Phenotypic differences in individuals with autism spectrum disorder born preterm and at term gestation

Abstract

The objective of the study was to characterize the phenotype of males and females with autism spectrum disorder born preterm versus those born at term. Descriptive statistical analyses identified differences between male and female autism spectrum disorder subjects born preterm compared to term for several phenotypic characteristics and comorbidities. Of the 115 (13.0% of 883) born preterm, a greater percentage of males had sleep apnea (13.8% vs 2.5%, p < 0.0001), seizure disorders (17.0% vs 8.5%, p = 0.01), and attention-deficit/hyperactivity disorder (14.9% vs 6.6%, p = 0.005). Females born preterm were more likely to be nonverbal (22.2% vs 4.6%, p = 0.001). In summary, phenotypic differences were observed, especially among males. The results may have implications for understanding the underpinnings of a subset of individuals with autism spectrum disorder and contribute to the development of focused treatments for autism spectrum disorder among children born preterm.

Keywords

autism spectrum disorder subtypes comorbidities descriptive analyses preterm birth

Individuals with autism spectrum disorder (ASD) have persistent deficits in social and language functioning, as well as repetitive and restricted behaviors. While there is a significant genetic component contributing to the development of ASD, differences in the prevalence of pre- and peri-natal environmental factors have also been reported (Kolevzon et al., 2007). Preterm birth can affect a wide array of organs and systems, and the brain is particularly susceptible, leading to potential neurological deficits (Saigal and Doyle, 2008). Preterm birth has been evaluated in several epidemiologic studies relating pre- and peri-natal environmental factors to later development of ASD. While several studies have identified an association between preterm birth (Eaton et al., 2001; Kolevzon et al., 2007; Larsson et al., 2005) and especially very early preterm birth and ASD (Johnson et al., 2010; Limperopoulos et al., 2008), this result has not been universally found (Glasson et al., 2004; Hultman et al., 2002). Interestingly, male sex is a risk factor for preterm birth (Di Renzo et al., 2007), and an increased male:female ratio has been consistently observed in ASD (Werling and Geschwind, 2013). The objective of this study was to characterize the phenotype of males and females with ASD who were born preterm compared to those born ≥37 weeks’ gestation. Going forward, those born ≥37 weeks’ gestation will be referred to as “term,” which is inclusive of individuals who were born post-term.

Methods

Study population

Participants were part of a larger sample of subjects with ASD involved in a comprehensive longitudinal assessment of medication management. The study was approved by the Institutional Review Board at Indiana University. Participants included consecutive individuals, ≤18 years or age referred for assessment and treatment at an academic tertiary care ASD center between July 2004 and April 2012. Subjects met Diagnostic and Statistical Manual of Mental Disorders (4th ed., text rev.; DSM-IV-TR) criteria for autistic disorder, pervasive developmental disorder not otherwise specified (PDD-NOS), or Asperger’s disorder. For each subject, demographic characteristics, anthropometrics (e.g. weight, height, and body mass index (BMI)), ASD diagnostics subtype, psychiatric and medical comorbidities, as well as detailed information on target behavioral symptoms, medications prescribed, medication dose, treatment duration, and adverse effects were collected at baseline. Intellectual disability data were collected from clinical interviews, as well as review of neuropsychological testing and school reports when available (Wink et al., 2014).

Variable definitions

Subjects were considered preterm if their gestational age at birth was less than 37 weeks and were considered term if their gestational age at birth was ≥37 weeks. Gestational age was based on parent/guardian report and medical records when available. Parents or guardians were asked “Was your/the child born term or preterm, at how many weeks’ gestation?” and “Did your/the child require resuscitative measures or neonatal intensive care unit (NICU) stay after birth?” BMI z scores and percentiles were calculated using 2000 Centers for Disease Control (CDC) growth chart standards (Kuczmarski et al., 2002) by employing a SAS program macro provided by the CDC (http://www.cdc.gov/nccdphp/dnpao/growthcharts/resources/sas.htm). Nonverbal was defined as not able to say any words and/or if described in the medical chart as nonverbal. Significant language delay was defined as saying only two to three words by 2 years of age and/or if described as “language delay” or “speech delay” in the medical chart. Language regression was designated if the child was speaking any words at all and then either stopped speaking or speech was reduced to only a few words. In addition, a child was determined to have language regression if described in the medical chart.

Statistical methods

All statistical analyses were performed with SAS software version 9.3 (SAS Institute, Cary, NC, USA). Subjects were excluded if they were diagnosed with known genetic anomalies. Means with standard deviations for continuous variables and frequencies and percents were calculated for each study group. A chi-square test or t-test determined whether phenotypic characteristics were significantly different between across study groups. Analyses were also conducted separately for each sex to determine whether observed differences between preterm and term subjects were evident among males and females. If a statistical difference between preterm and term was identified among males, but not females (or vice versa), a direct statistical comparison was conducted. Logistic regression (for dichotomous characteristics) and linear regression (for continuous characteristics) were employed with the phenotypic characteristic as the dependent variable. Models included sex (male/female), preterm (yes/no), and an interaction term between sex and preterm. The p values for the interaction of <0.05 would suggest the observed differences for preterm versus term varied by sex.

Results

Data were collected on 1069 individuals with a diagnosis of ASD. After excluding those who were >18 years of age (n = 109), had fragile X syndrome (n = 58), Angelman syndrome (n = 3), Down syndrome (n = 7), mitochondrial disorders (n = 2), tuberous sclerosis (n = 1), and other genetic anomalies (n = 6), the analytic population included a total of 883 individuals, with 115 (13.0%) born preterm. Comparing individuals born preterm to those born at term (Table 1), there was a statistically significant difference between the frequency of sleep apnea (13.0% vs 3.0% for preterm vs term, respectively, p < 0.0001) that when stratified by sex was significantly higher among males (13.8% vs 2.5%, p < 0.0001), but not females (9.5% vs 5.3%, p = 0.44) (Table 2). Similarly, seizure disorders were more frequent among males born preterm versus term (17.0% vs 8.5%, p = 0.01), and attention-deficit hyperactivity disorder (ADHD) (14.9% vs 6.6%, p = 0.005) was again significantly different only among males. Additional differences were observed for factors related to language. Subjects born preterm were more likely to be nonverbal (9.6% vs 4.6%, p = 0.02), and this effect was statistically significant among females (23.8% vs 3.8%, p = 0.001), but not males (6.4% vs 4.7%, p = 0.49). Finally, anthropometric differences were observed. Individuals born preterm had a significantly lower average weight compared with those born at term (70.0 vs 81.7 lbs, p = 0.02), though the lower average height, BMI, and BMI z scores were not statistically different. In a direct comparison of males and females, only the difference between nonverbal status for term versus preterm differed by sex (p interaction = 0.03), data not shown.

Table 1.

Demographic, comorbidity, and risk factor differences in individuals born preterm and those born at term with autism spectrum disorders.

	Full cohort (n = 883)	Preterm (n = 115)	Term (n = 768)	p value
Characteristics
Age (years)	8.3 (3.8)	8.3 (3.7)	8.3 (3.9)	0.93
Sex (%male)	728 (82.6)	94 (81.7)	635 (82.7)	0.80
Race				0.78
Asian	9 (1.0)	0	9 (1.2)
Native Hawaiian/Pacific Islander	1 (0.1)	0	1 (0.1)
Black/African American	78 (9.0)	12 (10.5)	66 (8.7)
White	654 (75.3)	88 (77.2)	566 (75.0)
More than one race	15 (1.7)	2 (1.8)	13 (1.7)
Unknown/not reported	112 (12.9)	12 (10.5)	100 (13.3)
Significant language delay	495 (56.1)	70 (60.9)	425 (55.3)	0.27
Language regression	138 (15.6)	17 (14.8)	121 (15.8)	0.79
Nonverbal	46 (5.2)	11 (9.6)	35 (4.6)	0.02
ASD diagnosis				0.50
Autistic disorder	430 (48.8)	59 (51.3)	371 (48.4)
PDD-NOS	363 (41.2)	48 (41.7)	315 (41.1)
Asperger’s disorder	88 (10.0)	8 (7.0)	80 (10.4)
BMI	19.3 (5.4)	18.3 (4.8)	19.4 (5.5)	0.07
Weight	80.1 (46.3)	70.0 (39.6)	81.7 (47.1)	0.02
Height	52.3 (9.4)	50.6 (9.5)	52.6 (9.3)	0.06
BMI z score^a	−0.07 (8.6)	−0.66 (11.4)	0.02 (8.1)	0.47
BMI categories^b				0.66
Underweight	197 (23.4)	21 (19.4)	176 (24.0)
Healthy weight	367 (43.6)	52 (48.2)	315 (42.9)
Overweight	103 (12.2)	14 (13.0)	89 (12.1)
Obese	175 (20.8)	21 (19.4)	154 (19.4)
Comorbidities
Mental retardation diagnosis				0.28
None	558 (63.4)	63 (54.8)	495 (64.7)
Mild	68 (7.7)	10 (8.7)	58 (7.6)
Moderate	55 (6.3)	10 (8.7)	45 (5.9)
Severe	15 (1.7)	3 (2.6)	12 (1.6)
Profound	2 (0.2)	0	2 (0.3)
Borderline intellectual functioning	37 (4.2)	4 (3.5)	33 (4.3)
Unspecified	120 (13.6)	23 (20.0)	97 (12.7)
Unknown	25 (2.8)	2 (1.7)	23 (3.0)
Seizure disorder	82 (9.3)	18 (15.7)	64 (8.3)	0.01
Chronic ear infections	212 (24.0)	27 (23.5)	185 (24.1)	0.89
Sleep apnea	38 (4.3)	15 (13.0)	23 (3.0)	<0.0001
Depression	28 (3.2)	4 (3.5)	24 (3.1)	0.84
Bipolar disorder	22 (2.5)	2 (1.7)	20 (2.6)	0.58
Anxiety disorder	46 (5.2)	8 (7.0)	38 (5.0)	0.37
OCD	17 (1.9)	1 (0.9)	16 (2.1)	0.38
Schizophrenia	3 (0.3)	0	3 (0.4)	0.50
Tic disorder	29 (3.3)	3 (2.6)	26 (3.4)	0.66
ADHD	62 (7.0)	15 (13.0)	47 (6.1)	0.01
Echolalia	356 (40.3)	44 (38.3)	312 (40.6)	0.63
Risk factors
Drug exposure in utero	211 (23.9)	35 (30.4)	176 (22.9)	0.08
First degree relative with ASD	100 (11.3)	17 (14.8)	839 10.8)	0.21
Relative with language disorder	50 (5.7)	7 (6.1)	43 (5.6)	0.83

ASD: autism spectrum disorder; PDD-NOS: pervasive developmental disorder not otherwise specified; BMI: body mass index; OCD: obsessive–compulsive disorder; ADHD: attention-deficit hyperactivity disorder.

Values presented are means (standard deviations) for continuous variables and number (%) for categorical variables.

BMI z scores calculated for individuals between the ages of 2 and 20 years and who have data on height and weight using CDC Growth Charts.

Underweight <5th percentile, healthy weight 5th to 84th percentile, overweight 85th to 95th percentile, and obese >95th percentile.

Table 2.

Characteristics of patients born preterm and those born at term among males and females with ASD.

	Females			Males
	Preterm (n = 21)	Term (n = 133)	p value	Preterm (n = 94)	Term (n = 635)	p value
Characteristics
Age (years)	7.8 (3.4)	8.5 (4.0)	0.42	9.4 (8.4)	8.3 (3.8)	0.78
Race			0.003			0.52
Asian	0	0		0	9 (1.40
Native Hawaiian/Pacific Islander	0	0		0	1 (0.2)
Black/African American	0	10 (7.6)		11 (11.8)	56 (9.0)
White	1 (4.8)	100 (76.3)		71 (76.4)	466 (74.7)
More than one race	17 (81.0)	0		0	13 (2.1)
Unknown/not reported	2 (9.5)	21 (16.0)		11 (11.8)	79 (12.7)
Missing	0	2 (1.5)		1 (1.1)	11 (1.70
Language delay	13 (61.9)	71 (53.4)	0.47	57 (60.6)	354 (55.8)	0.37
Language regression	4 (19.1)	22 (16.5)	0.78	13 (13.8)	99 (15.6)	0.66
Nonverbal	5 (23.8)	5 (3.8)	0.001	6 (6.4)	30 (4.7)	0.49
ASD diagnosis			0.41			0.61
Autistic disorder	10 (47.6)	69 (51.9)		49 (52.1)	302 (47.7)
PDD-NOS	11 (52.4)	56 (42.1)		37 (39.4)	259 (40.9)
Asperger’s disorder	0	8 (6.0)		8 (8.5)	72 (11.4)
BMI	18.5 (4.9)	20.0 (5.5)	0.30	18.3 (4.7)	19.3 (5.5)	0.12
Weight	61.9 (30.7)	84.1 (44.9)	0.04	71.9 (41.4)	81.1 (47.6)	0.10
Height	46.7 (8.7)	53.0 (8.7)	0.006	51.5 (9.5)	52.5 (9.5)	0.41
BMI z score	0.54 (1.7)	0.61 (1.8)	0.88	−0.93 (12.6)	−0.10 (8.8)	0.47
BMI categories						0.72
Underweight	4 (20.0)	30 (23.6)	0.74	17 (19.3)	146 (24.1)
Healthy weight	9 (45.0)	53 (41.7)		43 (48.9)	262 (43.2)
Overweight	4 (20.0)	16 (12.6)		10 (11.4)	73 (12.0)
Obese	3 (15.0)	28 (22.1)		18 (20.5)	126 (20.8)
Comorbidities
Mental retardation diagnosis			0.20			0.26
None	9 (42.9)	73 (55.3)		54 (57.5)	422 (66.7)
Mild	2 (9.5)	16 (12.1)		8 (8.5)	42 (6.6)
Moderate	1 (4.8)	8 (6.1)		9 (9.6)	37 (5.9)
Severe	3 (14.3)	4 (3.0)		0	8 (1.3)
Profound	0	0		0	2 (0.3)
Borderline intellectual functioning	0	9 (6.8)		4 (4.3)	24 (3.8)
Unspecified	5 (23.8)	19 (14.4)		19 (19.2)	78 (12.3)
Unknown	1 (4.8)	3 (2.3)		1 (1.1)	20 (3.2)
Seizure disorder	2 (9.5)	10 (7.5)	0.75	16 (17.0)	54 (8.5)	0.01
Chronic ear infections	4 (19.1)	32 (24.1)	0.61	23 (24.5)	153 (24.1)	0.94
Sleep apnea	2 (9.5)	7 (5.3)	0.44	13 (13.8)	16 (2.5)	<0.0001
Depression	0	1 (0.8)	0.69	4 (4.3)	23 (3.6)	0.76
Bipolar	0	2 (1.5)	0.57	2 (2.1)	18 (2.8)	0.70
Anxiety	2 (9.5)	7 (5.3)	0.44	6 (6.4)	31 (6.4)	0.54
OCD	0	1 (90.8)	0.69	1 (1.1)	15 (2.4)	0.42
Schizophrenia	0	0	N/A	0	3 (0.5)	0.50
Tic disorder	0	4 (3.0)	0.42	3 (3.2)	22 (3.5)	0.89
ADHD	1 (4.8)	5 (3.8)	0.83	14 (14.9)	42 (6.6)	0.005
Echolalia	7 (33.3)	57 (42.9)	0.41	37 (39.4)	255 (40.2)	0.88
Risk factors
Drug exposure in utero	7 (33.3)	30 (22.6)	0.28	28 (30.0)	146 (23.0)
First degree relative with ASD	5 (23.8)	15 (11.3)	0.11	12 (12.8)	68 (10.7)	0.15
Relative with language disorder	2 (9.5)	9 (6.8)	0.65	5 (5.3)	34 (5.4)	0.55
						0.99

Values presented are means (standard deviations) for continuous variables and number (percent) for categorical variables.

Discussion

For a number of variables and phenotypic characteristics examined, individuals with ASD who were born preterm compared with those born at term gestation were not appreciably different. However, some potentially meaningful differences were found. A greater proportion of preterm males reported sleep apnea, seizure disorders, and ADHD. Finally, preterm subjects, particularly females, were more often nonverbal. More research is necessary to explore these potential differences further and to determine whether they may be important areas for early and targeted intervention.

Sleep disorders are prevalent among children with ASD, with up to 83% of children affected (Richdale, 1999). While the prevalence of sleep apnea specifically is unclear, a prior report found 3.4% of parents reported apnea (Gail Williams et al., 2004), which is consistent with the 3% of individuals in this study affected by sleep apnea who were born at term. However, this is in contrast to the 13.0% of subjects identified with sleep apnea among individuals born preterm. Prematurity (independent of ASD) is associated with sleep-disordered breathing. Prevalence estimates derived from cardiorespiratory monitoring during sleep identified an obstructive apnea index ≥1 event per hour in 6.1% of children aged 8 to 11 years who were born preterm (Rosen et al., 2003). Since sleep apnea is related to cognitive and behavioral deficits in neurotypical children, potentially due to cellular injury of the central nervous system (Beebe and Gozal, 2002), addressing sleep apnea could be especially vital for children with ASD. In fact, a case report of a child treated for sleep apnea found improvement in autistic symptoms, measured by parent report and a more objective measure, the Autism Diagnostic Observation Schedule (ADOS) (Malow et al., 2006). If additional studies confirm the high prevalence of sleep apnea among individuals with ASD born preterm, treatment of sleep apnea may be an important target for these children.

A recent change with the introduction of the Diagnostic and Statistical Manual of Mental Disorders (5th ed.; DSM-5), allows for the co-occurrence of ASD and ADHD. Recent studies suggest preterm birth is a common risk factor for both ASD and ADHD (Johnson and Marlow, 2011; Linnet et al., 2006). Therefore, it is interesting that the comorbidity of these disorders was found to be higher among subjects with ASD born preterm in this study. It should be noted that prevalence of ADHD in term infants in the present analysis is lower than 2011 estimates for the United States (11.0%) and the state of Indiana (13.0%) (Visser et al., 2014). The relationship between ASD and ADHD and a potential contribution from preterm birth is an important area for follow-up studies.

Seizure disorders were also more prevalent among males with ASD born preterm compared with term. However, the overall prevalence of 9.3% and the higher prevalence among males (17.0%) did not reach the 20%–25% conservative estimate for comorbid epilepsy often reported in ASD (Canitano, 2007). These numbers should therefore be interpreted with caution, as seizure disorders may be underreported in this sample. Alternatively, the mean age of males in this sample is 8.3 years and it is possible that many of the participants had not reached the second peak age of seizure onset that has been described in ASD (Tuchman et al., 2010). It should be noted that Table 2 does not include direct statistical comparisons by sex. When compared directly, the difference between term and preterm for seizure disorders, sleep apnea, and ADHD did not vary significantly by sex (p_interaction = 0.54, 0.20, 0.57, respectively). However, there were few preterm females, limiting our ability to find statistically significant differences.

The preterm group in this study was more likely to be nonverbal especially among females. While there were only 27 preterm females, the difference was statistically significant. In addition, the difference for preterm versus term varied significantly by sex (p_interaction = 0.03). However, for other measures of language impairment (including language regression or delay), there were no differences observed. Again, confirmation from longitudinal follow-up studies will be important to determine whether prematurity is a risk factor for nonverbal status. It will also be important to determine what factors associated with prematurity may affect the development of speech. Doing so may aid the understanding of the etiology of communication deficits associated with ASD. This could lead to enhanced early targeted treatment for those individuals identified as high-risk for nonverbal status.

Several limitations should be considered when interpreting the results of this study. Despite having a large longitudinal registry, the sample was relatively small for making direct comparisons by sex, given the low number of females, especially those born preterm. However, given the potential for differences by gender, we felt it was appropriate to present the stratified results. The majority of subjects were not administered the ADOS, a gold standard diagnostic test. However, all diagnoses were based upon DSM-IV-TR diagnostic criteria by clinicians with years of experience in caring for individuals with ASD across the lifespan. The data are included in a registry of all participants presenting to a tertiary care clinic and not a population-based sample. This suggests that the results may not be generalizable to all individuals with ASD. For example, individuals may have been referred to this academic ASD center if their symptoms were more severe or if they had symptoms more amenable to pharmacologic treatments. In addition, data were missing for several variables of interest. For example, full-scale and verbal IQ scores were available for only a small proportion of individuals and measures of head circumference were unavailable. Varying levels of risk of ASD have been observed for infants born preterm compared with very early preterm. Data describing the exact gestational age were not available for these analyses. It would be important in future studies to identify phenotypic differences between children born early preterm (<32 weeks), preterm (32.1–34 weeks), and late preterm (34.1–37 weeks), and even studies to identify phenotypic differences between infants born between 37 and 39 weeks would be warranted. However, the results remain important as 37 weeks is still a common clinical milestone for labeling a baby as preterm. Although data were acquired from clinical interview and parent reports, which may be affected by diminished recall, medical records were used to confirm data, when available. In addition, for our primary variable of interest, preterm birth, parent report has been demonstrated to be accurately reported (Daly et al., 1994). Finally, we performed a large number of statistical tests to compare differences between the two groups. Given the number of tests conducted, it is likely that some statistically significant differences between the groups arose simply by chance. We have chosen to present unadjusted p values given the exploratory nature of these analyses and suggest confirmation in independent samples.

The strengths of the study include the large sample of individuals with ASD and detailed data describing their phenotypic characteristics and comorbidities. While the majority of phenotypic characteristics examined did not vary significantly between subjects with ASD born preterm versus those born at term, several potentially meaningful differences were observed that should be explored further in independent datasets. In addition, studies will require innovative methods to identify the causal role (or lack of) of preterm birth on the development of ASD.

Footnotes

Funding

This research received no specific grant from any funding agency in the public, commercial or not-for-profit sectors.

References

Beebe

Gozal

(2002) Obstructive sleep apnea and the prefrontal cortex: towards a comprehensive model linking nocturnal upper airway obstruction to daytime cognitive and behavioral deficits. Journal of Sleep Research 11: 1–16.

Canitano

(2007) Epilepsy in autism spectrum disorders. European Child & Adolescent Psychiatry 16: 61–66.

Daly

Lindgren

Giebink

(1994) Validity of parental report of a child’s medical history in otitis media research. American Journal of Epidemiology 139: 1116–1121.

Di Renzo

Rosati

Sarti

. (2007) Does fetal sex affect pregnancy outcome? Gender Medicine 4: 19–30.

Eaton

Mortensen

Thomsen

. (2001) Obstetric complications and risk for severe psychopathology in childhood. Journal of Autism and Developmental Disorders 31: 279–285.

Gail Williams

Sears

Allard

(2004) Sleep problems in children with autism. Journal of Sleep Research 13: 265–268.

Glasson

Bower

Petterson

. (2004) Perinatal factors and the development of autism: a population study. Archives of General Psychiatry 61: 618–627.

Hultman

Sparen

Cnattingius

(2002) Perinatal risk factors for infantile autism. Epidemiology 13: 417–423.

Johnson

Hollis

Kochhar

. (2010) Autism spectrum disorders in extremely preterm children. Journal of Pediatrics 156: 525–531.e2.

10.

Johnson

Marlow

(2011) Preterm birth and childhood psychiatric disorders. Pediatric Research 69: 11R–18R.

11.

Kolevzon

Gross

Reichenberg

(2007) Prenatal and perinatal risk factors for autism: a review and integration of findings. Archives of Pediatrics & Adolescent Medicine 161: 326–333.

12.

Kuczmarski

Ogden

Guo

. (2002) 2000 CDC growth charts for the United States: methods and development. Vital and Health Statistics 11: 1–190.

13.

Larsson

Eaton

Madsen

. (2005) Risk factors for autism: perinatal factors, parental psychiatric history, and socioeconomic status. American Journal of Epidemiology 161, 916–925; discussion 926–928.

14.

Limperopoulos

Bassan

Sullivan

. (2008) Positive screening for autism in ex-preterm infants: prevalence and risk factors. Pediatrics 121: 758–765.

15.

Linnet

Wisborg

Agerbo

. (2006) Gestational age, birth weight, and the risk of hyperkinetic disorder. Archives of Disease in Childhood 91: 655–660.

16.

Malow

Mcgrew

Harvey

. (2006) Impact of treating sleep apnea in a child with autism spectrum disorder. Pediatric Neurology 34: 325–328.

17.

Richdale

(1999) Sleep problems in autism: prevalence, cause, and intervention. Developmental Medicine & Child Neurology 41: 60–66.

18.

Rosen

Larkin

Kirchner

. (2003) Prevalence and risk factors for sleep-disordered breathing in 8- to 11-year-old children: association with race and prematurity. Journal of Pediatrics 142: 383–389.

19.

Saigal

Doyle

(2008) An overview of mortality and sequelae of preterm birth from infancy to adulthood. Lancet 371: 261–269.

20.

Tuchman

Cuccaro

Alessandri

(2010) Autism and epilepsy: historical perspective. Brain & Development 32: 709–718.

21.

Visser

Danielson

Bitsko

. (2014) Trends in the parent-report of health care provider-diagnosed and medicated attention-deficit/hyperactivity disorder: United States, 2003–2011. Journal of the American Academy of Child & Adolescent Psychiatry 53: 34–46.e2.

22.

Werling

Geschwind

(2013) Sex differences in autism spectrum disorders. Current Opinion in Neurology 26: 146–153.

23.

Wink

Early

Schaefer

. (2014) Body mass index change in autism spectrum disorders: comparison of treatment with risperidone and aripiprazole. Journal of Child and Adolescent Psychopharmacology 24: 78–82.