An incidence study of diagnosed autism-spectrum disorders among immigrants to the Netherlands

Abstract

Objective

To estimate the risk of developing autism-spectrum disorder (ASD) in children born to immigrants as compared with children of Dutch-born parents.

Method

Retrospective, population-based cohort study of all live births (n = 106 953) between 1998 and 2007 in a circumscribed geographical region in the Netherlands. Cohort members were linked to the Psychiatric Case Register to identify diagnosed cases.

Results

A total of 518 cases of ASD were identified, including 150 children with autism and 368 children with Asperger syndrome or Pervasive Developmental Disorder Not Otherwise Specified (PDD-NOS). Children born to migrants from developing countries were at significantly lower risk of ASD [rate ratio (RR) = 0.6, 95% confidence interval (CI) 0.5–0.9] than children of Dutch-born parents. Within the ASD group, the risk for the subgroup with Asperger syndrome and PDD-NOS was reduced (RR = 0.4, 95% CI 0.3–0.6), whereas that for narrowly defined autism was non-significantly increased (RR = 1.4, 95% CI 0.9–2.4). Migrant groups did not differ in age at diagnosis.

Conclusion

The results echo Swedish findings indicating a reversal of risk gradient in children of parents from developing countries, specifically a decreased risk for high-functioning and increased risk for low-functioning autism.

Significant outcomes

• The incidence of diagnosed Asperger syndrome and Pervasive Developmental Disorder Not Otherwise Specified (PDD-NOS) is significantly decreased among children of parents born in developing countries.
• The results of this study support Swedish findings indicating reverse risk gradients across high-functioning (lower risk) and low-functioning (higher risk) autism in children of parents from developing countries.

Limitations

• The limited number of cases precluded in-depth examination of putative risk factors for autism-spectrum disorder (ASD) that may also be associated with migration or migrant status.
• Group comparisons based on administrative incidence may be biased by differential help-seeking, referral and service use.

Introduction

The long-held axiom that autism is an almost exclusively heritable neurodevelopmental disorder has recently been challenged by a twin study showing that 55% of the liability to autism can be explained by the shared twin environment [1]. Parental immigration as a risk factor for autism-spectrum disorder (ASD) in the child is receiving renewed attention. In the UK, mothers born outside of Europe were found to have a significantly higher risk of having a child with ASD compared with native-born mothers, with the highest risk observed for the Caribbean group [2]. Two recent Swedish studies [3, 4] reported differential results contingent on level of functioning of the child. Maternal birth outside Northern Europe was associated with a reduced incidence of high-functioning and an elevated incidence of low-functioning subtypes of ASD. While most European findings point to increased rates [2, 5, 6], studies from the US have reported no differences or even decreased rates of ASD according to maternal migrant status or ethnic origin [7-9]. Variation in study results, as well as differences in diagnostic categorization, migrant group distribution, case ascertainment and confounding variables, challenge the interpretation of these findings [10].

Aims of the study

The purpose of this study was to examine whether a history of parental migration influenced a child's risk of being diagnosed with ASD. We compared the risk in children of two Dutch-born parents to the risk in children with at least one parent born in Morocco, Turkey, Surinam, the Netherlands Antilles or other countries with developing or developed economies. We also explored the effect of parental migration on the incidence of separate ASD subtypes, i.e., autistic disorder, Asperger syndrome and Pervasive Developmental Disorder Not Otherwise Specified (PDD-NOS).

Material and methods

Study population and data extraction

A retrospective cohort study was carried out using two data sources. The first source pertained to a birth-cohort, with information on the parents, that was selected from the civil registry database of Statistics Netherlands (Centraal Bureau voor de Statistiek, CBS), consisting of all live births in the urban area of Utrecht and surrounding semi-rural municipalities between January 1, 1998 and December 31, 2007 (n = 106 953; hereafter: CBS-cohort). The CBS is responsible for collecting and processing all individual and population health care data in the Netherlands for the purpose of research by health authorities and academic institutions [11]. Under Dutch privacy law, use of personal (health care) data for the purpose of scientific research is allowed, provided that data cannot be traced to the individual. For 514 individuals (0.5%), the mother could not be identified, for 3735 (3.5%) the father and for 378 individuals (0.4%) neither the mother nor the father.

The second source of the data originates from The Psychiatric Case Registry of the central part of the Netherlands (PCR-MN), which is in operation since 1999. It contains anonymized information on all patients who attended in- or out-patient facilities for mental health care until December 31^st 2009, including date of birth, gender, postal code and at least one DSM-IV diagnosis [12]. The PCR-MN does not receive information from GPs, paediatricians or from the small number of psychiatrists working in private practice who rarely see patients with autism. Thus, it is likely that the vast majority of children referred to psychiatric services are recorded by PCR-MN.

Linkage between PCR-MN and CBS-cohort was based on date of birth of the child, gender and a part of the postal code. The postal codes included in the CBS-cohort are identical to those covered by the PCR-MN region. A unique match with a record in the CBS-cohort could be established in 89% of the patients registered in the PCR-MN with autistic disorder (DSM-IV code 299.00), Asperger syndrome (299.80) or PDD-NOS (299.80). Country of birth and parental country of birth are known for all children in the CBS-cohort, because it is part of the information provided at compulsory child registration after birth. In the PCR-MN, parental country of birth was incomplete for 27% of the children. However, we used the available information from the PCR-MN to verify whether ethnic minorities were disproportionally present among cases that could not be linked to the CBS-cohort. The proportions of children for whom successful record linkage was possible were 84% for the native Dutch children, 86% for the Moroccan, 83% for the Turkish, 100% for the Surinamese and Dutch Antillean children and 91% for those from other countries of origin. Thus, in the cohort resulting from the merge between PCR-MN and CBS-cohort, the occurrence of diagnosed ASD may be slightly underestimated, but not differentially between immigrants or native-born children.

For children with a diagnosis of ASD, the period at risk ended at the time of diagnosis, i.e. at the age at diagnosis. For children without a diagnosis of ASD, the period at risk ended at emigration outside the catchment area, death or 31 December 2009, whichever came first. In addition, to evaluate whether any difference between migrants and non-migrants in rates of ASD or its subtypes may be explained by differences in time until diagnosis, age at diagnosis was estimated for each migrant group.

In line with the CBS definition of immigrant status, a Dutch-born subject with two Dutch-born parents was considered native Dutch. A Dutch-born subject with at least one foreign-born parent was classified according to the country of birth of that parent. When the parents were born in different foreign countries, the maternal country of birth was decisive for assignment to a particular group. In an effort to distinguish the effect of maternal from paternal migration, we consecutively reclassified children based on the country of birth of the mother only. Most migrants in the study area originated from Turkey, Morocco, Surinam and the Netherlands Antilles. The latter two groups of migrants from former Dutch colonies with relative geographical proximity were collapsed into a single group in the analyses. Migrants were categorized according to their source country's economic situation to capture the putative influence of discrepancy in the level of development between source and host society. Using the United Nations Conference on Trade and Development (UNCTAD) classification for the year 2011 [13], countries were divided into three categories: 1) developing economies, 2) transition economies (e.g., Croatia, Ukraine), and 3) developed economies. UNCTAD group 2 including predominantly Eastern European countries was adjoined with group 3, reasoning that these countries mostly resemble countries with developed economies in terms of cultural and ethnic characteristics. This combined group is hereafter referred to as ‘developed countries’. UNCTAD group 1 is referred to as ‘developing countries’.

Statistical analysis

Group comparisons on demographic variables and age at diagnosis were performed using t tests for independent groups. Incidence rates of any ASD, and separately for narrowly defined autism vs. Asperger syndrome and PDD-NOS, were estimated for each migrant group. Multivariable Poisson regression analysis was used to estimate rate ratios (RRs), adjusted for gender and paternal age. These variables were selected a priori based on the reported association with the incidence of ASD [14]. We adjusted for advanced maternal age at birth, which has also been related to an increased risk for ASD [15], in separate analyses to minimize the possible effect of collinearity with paternal age. RRs with a confidence interval (CI) not including unity were considered statistically significant. Time of observation was used as offset and the variance was scaled to take possible over-dispersion into account. The Poisson regression analysis was performed with the Generalized Linear Models module of STATA, version 10.0 [16]. The overall between-group differences in risk were tested with the Wald chi-square statistic.

We used the Kaplan–Meier method to estimate the cumulative incidence for a diagnosis of ASD at age 10 in the entire birth-cohort. The age cut-off was set at a maximum of 10 years, because for an older age, the majority of the cohort was not followed up through the period of risk. Because migrants tend to settle in urban areas and urbanicity has been associated with an increased risk for autism [7], we performed subanalyses on a restricted urban sample of the inhabitants of the city of Utrecht only.

Results

Cohort description

During the 673 752 person-years at risk, 518 children developed ASD of whom 150 were diagnosed with an autistic disorder and 368 with Asperger syndrome or PDD-NOS. This corresponded to a cumulative incidence at age 10 of 1.0% (95% CI 0.9–1.1) for any ASD, or 0.2% (95% CI 0.1–0.3) for autism and 0.8% (95% CI 0.7–0.9) for Asperger syndrome and PDD-NOS.

Loss to follow-up occurred mostly because of individuals moving out of the catchment area (13.8% of total cohort), which was less common in families with a child diagnosed with ASD. Individuals diagnosed with ASD were predominantly male. About one-fourth of the children had one or two foreign-born parents. There was no large or significant difference in paternal or maternal age at birth between children with and without ASD (Table 1). Age at diagnosis was significantly lower for children with an autistic disorder compared with those with Asperger syndrome and PDD-NOS (t = 10.3, P < 0.01), also when grouped by parental country of birth in native Dutch, developed or developing countries (Table 2). There were no significant differences between migrant groups and native Dutch in terms of duration of follow-up or number of cases lost to follow-up, sex distribution or mean age at diagnosis.

Table 1. Basic demographic characteristics of denominator and numerator populations

Denominator population	Population at riska (n = 106 953)	Any ASD (n = 518)	Asperger syndrome and PDD-NOS (n = 368)	Autistic disorder (n = 150)
^a Total population at risk per study year. ^b Divided according to the United Nations Conference on Trade and Development (UNCTAD) classification of economic development. ‘Developed countries’ include countries with economies in transition and developed economies; ‘Developing countries’ include those with developing economies. ^c Born in the Netherlands and both parents born in the Netherlands. ^d European countries (except for Turkey), Japan, US, Canada, Australia, New Zealand.
Gender, n (%)
Male	54 774 (51.0)	443 (85.5)	310 (84.2)	133 (88.7)
Female	52 179 (49.0)	75 (14.5)	58 (15.8)	17 (11.3)
Parental country of birthb, n (%)
Netherlandsc	80 354 (75.1)	420 (81.1)	320 (87.0)	100 (66.7)
Developed countriesd	6651 (6.2)	29 (5.6)	16 (4.3)	13 (8.7)
Developing countries	19 948 (18.7)	69 (13.3)	32 (8.7)	37 (24.7)
Paternal age at birth, mean (SD)	34.4 (5.2)	34.1 (5.2)	33.8 (5.2)	34.6 (6.0)
Maternal age at birth, mean (SD)	31.5 (4.6)	31.2 (4.6)	31.2 (4.6)	31.2 (4.7)
Age at diagnosis, mean (SD)	NA	6.2 (2.3)	6.8 (2.1)	4.7 (2.1)

Table 2. Paternal and maternal age at birth, person-years of follow-up and mean age at diagnosis by level of economic development of parental country of birth

Parental country of birth	Paternal age at birth, mean (SD)	Maternal age at birth, mean (SD)	Person-years of follow-upa	Any ASD		Asperger's syndrome and PDD-NOS		Autistic disorder
				N (%)	Mean age at diagnosis (SD)	N (%)	Mean age at diagnosis (SD)	N (%)	Mean age at diagnosis (SD)	t b	P
SD, standard deviation. ^a total follow-up period for cases and non-cases until emigration outside the catchment area, death or 31 December 2009, whichever came first. ^b t Test statistics and corresponding P-values for the comparisons of mean age at diagnosis for Asperger's syndrome and PDD-NOS, and autistic disorder.
Netherlands	34.3 (4.8)	31.9 (4.2)	509 819	420 (81.1)	6.3 (2.3)	320 (87.0)	6.8 yrs (2.1)	100 (66.7)	4.9 yrs (2.2)	7.8	<0.01
Developing countries
All	34.5 (6.6)	29.7 (5.5)	126 596	69 (13.3)	5.3 (2.2)	32 (8.7)	6.5 (2.1)	37 (24.7)	4.4 (1.8)	4.5	<0.01
Turkey	31.3 (5.6)	28.2 (5.1)	19 184	14 (2.7)	5.9 (2.4)	7 (1.9)	7.3 (1.3)	7 (4.7)	4.5 (2.5)	2.6	0.02
Morocco	35.1 (6.9)	29.2 (5.5)	58 469	16 (3.1)	5.7 (1.9)	7 (1.9)	6.5 (1.8)	9 (6.0)	5.0 (1.8)	1.7	0.12
Suriname and Dutch Antilles	34.6 (5.9)	31.4 (5.2)	16 733	17 (3.3)	5.7 (2.3)	11 (3.0)	6.5 (2.6)	6 (4.0)	4.4 (0.9)	1.9	0.08
Other	35.1 (6.3)	30.5 (5.5)	32 210	22 (4.2)	4.4 (2.0)	7 (1.9)	5.5 (2.3)	15 (10.0)	3.9 (1.7)	1.8	0.08
Developed countries
All	35.2 (5.9)	32.1 (4.7)	39 891	29 (5.6)	5.5 (2.5)	16 (4.3)	6.8 (2.5)	13 (8.7)	3.9 (1.2)	3.8	<0.01
Total			676 306	518		368		150

Incidence of ASD by parental country of birth

Incidence rates and RRs grouped by parental country of birth are displayed in Table 3. There were no major differences between crude RRs, RRs adjusted for gender and paternal age, and RRs adjusted for gender and maternal age. Parental birth in a developing country was associated with a significantly decreased risk of ASD (RR: 0.6, 95% CI 0.5–0.9). Children born to Moroccan immigrants were at a very low risk (RR: 0.3, 95% CI 0.2–0.6), whereas the risk for children from developed countries was similar to that of native Dutch.

Table 3. Incidence rates and rate ratios (RRs) of ASD and its subtypes in Utrecht and surroundings by level of economic development of parental country of birth

Parental country of birth	Any ASD		Asperger's syndrome and PDD-NOS		Autistic disorder
	Rate per 10 000 py (CI)	RR (95% CI)a	Rate per 10 000 py (CI)	RR (95% CI)a	Rate per 10 000 py (CI)	RR (95% CI)a
ASD, autistic spectrum disorder; py, person-years; RR, rate ratio; CI, confidence interval; df, degrees of freedom. ^a Adjusted for gender and paternal age. ^b Significance level of between-group differences in estimated RRs, excluding the combined category ‘developing countries, all’. ^c Significance level of differences in estimated RRs between three groups of parental country of birth: the Netherlands, developing countries and developed countries.
Netherlands	8.2 (7.4–9.0)	1.0	6.2 (5.6–7.0)	1.0	1.9 (1.6–2.3)	1.0
Developing countries
All	5.4 (4.2–6.9)	0.6 (0.5–0.9)	2.5 (1.7–3.5)	0.4 (0.3–0.6)	2.9 (2.0–4.0)	1.4 (0.9–2.4)
Turkey	7.3 (3.9–12.2)	0.8 (0.5–1.5)	3.6 (1.4–7.5)	0.5 (0.2–1.2)	3.6 (1.4–7.5)	1.8 (0.6–4.9)
Morocco	2.7 (1.5–4.4)	0.3 (0.2–0.6)	1.2 (0.4–2.4)	0.2 (0.1–0.4)	1.5 (0.7–2.9)	0.8 (0.3–1.9)
Suriname and Dutch Antilles	10.1 (5.9–16.2)	1.2 (0.7–2.1)	6.5 (3.2–11.7)	1.0 (0.5–1.9)	3.5 (1.3–7.8)	1.8 (0.6–5.4)
Other	6.8 (4.2–10.3)	0.8 (0.5–1.3)	2.1 (0.8–4.4)	0.3 (0.2–0.7)	4.6 (2.6–7.6)	2.3 (1.1–4.8)
Developed countries
All	7.2 (4.8–10.4)	0.9 (0.6–1.3)	4.0 (2.2–6.5)	0.6 (0.4–1.1)	3.2 (1.7–5.5)	1.6 (0.8–3.5)
		P < 0.01 (df = 5)b P = 0.09 (df = 2)c		P < 0.01 (df = 5)b P < 0.01 (df = 2)c		P = 0.14 (df = 5)b P = 0.22 (df = 2)c

Incidence of ASD subtypes by parental country of birth

Second-generation migrants were at decreased risk of Asperger syndrome or PDD-NOS. This finding was significant for children from developing countries (RR: 0.4, 95% CI 0.3–0.6). A reversed pattern was observed for a diagnosis of autism, for which a non-significantly increased risk was observed of 1.6 (95% CI 0.8–3.4) and 1.4 (95% CI 0.9–2.4) for the developed and developing group respectively.

Children of Moroccan-born parents, were at significantly lower risk of receiving a diagnosis of Asperger syndrome or PDD-NOS, but not of autistic disorder. The group of children from developing countries other than Morocco, Turkey, Surinam or Netherlands Antilles was at significantly decreased risk of Asperger syndrome or PDD-NOS (RR: 0.3, 95% CI 0.2–0.7), but at significantly increased risk for autistic disorder (RR: 2.3, 95% CI 1.1–4.8).

Migrant groups from developing countries, especially those from Turkey (RR: 1.0, 95% CI 0.5–1.9) and Suriname and Dutch Antilles (RR: 1.8, 95% CI 0.8–3.7) were at a slightly higher risk of ASD in the strictly urban subsample compared with the total sample. However, differences in risk between Asperger syndrome or PDD-NOS and autistic disorder remained present in all immigrant groups (see Table 4).

Table 4. Rate ratios (RRs) of ASD and its subtypes in an exclusively urban study area (city of Utrecht) by level of economic development of parental country of birth

Parental country of birth		Any ASD		Asperger's syndrome and PDD-NOS		Autistic disorder
		N	RR (95% CI)a	N	RR (95% CI)a	N	RR (95% CI)a
ASD, autistic spectrum disorder; RR, rate ratio; CI, confidence interval; df, degrees of freedom. ^a Adjusted for gender and paternal age. ^b Significance level of between-group differences in estimated RRs, excluding the combined category ‘developing countries, all’. ^c Significance level of differences in estimated RRs between three groups of parental country of birth: the Netherlands, developing countries and developed countries.
Netherlands		110	1.0	86	1.0	24	1.0
Developing countries
All		46	0.6 (0.4–1.0)	23	0.5 (0.3–0.8)	23	1.2 (0.6–2.5)
Turkey		13	1.0 (0.5–1.9)	7	0.7 (0.3–1.5)	6	2.5 (0.8–7.7)
Morocco		13	0.4 (0.2–0.8)	6	0.2 (0.1–0.5)	7	1.0 (0.4–2.9)
Suriname and Dutch Antilles		9	1.8 (0.8–3.7)	6	1.5 (0.7–3.5)	3	2.6 (0.6–11.8)
Other		11	1.0 (0.5–1.9)	4	0.5 (0.2- 1.3)	7	2.8 (1.0–8.1)
Developed countries
All		9	0.7 (0.3–1.4)	4	0.4 (0.1–1.1)	5	1.7 (0.5–5.7)
			P = 0.03 (df = 5)b P = 0.09 (df = 2)c		P < 0.01 (df = 5)b P = 0.04 (df = 2)c		P = 0.27 (df = 5)b P = 0.64 (df = 2)c

Incidence of ASD and its subtypes by maternal and paternal country of birth separately

When reclassifying the sample based on birth country of the mother only, there was a minor decrease in estimated risk of ASD among children with a mother from a developed country (RR: 0.5, 95% CI 0.3–1.0) and among those with a Surinamese or Dutch Antillean mother (RR: 1.0, 95% CI 0.5–2.0). In other migrant groups the RR for ASD, autistic disorder and Asperger syndrome or PDD-NOS did not change more than 10% for the reclassified sample. To study the effect of specifically paternal migrant status, children with a foreign-born father and a Dutch-born mother (ASD total n = 37) were analysed as a separate group. The RR of ASD among children with a Dutch mother and a foreign father compared with native Dutch was 1.2 (95% CI 0.8–1.8) with a lower risk of Asperger or PDD-NOS (RR: 0.8, 95% CI 0.5–1.3) and a significantly higher risk of autistic disorder (RR: 2.6, 95% CI 1.4–5.0).

Discussion

This is the first epidemiological study on the incidence of diagnosed ASD in the Netherlands. A decreased risk was found for ASD among second-generation immigrants from developing countries. A closer look at specific diagnostic categories revealed that this was attributable to a significantly lower relative risk for Asperger syndrome and PDD-NOS, but not for autistic disorder, which showed a trend in the opposite direction, albeit statistically inconclusive. The study may have lacked the power to reveal significant differences in the incidence of autistic disorders as a function of parental migrant status. The restricted number of cases also precluded in-depth analyses on other relevant factors, for example, the timing of maternal migration. Asperger syndrome and PDD-NOS were registered under an identical DSM-IV code (299.80), hence we were unable to calculate incidence rates for each diagnostic group separately. The external validity of these findings may be limited for rural areas, as the study area comprised an urban and semi-rural area. However, it is unlikely that urbanicity acted as a confounder, because in a more homogeneous, strictly urban subsample, the association between the risk of ASD and migrant status remained similar, as well as the risk difference between ASD subtypes. A major strength of the study is that PCR-MN covers both in- and out-patient psychiatric health care facilities in a large, well-defined study area. Because of the nature of the Dutch health care system, characterized by compulsory medical insurance and compulsory diagnostic registration to receive funding, the diagnostic registration of treated patients is particularly complete [12]. This is supported by our reported cumulative incidence of ASD of roughly 1%, which is in accordance with two large-scale, prevalence studies using rigorous case finding methods [17, 18].

It is uncertain to what degree the decreased risk of Asperger syndrome or PDD-NOS and the increased risk of autistic disorder among migrants from developing countries represent valid findings, or, alternatively, the result of various potential biases. The findings concur with two recent Swedish studies [3, 4] showing a negative and a positive association between the risk for high and low-functioning autism, respectively, and maternal birth outside Northern Europe. It was argued that the stressful circumstances of migration contribute to the increased risk for autism with co-morbid intellectual disability, in particular when operating in utero [4]. Similar findings have been reported in the US [19], where black ethnicity was positively associated with autistic disorder (RR= 2.6, 95% CI 1.3–5.0), with apparent reversal of risk for PDD-NOS and Asperger syndrome (RR= 0.5, 95% CI 0.2–1.0). Dealberto [20] suggested that maternal vitamin D insufficiency among migrant mothers may be responsible for increased rates of especially the strict definition of autism. Our findings did not uniformly support this hypothesis, given that autistic disorder was not more common in migrant groups in the Netherlands known to have low serum vitamin D, e.g., those from Moroccan origin [21]. In addition, we demonstrated that exclusively paternal foreign birth is also associated with a reverse risk gradient of ASD subtypes in the child, in particular, a significantly increased risk for autistic disorder. This suggests that the effect of parental migration on ASD risk is not fully explained by maternal migration history, but may also be mediated by the migration history of the father.

The low risk of Asperger syndrome and PDD-NOS for second-generation migrants may also be because of a low detection rate. Underutilization of child psychiatric services has been reported for non-Dutch youth in the Hague [22]. Parents originating from a different cultural background may not view the disturbed behaviour of their children as a medical problem or may be reluctant to seek help for fear of stigma [23, 24]. This would leave cases of less severe forms of ASD, such as Asperger syndrome or PDD-NOS, particularly undetected. Children with autistic disorder, on the other hand, often have concurrent intellectual disabilities, decreasing their possibilities to complete a school education in the regular stream, increasing the probability of help-seeking behaviour [25]. It is also possible that children with higher functioning forms of autism are not diagnosed at all, because clinicians may be inclined to attribute minor social and communicative problems to a child's ethnic background [26]. A study from the Netherlands found that paediatricians more often chose autism as the most likely diagnosis when judging clinical vignettes of European majority cases than vignettes including non-European minority cases (Moroccan or Turkish) [27]. However, our finding that age at diagnosis was similar across groups does not fit the notion that children born to immigrants are less likely to be diagnosed with ASD. Nevertheless, diagnosing minor behavioural problems among Dutch youth as Asperger syndrome and PDD-NOS can result in an overestimation of these disorders, leading to an artificially decreased observed incidence among immigrants.

Declaration of interest

None.