies can specify the time frame to more accurately explore the relationship between constipation and probiotic use in Turkish individuals with ASD. Similarly, one reason that no significant relationship emerged between probiotic use and ASD severity may be that unlike other studies that specifically investigated this relationship (e.g. Shaaban et al. 2018), this study did not compare the ASD severity levels before and after probiotics use. Another reason may be that the time and duration of probiotic use in relation to when ASD severity was evaluated was not known in the present study and can be investigated in future studies. Given other findings showing a strong positive association between gastrointestinal problems and ASD severity (Adams et al. 2011) and the recent interest in the role of the gut-brain axis in ASD pathogenesis and prognosis (Doenyas 2018), referring children with ASD to dietitians, endocrinologists, or laboratories for gut microbiota analyses may not only alleviate their accompanying conditions that may increase the severity of their ASD symptoms, but also enable finding personalized treatments for their own physiological profiles based on the newly emerging personalized dietary interventions that are combined with prebiotic and probiotic supplements for individuals with ASD (Doenyas 2019, Grimaldi et al. 2018).

Limitations and future directions

The main limitation of this study is that the assessment tools are based on the reports of the parents and the examination process was not under control of professionals. The evaluation forms have been created in accordance with the clinicians’ own experiences and observations in child neurology. Consequently, this study may have biased results based on parents' psychological status and physical environments. Further research should evaluate the children through more objective forms and evaluations completed by professionals.

This focus of this study was on the general physiological and behavioral characteristics of individuals with ASD in Turkey. Therefore, some questions were asked as ‘Has your child with ASD ever…’ and others as ‘Does your child with ASD…’ In future studies, for specific research questions that stem from the general information provided here, questions can be used that are more focused on time frames of nutritional supplement utilization, dietary intervention administration, and gastrointestinal problems. Along these lines, it can be recommended for future studies to probe allergic conditions in more detail and identify the type of allergies such as food or pet allergies, to inquire about subtypes of obsessive behaviors such as counting, ordering, or arranging, to investigate the gastrointestinal symptoms of Turkish individuals with ASD in more detail beyond constipation, and to compare these symptoms between Turkish individuals with ASD who receive and who do not receive probiotic supplements. In addition, to be more specific than the version asked in the present study, future studies can ask about the type of dietary intervention, whether the families consulted a dietitian or not, and the regimen and frequency of using probiotics and other supplements to investigate if possible relationships exist between these variables and ASD severity.

Conclusion

This study presents epidemiological data about ASD in Turkey. It found a male-to-female ratio of 4.6:1 and the average age at diagnosis to be 31.06 ± 11.88 months for ASD in Turkey. In this sample of Turkish individuals with ASD, 3 out of 4 were found to have obsessive behaviors, and 1 out of 4 to have allergic conditions, inappropriate sexual behaviors, self-harming behaviors, and harmful behaviors towards others. In this sample, 2 in 3 participants were reported to be picky eaters, parents reported 1 in 3 to receive a dietary treatment for at least 3 months, and almost half of the parents said they used vitamin supplements for their child with ASD. These findings give information from a Turkish sample of individuals with ASD about gender ratios, accompanying behaviors, and dietary interventions that are topics of interest for the ASD population.

ASD is a heterogeneous neurodevelopmental condition resulting from the interaction of many environmental and genetic factors. This study aimed to reveal multiple variables that may be relevant to understanding this complex condition, such as childbirth-related factors, age of diagnosis, accompanying conditions, gastrointestinal problems, selective eating, and dietary interventions for Turkish individuals with ASD. We hope that this epidemiological depiction of youth with ASD in Turkey will contribute information from a Turkish sample to the literature to help understand the etiology and correlates of this condition that yet eludes a single comprehensive explanation or treatment.

Conflict of interest

No potential conflict of interest was reported by the authors.