Advertisement

How has DSM-5 Affected Autism Diagnosis? A 5-Year Follow-Up Systematic Literature Review and Meta-analysis

  • Kristine M. KulageEmail author
  • Johanna Goldberg
  • John Usseglio
  • Danielle Romero
  • Jennifer M. Bain
  • Arlene M. Smaldone
OriginalPaper

Abstract

We conducted a 5-year follow-up systematic review and meta-analysis to determine change in frequency of autism spectrum disorder (ASD) diagnosis since diagnostic and statistical manual 5 (DSM-5) publication and explore the impact of Social Communication Disorder (SCD). For 33 included studies, use of DSM-5 criteria suggests decreases in diagnosis for ASD [20.8% (16.0–26.7), p < 0.001], DSM-IV-TR Autistic Disorder [10.1% (6.2–16.0), p < 0.001], and Asperger’s [23.3% (12.9–38.5), p = 0.001]; pervasive developmental disorder-not otherwise specified decrease was not significant [46.1% (34.6–58.0), p = 0.52]. Less than one-third [28.8% (13.9–50.5), p = 0.06] of individuals diagnosed with DSM-IV-TR but not DSM-5 ASD would qualify for SCD. Findings suggest smaller decreases in ASD diagnoses compared to earlier reviews. Future research is needed as concerns remain for impaired individuals without a diagnosis.

Keywords

Autism Spectrum Disorder DSM-5 Diagnosis Asperger’s Disorder PDD-NOS Social Communication Disorder 

Introduction

Autism Spectrum Disorder (ASD) was first established as a unique diagnosis from schizophrenia in 1980 in the Third Edition of the Diagnostic and Statistical Manual of Mental Disorders (DSM)—the clinical diagnostic standard for mental disorders, including development disorders. Prior to 1980, the prevalence of autism estimated both in the United States (US) and globally ranged from 0.07 to 0.31 (Treffert 1970) to 0.49 (Wing and Gould 1979) per 1000 children. When the DSM, Fourth Edition, Text-Revision (DSM-IV-TR) was published in 2000 (American Psychiatric Association 2000), data from the first surveillance year (2000) of the Autism and Developmental Disabilities Monitoring (ADDM) Network estimated an ASD prevalence rate of 6.7 per 1000 or 1 in 150 children aged 8 years (Rice and Autism and Developmental Disabilities Monitoring Network Surveillance Year 2000 Principal Investigators 2007), a finding similar to that reported by Mattila et al. in a study of Finnish children (Mattila et al. 2011). The most recent estimate from the ADDM Network (2014) illustrates a further increase in prevalence to 16.8 per 1000 or 1 in 59 American children (Baio et al. 2018) and is consistent with estimates of the increase in diagnosis rate obtained by parent self-report via national surveys (Kogan et al. 2018; Schieve et al. 2006). While estimates by country and the methods by which they are derived may vary, the increasing prevalence of autism as a global issue clear (Adak and Halder 2017; Elsabbagh et al. 2012; Fombonne et al. 2009; Levy et al. 2009). Collectively, this has prompted public health concerns, an expansion of research efforts, and a continued need for services (Baio et al. 2018).

Changes in the criteria for autism diagnosis published in the Fifth Edition of the DSM (DSM-5) (American Psychiatric Association 2013a) have stimulated much debate. First, the DSM-IV-TR contained ASD subtypes of Autistic Disorder (AD), Asperger’s Disorder, and pervasive developmental disorder-not otherwise specified (PDD-NOS) that were omitted in the DSM-5; instead, subtypes were collapsed into a single diagnostic category—ASD. The DSM-5 also reduced the core domains of impairment from three to two: (1) social interaction and social communication (previously two distinct categories of “social interaction” and “communication”) and (2) restricted, repetitive patterns of behavior, interests, or activities. In addition, while the DSM-IV-TR contained 12 distinct diagnostic criteria, the DSM-5 outlines only seven which are more general principles and behaviors. Finally, the DSM-5 allows for inclusion of historical behaviors in the ASD criteria, with the caveat that these behaviors must have been present in the early developmental period, while the previous edition was limited to current behaviors. Overall, these changes have caused concern that a higher threshold of symptoms is required for DSM-5 ASD diagnosis, thereby failing to capture some individuals who would have previously been diagnosed with ASD under the DSM-IV-TR and who may benefit from access to treatment and services (Maenner et al. 2014). Notably, while ADDM Network data on autism rates released just prior to publication of the DSM-5 identified a prevalence of 1 in 88 children aged 8 years old (Centers for Disease Control and Prevention 2012), the most recent ADDM Network prevalence estimate since DSM-5 publication was 1 in 59 children (Baio et al. 2018). However, data for this latest report are from 2014, and children included in this analysis would have primarily been evaluated under DSM-IV-TR ASD criteria (Baio et al. 2018). Therefore, the impact of DSM-5 criteria on ASD diagnosis rates remains unknown.

To date, three systematic literature reviews (one with a meta-analysis) which examined the potential impact of DSM-5 on ASD diagnosis rates have been published; two were conducted just prior to DSM-5 publication (Kulage et al. 2014; Sturmey and Dalfern 2014), and one was conducted a year after (Smith et al. 2015). All three determined that ASD rates could decrease by at least one-third. While numerous studies have quantified potential changes in ASD rates in the last 5 years, no new systematic literature reviews with meta-analyses have been conducted to synthesize data from studies comparing DSM-IV-TR and DSM-5 ASD rates. In addition, the impact of a new DSM-5 diagnosis, Social Communication Disorder (SCD)—defined as a primary deficit in social communication and interaction (SCI) without restrictive, repetitive behaviors (RRB) (Ohashi et al. 2015; Sumi et al. 2014; Swineford et al. 2014)—on ASD rates has not been specifically examined in a systematic review since DSM-5 publication. This is an important gap in the literature because not only must an ASD diagnosis be “ruled out” before an SCD diagnosis can be given, but SCD was also initially described by the American Psychiatric Association as potentially capturing individuals with symptoms of PDD-NOS but who would no longer meet criteria for ASD under DSM-5 (American Psychiatric Association 2013b).

To address these gaps on the impact of DSM-5 on ASD diagnosis rates, the aims of this follow-up systematic literature review and meta-analysis were to: (1) determine the change in frequency of ASD diagnosis in the first five years after publication of the revised DSM-5 ASD criteria; (2) identify the DSM-IV-TR autism subtypes most affected by the new criteria; and (3) assess the potential of an alternative diagnosis of SCD for individuals who meet DSM-IV-TR but not DSM-5 ASD diagnostic criteria.

Methods

Search Strategy and Inclusion Criteria

We followed the Preferred Reporting Items for Systematic Reviews and Meta-Analyses (PRISMA) guidelines (Moher et al. 2009) in conducting this literature review and meta-analysis. An a priori protocol was registered (PROSPERO 2017 CRD42017077533) in November 2017 and updated in October 2018; the protocol can be accessed from http://www.crd.york.ac.uk/PROSPERO/display_record.php?ID=CRD42017077533. We used Covidence (http://covidence.org), the web-based production platform for Cochrane Reviews, to manage our work flow. On October 26, 2017, we searched MEDLINE (PubMed), the Cumulative Index to Nursing and Allied Health Literature (EBSCO), Education Resources Information Center (ProQuest), and PsycInfo (Ovid) for original studies published from April 1, 2013, the end of coverage of the first literature review on this topic, through December 31, 2017. Subsequently, we re-ran the search on July 11, 2018 for studies published between January 1, 2018 and June 30, 2018. For search terms, two main domains were combined with the AND operator: one relating to DSM-5 and the other to autism diagnoses (e.g., Asperger’s) or other related diagnoses (e.g., SCD). The full search strategy by database is available online in Appendix 1. Both subject headings and free text were used. No language requirement was placed on the text. To supplement the database search, we hand-searched issues of the Review Journal of Autism and Developmental Disorders and conference proceedings of the International Society for Autism Research from 2013 to 2017. We conducted a grey literature search for conference proceedings in both BIOSIS and Embase and examined .gov and .org sites for seven pages of search results on Google.com.

All items found in the literature during the identification phase were screened by at least two authors who examined titles and abstracts for two inclusion criteria: studies needed to (1) present original data and (2) compare application of DSM-IV-TR and DSM-5 ASD diagnostic criteria to populations at risk for or previously diagnosed with ASD and/or one of three DSM-IV-TR ASD subtypes (AD, Asperger’s disorder, or PDD-NOS). If it was unclear whether a study met these criteria based on abstract review, we conservatively included the study for full-text review. During full-text review, at least two authors assessed each study and came to a consensus for inclusion based on the following criteria: studies needed to (1) report results as raw data or percentages of individuals meeting diagnostic criteria using both DSM-IV-TR and DSM-5 criteria separately or (2) provide sufficient information so that percentages could be calculated (for example, present DSM-5 sensitivity and specificity with DSM-IV-TR as the reference standard). We excluded studies if they (1) did not compare DSM-IV-TR and DSM-5 diagnostic criteria applied to the same population; (2) did not provide sufficient information for extracting raw data on changes in rates of ASD diagnoses under DSM-IV-TR as compared to DSM-5; (3) had been included in the first literature review and meta-analysis on this topic (Kulage at al. 2014); (4) examined a duplicate study sample; or (5) used an inappropriate study design/article type for purposes of this review (i.e., editorials, letters to the editor, case reports, review articles, qualitative studies, or summaries or press releases of another article). We then hand-searched reference lists of included studies to locate other studies that may not have been identified in the electronic search.

Data Extraction

Two authors independently extracted data from each study and four authors compared results to arrive at a consensus. We extracted the following study characteristics: continent; study design; data sources; funding information; sample size; sample demographics including gender, race, and ethnicity; number diagnosed with ASD and/or its subtypes under DSM-IV-TR criteria; the version of DSM-5 ASD diagnostic criteria used in the study (i.e., draft or final); the discipline of the rater(s) responsible for making the autism diagnosis; and the instruments used by raters. The change in frequency of ASD diagnosis when DSM-5 criteria were applied to the same sample and/or subsamples was then calculated, including number and percent reduction in diagnosis. For studies which examined SCD, we extracted information on the number of individuals with ASD and its subtypes under DSM-IV-TR criteria who did not meet DSM-5 criteria but would qualify for an alternative diagnosis of SCD. Finally, we collected data from studies which reported specificity and sensitivity of DSM-5 diagnostic criteria.

Quality Appraisal

To rate the scientific rigor of individual studies, we used the quality appraisal of reliability studies (QAREL) (Lucas et al. 2010) which was developed for use in systematic reviews and meta-analyses to assess the quality of studies which explore diagnostic reliability. This 11-item checklist examines seven principles including the appropriateness of subjects, qualification of examiners, examiner blinding, ordering of examination, suitability of the time interval between repeated measurements, appropriate test application and interpretation, and statistical analysis of intra or inter-rater agreement. Each QAREL item can be answered with “yes,” “no,” or “unclear,” with five items also including “not applicable” as an option. When raters agree upon the interpretation of criteria for each item, the QAREL has been demonstrated to be a reliable assessment tool for studies of diagnostic reliability (Lucas et al. 2013). In this study, two authors independently rated each study using QAREL, and then four authors collectively reviewed results and came to a consensus on each item.

Data Analysis

We conducted three meta-analyses. In the first pooled analysis, all included studies were examined to determine the change in frequency of ASD diagnosis based on DSM-5 criteria. For the second pooled analysis, we included studies that explored differences in ASD diagnosis by DSM-IV-TR subtype. For each, data were extracted as the number of individuals meeting DSM-IV-TR ASD diagnostic criteria and the number no longer meeting ASD diagnostic criteria under DSM-5; we then computed the proportion of those who would not retain an ASD diagnosis. Pooled effects were estimated for the proportion of individuals who no longer met criteria for ASD diagnosis using a random effects meta-analysis model. For the third meta-analysis, we pooled data from studies that examined application of DSM-5 SCD criteria to ASD samples. Specifically, we extracted the number of individuals who met DSM-IV-TR ASD criteria but no longer met criteria for an ASD diagnosis under DSM-5 and, of those, the number who would alternatively meet criteria for SCD. Because of the small number of studies, and to obtain a more comprehensive assessment of the impact of the SCD diagnosis and its potential to capture these individuals, we also extracted the same data from the four studies that examined SCD that were included in the first review on this topic (Kulage et al. 2014). A pooled effect was estimated for the proportion of individuals who would meet criteria for SCD. Results are presented as forest plots using random effects meta-analysis models.

For pooled effects indicative of a statistically significant reduction (p < 0.05) in diagnoses when DSM-5 criteria were applied, we examined heterogeneity and publication bias. Heterogeneity was assessed using Cochran’s Q and I2 statistics and was considered to be present if the Cochran’s Q p-value was < 0.05 or I2 was > 50% (Higgins et al. 2003). To examine differences between studies that might explain heterogeneity, we conducted subgroup analyses by sample age; continent where the study was conducted; study design; instrument used to make an ASD diagnosis; discipline of the rater (MD, PhD, or both) responsible for making the diagnosis; version of DSM-5 ASD diagnosis criteria used (draft or final); study funding source; and three risk of bias domains: whether order of examination varied, measurement of intra and/or interrater agreement, and whether raters making the diagnosis were blinded to the results of the reference standard (i.e., DSM-IV-TR diagnosis). To examine the risk of publication bias, we constructed a funnel plot, examined it visually, and conducted a Classic fail-safe N test, which is used to determine the number of additional studies needed to change interpretation of publication bias (Persaud 1996). Data were analyzed using Comprehensive Meta-Analysis statistical software (Biostat, Inc., Englewood, NJ).

Results

Figure 1 presents details of the literature search. A total of 898 records were initially identified from the database and supplemental search phases; following removal of duplicates, 600 articles were deemed eligible for screening. After screening titles and abstracts, 509 items were excluded, leaving an initial group of 91 studies for full-text assessment. However, prior to full-text assessment, the reference lists of the 91 studies were hand-searched, and two additional publications were identified, creating a total of 93 for full-text review. Sixty studies were subsequently excluded after the full-text review, including seven which used the same sample as a study (Matson et al. 2012) that was included in the first review on this topic (See Appendix 2 online for list of excluded references and rationale for exclusion). Therefore, a total of 33 studies were included in the systematic review and meta-analysis; of these, 19 studies that examined ASD subtypes and nine studies that examined SCD (five studies identified in this review and four studies from the previous review) were eligible for the additional analyses.

Fig. 1

PRISMA flow diagram for the systematic literature review

Study Quality

Figure 2 summarizes the results of the quality appraisal of the 33 studies. All but one study (Kim et al. 2014) used an appropriate sample of subjects. In the majority of studies, appropriately credentialed raters provided diagnoses, correctly applied and interpreted the instruments or criteria for diagnoses, and employed an appropriate time-interval between DSM-IV-TR and DSM-5 measurement. However, only eight studies (Baio et al. 2018; Helles et al. 2015; Hiller et al. 2014; Kim et al. 2014; Mazurek et al. 2017; Mugzach et al. 2015; Taheri et al. 2014; Young and Rodi 2014) reported inter and/or intra-rater reliability, and variation in the order of examination could only be verified in three studies (Mazurek et al. 2017; Mugzach et al. 2015; Young and Rodi 2014). The risk for bias in relation to study blinding across studies was largely unclear. Only one study specified that raters were blinded to the findings of other raters (Wong and Koh 2016), and no studies definitively indicated that raters were blinded to their own prior findings. In addition, in only four studies could we determine that the raters were blinded to both clinical information and additional cues not part of the diagnosing process (Mazurek et al. 2017; Taheri et al. 2014; Turygin et al. 2013; Wong and Koh 2016), and only three studies reported that raters were blinded to the results of DSM-IV-TR when applying DSM-5 criteria (Magana and Vanegas 2017; Sung et al. 2018; Wong and Koh 2016).

Fig. 2

Study quality appraisal results using the QAREL checklist

Characteristics of the Included Studies

Study Year, Type, and Continent

Table 1 provides a descriptive summary of each study. Publication years for articles ranged from 2013 to 2018 with the majority (61%) published in the 2 years immediately following the release of the DSM-5. Fifty-five percent (n = 18) of studies were prospective (Barton et al. 2013; Beighley et al. 2014; Dawkins et al. 2016; Helles et al. 2015; Jashar et al. 2016; Konst et al. 2014; Magana and Vanegas 2017; Mazurek et al. 2017; Ocakoglu et al. 2015; Romero et al. 2016; Signorelli et al. 2015; Sumi et al. 2014; Sung et al. 2018; Tartaglia et al. 2017; van Steensel et al. 2015; Wheeler et al. 2015; Yaylaci and Miral 2017; Young and Rodi 2014), and the remaining were retrospective. While 16 studies were conducted in North America (15 in the US and one in Canada), the majority of studies were conducted globally, including seven in Asia (Kim et al. 2014; Ocakoglu et al. 2015; Ohashi et al. 2015; Sumi et al. 2014; Sung et al. 2018; Wong and Koh 2016; Yaylaci and Miral 2017), six in Europe (Helles et al. 2015; Romero et al. 2016; Signorelli et al. 2015; Solerdelcoll Arimany et al. 2017; van Steensel et al. 2015; Zander and Bolte 2015); and three in Australia (Christiansz et al. 2016; Hiller et al. 2014; Young and Rodi 2014); one was unreported (Mugzach et al. 2015).

Table 1

Characteristics of 33 included studies

Author, location, study type, data sources, and funding sources

Sample characteristics (i.e., number, age, gender, race, ethnicity)

Clinician type and diagnostic instruments

DSM-IV-TR diagnoses (including subtypes)

DSM-5 diagnoses

Reduction in diagnoses using DSM-5 criteria

Baio et al. (2018)

N = 4920

MD and PhD/PsyD

4658 ASDa

4236 ASD

9.1% ASD

US

Age 8 years

ASD case determination criteria for DSM-IV-TR and DSM-5

   

Retrospective

     

State vital records for 11 ADDM network sites

     

Federal funding

     

Barton et al. (2013)

N = 422

MD and PhD/PsyD

284 ASD

239 ASDb

15.8% ASD

US

Ages 16.79–39.36 months

ADI-R

   

Prospective, cross sectional

76.1% male

ADOS

   

Toddlers with siblings with ASD from pediatric offices in Atlanta, Connecticut, and bordering states; Connecticut early intervention program

69.9% white; 10.4% black

M-CHAT-R

   

Federal funding

8.3% Hispanic; 6.4% not reported; 3.1% Asian/Pacific Islander; 1.9% multi-racial

    

Beighley et al. (2014)

N = 261

Not reported

135 ASD

51 ASD

62.2% ASD

US

Ages 16–87 years

DSM-IV-TR/ICD-10 Checklist

   

Prospective, cross sectional

53.6% male

    

Louisiana state developmental center

76.6% white, 23% black, 0.4% Hispanic

    

No funding reported

     

Christiansz et al. (2016)

N = 185

PhD/PsyD

126 ASD

106 ASD

15.9% ASD

Australia

Ages 20–55 months and 1.7–4.6 years

ADI-R

103 AD

90 AD

12.6%

Retrospective

83.2% male

ADOS

23 PDD-NOS

16 PDD-NOS

30.4%

Early childhood services, pediatricians, and public regional autism assessment programme

     

Federal funding

     

Dawkins et al. (2016)

N = 183

PhD/PsyD

142 ASD

134 ASD

5.6% ASD

US

Ages 1–62 years

ADOS-2

   

Prospective, cross sectional

78.6% male

CARS2-HF

   

TEACCH Autism Program Centers in North Carolina

61.2% white, 20.9% black, 7.7% Hispanic

CARS2-ST

   

No funding reported

     

Foley-Nicpon et al. (2017)

N = 45

Not reported

45 ASD

45 ASD

0% ASD

US

Ages 5.5–17.8 years

ADI-R

16 AD

16 AD

0% AD

Retrospective

75.6% male

ADOS

17 Asperger’s

17 Asperger’s

0% Asperger’s

University center psychology clinic for gifted and talented students

  

12 PDD-NOS

12 PDD-NOS

0% PDD-NOS

Federal funding

     

Harstad et al. (2015)

N = 227

MD and PhD/PsyD

156 ASD

120 ASDb

23.1% ASD

US

Ages 1.3–18 years

ADOS-2

114 AD

103 AD

9.6% AD

Retrospective

83.7% male

Bayley Scales, 3rd edition

5 Asperger’s

3 Asperger’s

40.0% Asperger’s

Multidisciplinary developmental behavioral pediatric clinic

 

DAS-II

37 PDD-NOS

14 PDD-NOS

62.2% PDD-NOS

No funding reported

     

Helles et al. (2015)

N = 50

MD and PhD/PsyD

39 ASD

31 ASD

20.5% ASD

Sweden

Ages 23–43 years

ASDI

9 AD

9 AD

0% AD

Prospective

100% male

DISCO-II

22 Asperger’s

20 Asperger’s

9.1% Asperger’s

Gothenburg Child Neuropsychiatric Clinic

 

Gillberg’s Criteria for Asperger’s Syndrome

8 PDD-NOS

2 PDD-NOS

75.0% PDD-NOS

Non-federal funding

     

Hiller et al. (2014)

N = 114

Not reported

114 ASD

87 ASD

23.7% ASD

Australia

Mean ages: Male = 8.76 ± 3.91 years Female = 8.06 ± 4.03 years

ADI-R

   

Retrospective

55.3% male

ADOS

   

Private practice specializing in diagnostic assessments for PDDs

 

ADEC

   

Non-federal funding

 

CARS

   

Jashar et al. (2016)

N = 281

MD or PhD/PsyD

203 ASD

146 ASD

28.1% ASD

US

Ages 16–39 months

M-CHAT

134 AD

116 AD

13.4% AD

Prospective

77.9% male

VABS

69 PDD-NOS

30 PDD-NOS

56.5% PDD-NOS

Pediatrician offices

76.5% white, 6.4% black, 10.3% Hispanic

    

Federal funding

     

Kim et al. (2014)

N = 292

MD and PhD/PsyD

206 ASD

184 ASD

10.7% ASD

South Korea

Ages 7–12 years

ADI-R

114 AD

112 AD

1.8% AD

Retrospective file review

 

ADOS

34 Asperger’s

31 Asperger’s

8.8% Asperger’s

All children born from 1993 to 1999 in a suburb of Seoul, South Korea

 

BASC II-PRS

58 PDD-NOS

41 PDD-NOS

29.3% PDD-NOS

Federal and non-federal funding

     

Konst et al. (2014)

N = 1722

PhD/PsyD

1104 ASD

605 ASD

45.2% ASD

US

Ages 17–37 months

BISCUIT – Part I

   

Prospective, cross sectional

71.5% male

BISCUIT – Part II

   

State-funded early intervention program for children at-risk for developmental disability

49.2% white, 38.2% black

    

No funding reported

10.3% other, 2.3% Hispanic

    

Maenner et al. (2014)

N = 7597

Not reported

6577 ASD

5339 ASD

18.8% ASD

US

Age 8 years

Not reported

   

Retrospective

82.3% male

    

ADDM Network, 11 sites in 2006 and 14 sites in 2008

55.8% white, 22.5% black; 12.2% Hispanic, 2.9% Asian/Pacific Islander

    

Federal and non-federal funding

0.6% other

    

Magana and Vanegas (2017)

N = 29

Not reported

20 ASD

23 ASDc

0% ASD

US

Ages 4–16 years

ADI-R (Spanish)

   

Prospective

100% Hispanic

    

Clinics and parent support groups across two Midwestern cities

     

Federal and non-federal funding

     

Mazurek et al. (2017)

N = 439

MD and PhD/PsyD

278 ASD

249 ASDd 

10.4% ASD

US

Ages 2-17.7 years

ABC

229 AD

222 AD

3.0% AD

Prospective

79% male

ADOS-2

25 Asperger’s

20 Asperger's

20.0% Asperger’s

Six autism centers affiliated with the Autism Treatment Network

78% white

CBCL

24 PDD-NOS

6 PDD-NOS

75.0% PDD-NOS

Federal and non-federal funding

     

Mugzach et al. (2015)

N = 2642

Not reported

2642 ASD

2485 ASD

5.9% ASD

Country not reported

 

ADI-R

   

Retrospective

     

Data source not reported

     

Federal and non-federal funding

     

Ocakoglu et al. (2015)

N = 28

Not reported

28 ASD

18 ASD

37.5% ASD

Turkey

Ages birth to 6 years

ABC

28 PDD-NOS

18 PDD-NOS

37.5% PDD-NOS

Prospective

82.1% male

CARS

   

Children diagnosed with PDD-NOS by Ege University Disabled Health Committee in 2010–2011

     

No funding reported

     

Ohashi et al. (2015)

N = 68

MD

40 ASD

27 ASD

32.5% ASD

Japan

Ages 6.2–14.9 years

PARS

3 AD

3 AD

0% AD

Retrospective

63.2% male

 

16 Asperger’s

13 Asperger’s

18.8% Asperger’s

Department of Psychology and Development, Nagoya City University Hospital

  

21 PDD-NOS

11 PDD-NOS

47.6% PDD-NOS

No funding reported

     

Rieske et al. (2015)

N = 424

Not reported

300 ASD

192 ASDb

36.0% ASD

US

Ages 2–18 years

DSM-IV-TR/ICD-10 Checklists

   

Retrospective database review

72.9% male

ASD-CC

   

Outpatient clinics, schools, and community organizations

69.1% white, 17.4% unknown; 8.3% black

    

No funding reported

5.2% other

    

Romero et al. (2016)

N = 123

MD

123 ASD

57 ASD

53.6% ASD

Spain

Ages 5–15 years

DSM-IV-TR/ICD-10 Checklists

34 AD

17 AD

50.0% AD

Prospective

82% male

IDEA

27 Asperger’s

14 Asperger’s

48.1% Asperger’s

Schools in Magala, Spain

100% white

 

62 PDD-NOS

26 PDD-NOS

58.0% PDD-NOS

No funding reported

     

Signorelli et al. (2015)

N = 15

Not reported

15 ASD

3 ASD

80.0% ASD

Italy

Adults; Ages not provided

ADI-R

15 Asperger’s

3 Asperger’s

80.0% Asperger’s

Prospective

 

ADOS

   

Clinic-based sample

 

VABS

   

No funding reported

     

Solerdelcoll Arimany et al. (2017)

N = 118

Not reported

88 ASD

77 ASDb

12.5% ASD

Spain

Ages 3–17 years

ADI-R

18 AD

16 AD

11.1% AD

Retrospective chart review

86.5% male

 

47 Asperger’s

44 Asperger’s

6.4% Asperger’s

Department of Child and Adolescent Psychiatry and Psychology, Institute of Neurosciences, Hospital clinic, Barcelona, Spain

  

23 PDD-NOS

17 PDD-NOS

26.0% PDD-NOS

No funding reported

     

Sumi et al. (2014)

N = 180

MD and PhD/PsyD

64 ASD

62 ASD

3.1% ASD

Japan

Ages 2–5 years

PARS

8 AD

8 AD

0% AD

Prospective

75% male

 

27 Asperger’s

27 Asperger’s

0% Asperger’s

Nagoya West District Care Center for disabled children

96% Asian/Pacific Islander, 4% other

 

29 PDD-NOS

27 PDD-NOS

6.9% PDD-NOS

No funding reported

     

Sung et al. (2018)

N = 110

Not reported

92 ASD

77 ASD

16.3% ASD

Singapore

Ages 5.1–19.6 years

ADI-R

30 AD

30 AD

0% AD

Prospective

80.9% male

ADOS-2

32 Asperger’s

29 Asperger’s

3.3% Asperger’s

The Child Guidance Clinic under the Institute of Mental Health

81.8% Chinese, 6.4% Malay, 6.4% other, 5.4% Indian

 

30 PDD-NOS

18 PDD-NOS

40.0% PDD-NOS

No funding reported

     

Taheri et al. (2014)

N = 22

PhD/PsyD

22 ASD

12 ASDb

45.5% ASD

Canada

Ages 5–19 years

CARS

16 AD

11 AD

31.3% AD

Retrospective chart review

95.5% male

VABS-II

6 PDD-NOS

1 PDD-NOS

83.3% PDD-NOS

The TRE-ADD Program

     

No funding reported

     

Tartaglia et al. (2017)

N = 98

MD and PhD/PsyD

29 ASD

29 ASD

0% ASD

US

Ages 3–22 years

ADI-R

   

Prospective

100% male

ADOS

   

Hospital-based outpatient clinics and national SCA support organizations

88.8% white

    

Federal and non-federal funding

     

Turygin et al. (2013)

N = 142

PhD/PsyD

66 ASD

44 ASDb

33.3% ASD

US

Ages 2–16 years

BASC-2

   

Retrospective database review

62.7% male

    

University-affiliated clinic in Louisiana

76.1% white, 12% black

    

No funding reported

7% Hispanic

    

Van Steensel et al. (2015)

N = 90

Not reported

88 ASD

63 ASD

28.4% ASD

Netherlands

Ages 7–17 years

ADI-R

   

Prospective

76.7% male

CSBQ

   

Outpatient mental health centers

     

No funding reported

     

Wheeler et al. (2015)

N = 758

Not reported

276 ASD

191 ASD

30.8% ASD

US

Ages 2–67 years

Survey questions; no standardized instrument

   

Prospective (survey)

84.3% male

    

Registry of fragile X syndrome

89.8% white, 3.8% Hispanic, 2.6% black

    

Federal and non-federal funding

     

Wong and Koh (2016)

N = 206

MD and PhD/PsyD

202 ASD

184 ASD

8.9% ASD

Singapore

Mean age = 3 years, 10 months

ADOS

174 AD

165 AD

5.2% AD

Retrospective

85.9% male

ADOS-2

2 Asperger’s

1 Asperger’s

50.0% Asperger’s

Specialist multidisciplinary clinic for developmental concerns

67% Chinese, 18% Malay, 10% Indian, 5% other

 

4 PDD-NOS

1 PDD-NOS

75.0% PDD-NOS

No funding reported

  

22 Non-specified

17 Non-specified

22.7% Non-specified

Yaylaci and Miral (2017)

N = 150

MD

149 ASD

120 ASD

19.5% ASD

Turkey

Ages 3–15 years

ABC

139 AD

120 AD

13.6% AD

Prospective, cross sectional

77.3% male

CARS

4 Asperger’s

0 Asperger’s

100% Asperger’s

Data source not reported

  

6 PDD-NOS

0 PDD-NOS

100% PDD-NOS

No funding reported

     

Young and Rodi (2014)

N = 233

PhD/PsyD

210 ASD

120 ASDb

42.9% ASD

Australia

Ages 1–54 years

ADI-R

76 AD

56 AD

26.3% AD

Prospective

72.5% male

AQ

114 Asperger’s

64 Asperger’s

43.9% Asperger’s

Private practice offering services by psychologists and pathologists

 

CARS

20 PDD-NOS

0 PDD-NOS

100% PDD-NOS

No funding reported

 

CAST

   
  

SCQ

   

Zander and Bolte (2015)

N = 171

Not reported

127 ASD

115 ASD

9.4% ASD

Sweden

Ages 20–47 months

ABC

68 AD

66 AD

2.9% AD

Retrospective chart review

 

ADI-R

59 PDD-NOS

49 PDD-NOS

16.9% PDD-NOS

Subset from previous study sample from the Neuropsychiatric Resource Team Southeast, Division of Child and Adolescent Psychiatry, Stockholm County Council

 

ADOS-2

   

Federal and non-federal funding

 

VABS-II

   

aThe abbreviation of “ASD” under DSM-IV-TR refers to group of three diagnoses under the autism spectrum: Autistic Disorder (AD), Asperger’s Disorder, and Pervasive Developmental Disorder-Not Otherwise Specified (PDD-NOS), and “ASD” under DSM-5 refers to a diagnosis of Autism Spectrum Disorder

bStudy used draft instead of final published DSM-5 criteria to diagnose ASD

cThree individuals met DSM-5 but not DSM-IV-TR ASD criteria

dOne individual met DSM-5 but not DSM-IV-TR ASD criteria

ADDM Autism and Developmental Disabilities Monitoring; ADI-R Autism Diagnostic Interview—revised; ADOS Autism Diagnosis Observation Schedule; M-CHAT-R Modified Checklist for Autism in Toddlers—revised; ICD-10 international statistical classification of diseases and related health problems, 10th edition; ADOS-2 Autism Diagnosis Observation Schedule, 2nd edition; CARS2-HF Childhood Autism Rating Scale, 2nd edition (high-functioning clinical tool); CARS2-ST Childhood Autism Rating Scale, 2nd edition (standard clinical tool); TEACCH Treatment and Education of Autistic and Communication related handicapped Children; DAS-II Differential Ability Scale, 2nd edition; ASDI Autism Spectrum Disorder Interview; DISCO-II Diagnostic Interview for Social and Communication Disorders, 2nd edition; ADEC Autism Detection in Early Childhood; CARS Childhood Autism Rating Scale; M-CHAT Modified Checklist for Autism in Toddlers; VABS Vineland Adaptive Behavior Scale; BASC II-PRS Behavior Assessment System for Children II—Parent Report Scale; BISCUIT-Part I Baby and Infant Screen for Children with aUtIsm Traits-Part I; BISCUIT-Part II Baby and Infant Screen for Children with aUtIsm Traits-Part II; ABC Autism Behavior Checklist; CBCL Child Behavior Checklist; PARS Pervasive Developmental Disorder-Autism Society Japan Rating Scale; ASD-CC Autism Spectrum Disorders-Comorbidity for Children; IDEA Autism Spectrum Disorder Inventory; TRE-ADD Treatment, Research, and Education for Autism and Developmental Disorders; VABS-II Vineland Adaptive Behavior Scale, 2nd edition; SCA sex chromosome aneuploidy; BASC-2 Behavior Assessment System for Children, 2nd edition; CSBQ Children’s Social Behavior Questionnaire; AQ Autism Spectrum Quotient; CAST Childhood Asperger Syndrome Test; SCQ Social Communication Questionnaire

Demographics

Samples were heterogeneous in terms of size, age, and data sources. Sample sizes ranged from 15 (Signorelli et al. 2015) to 7597 (Maenner et al. 2014) individuals. The majority of studies (n = 24) restricted their samples to pediatric populations (i.e., ages ≤ 19 years). There were six studies limited to young children under the age of five (Barton et al. 2013; Christiansz et al. 2016; Jashar et al. 2016; Konst et al. 2014; Sumi et al. 2014; Zander and Bolte 2015); eight that included all children ages ≤ 19 years (Harstad et al. 2015; Mazurek et al. 2017; Ocakoglu et al. 2015; Rieske et al. 2015; Solerdelcoll Arimany et al. 2017; Turygin et al. 2013; Wong and Koh 2016; Yaylaci and Miral 2017); and 10 studies with older children ages 5–19 years (Baio et al. 2018; Foley-Nicpon et al. 2017; Hiller et al. 2014; Kim et al. 2014; Maenner et al. 2014; Magana and Vanegas 2017; Ohashi et al. 2015; Romero et al. 2016; Taheri et al. 2014; van Steensel et al. 2015). Two studies included samples of children and adults ages ≥ five years (Beighley et al. 2014; Sung et al. 2018); two restricted inclusion to adults ≥ 20 years (Helles et al. 2015; Signorelli et al. 2015); four included all ages (Dawkins et al. 2016; Tartaglia et al. 2017; Wheeler et al. 2015; Young and Rodi 2014); and one did not report ages (Mugzach et al. 2015). Twenty-eight studies provided data on gender, race, and/or ethnicity of their samples. In the 27 studies which reported gender, 79.6% of the cumulative sample population was male (11,367 of 14,276). For the 16 studies which reported figures on race and/or ethnicity, 61% of the cumulative sample population was white (7926 of 12,975). Nine studies specifically indicated their populations included individuals of Hispanic ethnicity; out of a total sample population of 11,395 individuals, only 1113 (9.8%) were Hispanic.

Data Sources and Funding Sources

Studies used a wide variety of data sources; for example, prospective studies included sources such as early intervention programs and centers; pediatric offices; developmental clinics; support groups; and organizational registries. For retrospective studies, data sources included state records (e.g., ADDM Network site records); hospital, university, and clinic records; private practices; public schools; community organizations; census records; and previous study samples. Fifteen studies reported receiving financial support from a variety of funding sources including federal (e.g., National Institutes of Health, Centers for Disease Control and Prevention) and non-federal (e.g., Autism Speaks, Simons Foundation Autism Research Initiative) entities (Baio et al. 2018; Barton et al. 2013; Christiansz et al. 2016; Foley-Nicpon et al. 2017; Helles et al. 2015; Hiller et al. 2014; Jashar et al. 2016; Kim et al. 2014; Maenner et al. 2014; Magana and Vanegas 2017; Mazurek et al. 2017; Mugzach et al. 2015; Tartaglia et al. 2017; Wheeler et al. 2015; Zander and Bolte 2015).

Diagnostic Instruments, Raters, and DSM-5 Criteria Version

The most common screening instruments used in combination with clinical impressions to diagnose ASD were the Autism Diagnostic Interview—Revised (ADI-R) and the Autism Diagnosis Observation Schedule (ADOS) with more than half of studies (55%) using either one of these or both. Other objective tools coupled with clinical impressions included a wide variety of checklists, scales, and diagnostic instruments focused on identifying and measuring autism characteristics, developmental delays, and social behavior deficiencies. Clinicians who interpreted findings of the instruments to make the diagnosis of ASD included physicians (e.g., child psychiatrists, behavioral pediatricians), psychologists (e.g., PhD and/or PsyD), and teams of physicians and psychologists. The majority of studies (78.8%) used the final published version of the DSM-5 (American Psychiatric Association 2013a) to diagnose ASD, and the 2011 draft version of the criteria (You et al. 2011) was used to diagnose ASD in the remaining studies (Barton et al. 2013; Harstad et al. 2015; Rieske et al. 2015; Solerdelcoll Arimany et al. 2017; Taheri et al. 2014; Turygin et al. 2013; Young and Rodi 2014).

Changes in ASD Diagnosis Rates since DSM-5 Publication

The percent reduction in DSM-IV-TR ASD diagnoses using DSM-5 criteria ranged from 0% (Foley-Nicpon et al. 2017; Magana and Vanegas 2017; Tartaglia et al. 2017) to 80% (Signorelli et al. 2015). Overall, 91% of studies reported ASD diagnosis reduction rates between 0 and 50% when applying DSM-5 criteria, with the majority of studies (60.6%) reporting reduction rates of 0–25% and 30.3% demonstrating reduction rates of 26–50%. Only three studies (9.1%) reported ASD diagnosis rates > 50% (Beighley et al. 2014; Romero et al. 2016; Signorelli et al. 2015); of note, the highest reduction rate of 80% was in a sample of 15 individuals, all of whom were adults (Signorelli et al. 2015).

DSM-IV-TR Subtypes most affected by DSM-5 ASD Criteria

Nineteen studies (57.5%) reported data on changes in ASD diagnosis under DSM-5 criteria according to one or more of the DSM-IV-TR ASD subtypes, and the reduction rates in ASD diagnosis varied widely by subtype. In the 17 studies that examined AD, reduction rates of ≤ 25% were demonstrated in the vast majority of studies (82.4%) with the remaining reporting reduction rates of 26–50%. For the 14 studies that looked at Asperger’s, the reduction rates were more equally spread with 57.1% of studies reporting reduction rates of ≤ 25 and 42.9% of studies reporting reduction rates ≥ 26%. Of note, Signorelli et al. (2015) reported a reduction rate in Asperger’s of 80% and Yaylaci and Miral (2017) reported a reduction rate of 100%. Highest overall reduction rates were seen for the PDD-NOS subtype. Only 16.7% of the eight studies which examined PDD-NOS saw ASD diagnosis reduction rates of ≤ 25%. The majority of studies (66.6%) reported PDD-NOS reduction rates in the 26–75% range with the remaining three studies (16.7%) finding reduction rates > 75%, two of which reported a 100% reduction rate (Yaylaci and Miral 2017; Young and Rodi 2014).

Impact of DSM-5 Social Communication Disorder (SCD) Diagnosis

Table 2 provides details on the five studies from the current review (Kim et al. 2014; Mazurek et al. 2017; Ocakoglu et al. 2015; Ohashi et al. 2015; Sumi et al. 2014) and four studies from the first review (Dickerson Mayes et al. 2013; Huerta et al. 2012; Taheri and Perry 2012; Wilson et al. 2013) that examined the proportion of individuals with DSM-IV-TR ASD who did not retain an ASD diagnosis under DSM-5 but alternatively met SCD criteria. Only three studies utilized US populations (Dickerson Mayes et al. 2013; Huerta et al. 2012; Mazurek et al. 2017). Five studies examined the impact of SCD on DSM-IV-TR ASD subtypes (Dickerson Mayes et al. 2013; Kim et al. 2014; Ocakoglu et al. 2015; Ohashi et al. 2015; Sumi et al. 2014). Individuals qualifying for an alternative SCD diagnosis included 2/2 (100%) for the AD subtype; 4/6 (66.7%) for the Asperger’s Disorder subtype; and 23/57 (40.4%) for the PDD-NOS subtype.

Table 2

Impact of Social Communication Disorder on Individuals who do not retain an ASD diagnosis under DSM-5

Study and country

DSM-IV-TR diagnoses (including subtypes)

DSM-5 diagnoses

SCD diagnoses N (% captured)

Dickerson Mayes (2013)a

25 ASDb

7 ASDc

5/18 (28%) ASD

US

25 PDD-NOS

7 PDD-NOS

5/18 (28%) PDD-NOS

Huerta et al. (2012)a

US and Canada

4,453 ASD

4,058 ASDc

75/395 (19%) ASD

Kim et al. (2014)

206 ASD

184 ASD

17/22 (77%) ASD

South Korea

114 AD

112 AD

2/2 (100%) AD

 

34 Asperger’s

31 Asperger’s

2/3 (67%) Asperger’s

 

58 PDD-NOS

41 PDD-NOS

13/17 (76%) PDD-NOS

Mazurek et al. (2017)

US

278 ASD

249 ASDd

2/30 (7%) ASD

Ocakoglu et al. (2015)

28 ASD

18 ASD

0/10 (0%) ASD;

Turkey

28 PDD-NOS

18 PDD-NOS

0/10 (0%) PDD-NOS

Ohashi et al. (2015)

40 ASD

27 ASD

5/13 (38%) ASD

Japan

3 AD

3 AD

AD = N/A

 

16 Asperger’s

13 Asperger’s

2/3 (67%) Asperger’s

 

21 PDD-NOS

11 PDD-NOS

3/10 (30%) PDD-NOS

Sumi et al. (2014)

64 ASD

62 ASD

2/2 (100%) ASD

Japan

8 AD

8 AD

AD = N/A

 

27 Asperger’s

27 Asperger’s

Asperger’s = N/A

 

29 PDD-NOS

27 PDD-NOS

2/2 (100%) PDD-NOS

Taheri and Perry (2012)a

129 ASD

82 ASDc

2/47 (4%) ASD

Canada

   

Wilson et al. (2013)a

80 ASD

61 ASDc

12/19 (63%) ASD

Europe

   

SCD social communication disorder; N/A not applicable

aStudy included in previous literature review

bThe abbreviation of “ASD” under DSM-IV-TR refers to group of three diagnoses under the autism spectrum: Autistic Disorder (AD), Asperger’s Disorder, and Pervasive Developmental Disorder-not otherwise specified (PDD-NOS), and “ASD” under DSM-5 refers to a diagnosis of Autism Spectrum Disorder

cStudy used draft instead of final published DSM-5 criteria to diagnose ASD

dOne participant met DSM-5 but not DSM-IV-TR ASD criteria

DSM-5 Sensitivity and Specificity

Seven studies reported the sensitivity and specificity of DSM-5 diagnostic criteria with ADI-R and/or ADOS. Of three studies that used both the ADI-R and ADOS (Barton et al. 2013; Christiansz et al. 2016; Sung et al. 2018), sensitivity and specificity values ranged from 0.84 to 0.93 and 0.54 to 0.83, respectively. For two studies that used the ADI-R alone (Magana and Vanegas 2017; Solerdelcoll Arimany et al. 2017), the sensitivity range was reported between 0.88 and 0.90 while the specificity range was between 0.57 and 0.86. The remaining two studies used the ADOS alone (Dawkins et al. 2016; Mazurek et al. 2017); the sensitivity range was 0.89 to 1.00 and the specificity range was 0.71 to 0.99.

Quantitative Synthesis

Results of the meta-analyses are provided in Figs. 3, 4, and 5. Data from 33 studies which examined changes in DSM-IV-TR ASD diagnosis when DSM-5 criteria were applied were pooled and represent data from 18,648 individuals. Using a random effects model, the pooled proportion suggests a 20.8% [95% confidence interval (CI) 16.0–26.7, p < 0.001] reduction in ASD diagnoses (Cochran’s Q = 1454.9, p < 0.001; I2 = 97.8) when DSM-5 criteria were applied (Fig. 3).

Fig. 3

Forest plots of the 33 studies included studies representing the proportion of individuals who met criteria for an Autism Spectrum Disorder (ASD) diagnosis under DSM-IV-TR but not for DSM-5 ASD. Squares represent effect sizes of individual studies with extended lines denoting 95% confidence intervals. Sizes of squares indicate the weight of each study based on sample size using random effects analysis. The diamond represents the estimated pooled effect size

Fig. 4

Forest plots of Autistic Disorder (top), Asperger’s Disorder (middle), and Pervasive Developmental Disorder-Not Otherwise Specified (PDD-NOS) (bottom) representing the proportion of individuals who met criteria for diagnosis under DSM-IV-TR criteria but not for DSM-5 Autism Spectrum Disorder. Squares represent effect sizes of individual studies with extended lines denoting 95% confidence intervals. Sizes of squares indicate the weight of each study based on sample size using random effects analysis. The diamond represents the estimated pooled effect size

Fig. 5

Forest plot of Social Communication Disorder (SCD) representing the proportion of individuals who met criteria for an Autism Spectrum Disorder (ASD) diagnosis under DSM-IV-TR criteria but not for DSM-5 and instead met the criteria for an alternative diagnosis of SCD. Squares represent effect sizes of individual studies with extended lines denoting 95% confidence intervals. Sizes of squares indicate the weight of each study based on sample size using random effects analysis. The diamond represents the pooled effect size

Figure 4 presents the pooled analyses that examined DSM-IV-TR diagnoses of AD, Asperger’s Disorder, and PDD-NOS when DSM-5 criteria were applied. Nineteen of 33 studies examined these subtypes: AD was examined in 17 studies with data representing 1285 individuals; Asperger’s Disorder was examined in 14 studies with data representing 387 individuals; and PDD-NOS was examined in 18 studies with data representing 519 individuals. Pooled effects suggest statistically significant reductions in ASD diagnoses of 10.1% (95% CI 6.2–16.0, p < 0.001) for those with AD (Cochran’s Q = 90.9, p < 0.001, I2 = 82.4) and 23.3% (95% CI 12.9–38.5, p = 0.001) for those with Asperger’s Disorder (Cochran’s Q = 65.4, p < 0.001, I2 = 80.1) when DSM-5 criteria were applied. The reduction in diagnoses for PDD-NOS was not statistically significant [46.1% (95% CI 34.6–58.0), p = 0.52] (Cochran’s Q = 80.3, p < 0.001; I2 = 78.8). For all models, heterogeneity was greater than expected by chance alone.

Figure 5 provides the pooled analysis that examined the number of individuals who met DSM-IV-TR ASD diagnosis but would not meet DSM-5 criteria and instead would qualify for an alternative diagnosis of SCD; these include data from nine studies representing 556 individuals. While the finding did not achieve statistical significance, the pooled effect suggests that less than one-third [28.8% (95% CI 13.9–50.5), p = 0.06] of those who met DSM-IV-TR ASD diagnostic criteria but not DSM-5 would meet SCD diagnostic criteria. Heterogeneity was greater than expected by chance alone (Cochran’s Q = 57.5, p < 0.001, I2 = 86.1). Although four of the studies that examined the impact of SCD used the draft version of DSM-5 ASD diagnostic criteria, there were no statistical differences between those and the five studies which used the final version of the criteria.

Subgroup Analyses

Table 3 presents results of subgroup analyses for ASD and the AD and Asperger’s subtypes. Of 10 variables explored, six were found to contribute to heterogeneity: age group (all models); continent where study was conducted (ASD); instruments administered to make the diagnosis (AD); clinician who made the diagnosis (all models); study funding sources (ASD and AD); and one risk of bias criterion – measures of intra and inter-rater agreement (ASD).

Table 3

Subgroup analyses

Variable

All studies

Autistic Disorder

Asperger’s Disorder

# Studies

Pooled result (%) (95% CI)

# Studies

Pooled result (%) (95% CI)

# Studies

Pooled result (%) (95% CI)

Study sample agea,b,c

      

 Young children < 5 years

6

17.2 (8.7, 31.3)

4

10.6 (6.3, 17.1)

1

1.8 (0.1, 23.0)

 Young (< 5 years) and older children (5–18 years)

8

20.3 (13.2, 30.0)

5

7.5 (4.1, 13.1)

5

29.2 (9.9, 61.0)

 Children (5–19 years)

10

21.7 (14.5, 31.1)

5

13.2 (2.7, 45.5)

5

15.4 (5.3, 37.1)

 Children and adults

2

36.4 (6.6, 82.3)

1

1.6 (0.1, 21.1)

--

--

 Adults only

2

49.2 (6.2, 93.4)

1

5.0 (0.3, 47.5)

2

39.2 (1.7, 96.0)

 All ages

4

18.7 (8.3, 36.9)

1

26.3 (17.7, 37.3)

1

43.9 (35.1, 53.1)

 Age not reported

1

5.9 (5.1, 6.9)

--

--

Continenta

      

 North America

16

21.6 (14.9, 30.2)

5

9.8 (4.5, 20.0)

3

19.2 (6.2, 46.1)

 Europe

6

29.2 (13.9, 51.2)

4

11.9 (1.9, 49.1)

4

29.0 (6.7, 70.0)

 Asia

7

15.9 (10.1, 24.2)

5

5.7 (2.3, 13.4)

6

16.9 (6.1, 39.0)

 Australia

3

26.4 (13.3, 45.7)

2

18.7 (8.6, 35.8)

1

43.9 (35.1, 53.1)

 Not reported

1

5.9 (5.1, 6.9)

--

--

--

--

Study design

      

 Prospective

18

25.5 (18.5, 33.9)

7

12.8 (6.0, 25.1)

8

31.2 (15.8, 52.2)

 Retrospective

15

16.8 (12.2, 22.7)

9

8.1 (4.6, 13.8)

6

13.6 (6.3, 27.1)

Instrumentsb

      

 ADI-R and ADOS

9

16.4 (10.8, 24.0)

5

3.9 (1.3, 11.5)

4

17.4 (2.6, 62.8)

 ADI-R

5

14.9 (4.3, 40.9)

2

21.1 (9.6, 40.1)

2

19.8 (2.2, 72.8)

 ADOS

4

11.0 (6.0, 19.4)

3

5.5 (2.8, 10.4)

3

25.8 (13.2, 44.1)

 Other

14

29.8 (18.6, 44.2)

6

19.6 (10.1, 34.8)

5

24.1 (6.9, 57.4)

 Not reported

1

18.8 (17.9, 19.8)

--

--

--

--

Clinician typea,b,c

      

 MD

3

34.0 (15.2, 59.8)

3

24.8 (6.4, 61.2)

3

44.6 (15.9, 77.4)

 PhD/PsyD

6

28.1 (17.6, 41.7)

3

21.3 (11.8, 35.3)

1

43.9 (35.1, 53.1)

 MD or PhD/PsyD

1

28.1 (22.3, 34.7)

1

13.4 (8.6, 20.3)

 Both MD and PhD/PsyD

9

11.9 (8.7, 16.1)

6

4.8 (2.8, 8.2)

6

15.0 (7.2, 28.9)

 Not reported

14

22.7 (15.1, 32.7)

4

4.6 (1.9, 10.7)

4

16.0 (2.1, 62.4)

DSM-5 criteria version

      

 Draft

7

28.0 (19.6, 38.4)

4

18.1 (9.3, 32.5)

3

24.9 (5.8, 63.8)

 Final

26

18.9 (13.8, 25.4)

13

7.8 (4.1, 14.1)

11

22.6 (10.4, 42.4)

Funding sourcea,b

      

 Both federal and non-federal

8

11.3 (6.4, 19.0)

3

2.7 (1.5, 4.9)

2

14.2 (6.2, 29.4)

 Federal only

5

14.3 (7.8, 24.8)

3

12.8 (9.1, 17.5)

1

2.8 (0.2, 32.2)

 Non-federal only

2

22.9 (16.9, 30.2)

1

5.0 (0.3, 47.5)

1

9.1 (2.3, 30.0)

 No funding reported

18

28.8 (21.9, 36.8)

10

15.1 (8.2, 26.1)

10

31.4 (16.5, 51.5)

 Risk of bias

      

Blinded to reference standard

      

 Low risk

3

11.1 (5.9, 19.7)

2

4.9 (2.6, 8.9)

2

18.3 (2.7, 64.9)

 Unclear risk

16

22.2 (13.8, 33.7)

7

12.0 (7.1, 19.5)

5

26.3 (10.4, 52.4)

 High risk

14

21.5 (15.2, 29.6)

8

9.1 (3.0, 24.6)

7

20.5 (6.5, 49.1)

Order of examination varied

      

 Low risk

2

22.9 (4.6, 64.8)

2

9.7 (1.0, 53.8)

2

32.7 (13.9, 59.4)

 Unclear risk

20

20.9 (15.0, 28.3)

8

10.6 (8.0, 14.1)

6

22.7 (8.8, 47.3)

 High risk

11

19.5 (10.6, 33.0)

7

8.6 (2.0, 29.7)

6

17.7 (4.5, 49.4)

Statistical measures of agreement a

      

 Low risk

8

17.1 (10.1, 27.4)

5

8.6 (2.2, 28.2)

4

18.9 (6.7, 42.8)

 Unclear risk

1

36.0 (30.8, 41.6)

 High risk

24

22.1 (16.7, 28.6)

12

10.8 (6.4, 17.6)

10

25.8 (10.5, 50.8)

MD physicians (e.g., child psychiatrists, behavioral pediatricians); PhD/PsyD psychologists; Both MD and PhD/PsyD teams of physicians and psychologists

aVariable contributing to heterogeneity (p < 0.05) in all studies

bVariable contributing to heterogeneity (p < 0.05) in Autistic disorder

cVariable contributing to heterogeneity (p < 0.05) in Asperger’s disorder

Publication Bias

Figure 6 displays the funnel plot representing differences in the proportion of those diagnosed with ASD using DSM-IV-TR versus DSM-5 criteria for all studies. The open circles indicate each of the 33 individual studies. The upper portion of the funnel plot displays symmetry. The three circles on the lower left side represent studies with small sample sizes and do not represent a major concern. Findings of the Classic fail-safe N test suggests that an additional 7765 studies would need to be added to significantly change the pooled effect. Funnel plots for the subtypes AD and Asperger’s Disorder are found online in Appendix 3; findings of the Classic fail-safe N test suggest that an additional 1455 and 135 studies, respectively, would need to be added to significantly change the pooled effect. The funnel plot for SCD is found online in Appendix 4; findings of the Classic fail-safe N test suggests that an additional 89 studies would need to be added to significantly change the pooled effect. The filled circle represents a study estimated to be missing from the analysis.

Fig. 6

Funnel plot represents differences in proportion of those diagnosed with ASD using DSM-5 versus DSM-IV-TR criteria. Plot shows the standard error of the difference in proportion (Y axis) versus the reported percent not captured by DSM-5 (X axis) using a random effects model. The vertical line indicates the pooled effect estimate. The open circles indicate each of the 33 individual studies included in the meta-analysis. The open diamond indicates the pooled effect size and 95% confidence interval for meta-analysis, and the filled diamond indicates pooled effect size and 95% confidence interval when missing studies suggested by publication bias analysis are included

Discussion

Current Study Findings

Despite advances in understanding pathophysiology in ASD, it remains a behaviorally defined clinical syndrome. As such, the diagnosis is often based on several variables including the parental historical presentation of concerns, demonstration of such behaviors during evaluations, clinical providers’ experience, rating instruments, and final determination based on clinically agreed upon diagnostic guidelines set forth by the DSM. Revisions in updated DSM classification may change an individual’s diagnosis. In reviewing studies published in the five years since publication of the DSM-5, which has more stringent criteria required for an ASD diagnosis, our study findings indicate that a significant number of individuals who qualified for a DSM-IV-TR ASD diagnosis would not meet DSM-5 criteria. With more than one-fifth of individuals with notable SCI difficulties coupled with disruptive RRBs who will no longer qualify for an ASD diagnosis, clinicians, researchers, and public health officials need to recognize that there are individuals lacking a diagnosis but remain in need of services. Early diagnosis and intensive treatment has been linked to improvement across many domains in autism (Reichow et al. 2018; Rogers 2016; Salomone et al. 2016; Schreibman et al. 2015); however, a recent study examining treatment patterns of ASD among children using nationally representative data found that nearly 30% of US children with ASD are not receiving behavioral or medication treatment (Xu et al. 2018). A variety of therapies provided by the board of education and insurance carriers are often limited based upon an ASD diagnosis and/or clearly defined developmental delays (Candon et al. 2018; Turcotte et al. 2016). Acknowledging their need for treatment, clinicians may be providing ASD diagnoses in addition to other comorbidities, which are common in children with ASD, notably attention-deficit hyperactivity disorder (ADHD), obsessive compulsive behaviors, mood disorders, sensory processing issues, or anxiety (Belardinelli et al. 2016; Ford 2014; Soke et al. 2018).

ADDM Network data also continue to demonstrate that ASD prevalence rates are rising even with tightened DSM-5 diagnostic criteria. If true positive diagnoses are actually increasing, parental awareness and acceptance, less stigmatization, better trained clinicians, more thorough data collection methods, and even increasing genetic tendencies could be contributing factors. In addition, comorbid diagnoses are now allowable for ASD under DSM-5, enabling clinicians to give multiple comorbid diagnoses of intellectual disability, ASD, and ADHD, which could also explain why ASD rates have continued to rise since publication of the DSM-5.

It is notable that findings from this current systematic literature review and meta-analysis indicate a smaller decrease in ASD diagnoses when comparing DSM-IV-TR to DSM-5 as compared to all earlier reviews. Additionally, all studies which examined DSM-IV-TR ASD subtypes were also found to have smaller decreases in ASD diagnosis when comparing DSM-IV-TR to DSM-5 as compared with the first review. This may be because clinicians now have a greater comfort level with interpreting DSM-5 criteria. It could also indicate that fewer individuals are failing to receive an ASD diagnosis than what previous studies anticipated. Nevertheless, these findings do show that approximately one in five individuals who would have received an ASD diagnosis under DSM-IV-TR would not receive a diagnosis under DSM-5 with only a minority being alternatively captured by SCD. Most recent ADDM Network data show a continued increase in prevalence of ASD; however, the majority of children included in the last data reported from surveillance year 2014 were diagnosed under DSM-IV-TR criteria (Baio et al. 2018). It will be important to examine the next release of ADDM Network data on autism rates, which is anticipated to be based solely on children diagnosed with DSM-5 criteria; considering the findings of our meta-analyses, we would predict there may be a decrease in autism rates reported. Regardless of whether ASD prevalence rates are on an upward or downward trend, the potential numbers of individuals who may have been previously eligible for a DSM-IV-TR diagnosis of ASD but would not qualify under DSM-5 as reported by this study remains alarming and points to a need for continued research on this topic.

Autism remains a behaviorally defined clinical disorder set forth by a multitude of clinicians experienced in caring for this population. These clinical criteria remain diagnostic despite the emergence of biomarkers in blood (Smith et al. 2018) and saliva (Hicks et al. 2018) samples, in addition to neuroimaging (Bi et al. 2018; Li et al. 2018; Shen et al. 2018; Zhao et al. 2018) and electrophysiological (Levin et al. 2017; Muhle et al. 2018; Righi et al. 2014) profiles. Moreover, the use of diagnostic tools to support or refute ASD diagnosis are often created and validated in homogeneous autism cohorts, such as male-dominant groups (Halladay et al. 2015). There is increasing awareness that females are likely being under- or misdiagnosed with ASD for numerous reasons, including ascertainment bias, differential presentation with more SCI deficits and less RRBs, and a role for a female protective effect which may alter the endophenotype (Goldman 2013; Jacquemont et al. 2014; Lai et al. 2015; Volkmar et al. 1993). Moreover, autism is being recognized and accepted in black, Hispanic, and other non-Caucasian individuals (Baio et al. 2018; Singh and Bunyak 2018).

Another question remains regarding who should assign the autism diagnosis. An individual may see a medical doctor, including a psychiatrist, developmental pediatrician, or neurologist, or they may see a psychologist. The use of different tools may aid in diagnosis. Interestingly, where both MDs and PhD/PsyDs were involved in the diagnosis there was the lowest decrease in ASD diagnosis rates between DSM-IV-TR and DSM-5. This would suggest a multidisciplinary evaluation may have more specificity in initial diagnosis than a single provider. An earlier diagnosis is crucial to identify the need for early intensive behavioral interventions which have been proven as the mainstay of ASD treatment (Dawson 2013; Orinstein et al. 2014; Weitlauf et al. 2014).

Findings of Other Prior Reviews and Meta-Analyses Versus Current Study Findings

The change in ASD diagnostic criteria with introduction of the DSM-5 has been of great interest to the public as well as clinicians and researchers. Three prior systematic literature reviews have studied the impact of the changes in DSM-5 ASD diagnosis criteria on autism rates (Kulage et al. 2014; Smith et al. 2015; Sturmey and Dalfern 2014). Table 4 summarizes the findings of these previous systematic reviews in comparison to the current study. All prior reviews were published within a period of less than two years after publication of the DSM-5 with 56% of the included studies being duplicative at the time of the third review (Smith et al. 2015). While general findings were consistent across studies, the estimated reduction in ASD rates under DSM-5 criteria varied widely across included studies, ranging from 7 to 62%. Only one previous study included a meta-analysis, reporting a pooled decrease of 31% in ASD across studies (Kulage et al. 2014). The current five-year follow-up study includes a large number of studies published since April 2013 with only nine being duplicative of articles included in previous reviews. Comparing current study findings for estimated ASD reduction to the first review, the pooled decrease is smaller (20.8% vs. 31%) but remains a concern.

Table 4

Summary of findings of systematic reviews examining the effects of DSM-5 criteria on the number of individuals diagnosed with ASD

Study

Study type

No. of articles

No. of duplicate articles included in prior reviews (%)

% Reduction in ASD diagnoses under DSM-5

No. of articles that included DSM-IV-TR ASD subtypes

% Reduction in ASD diagnoses by subtypes

Kulage et al. (2014)

Systematic Review with Meta-Analysis

14

N/A

31% (95% CI 20–44) pooled decrease across studies, p = 0.006

7

Pooled results: 22% AD, p < 0.001 70% Asperger’s, p = 0.38 70% PDD-NOS, p = 0.01

Sturmey and Dalfern (2014)

Systematic Review

12

9 (75%)

36.97% median overall reduction across studies; range = 7% − 54%

5

19.35% median reduction in more impaired group (i.e., AD); range = 0% − 26.3%

    

Pooled analysis not conducted

 

71.27% median reduction in less impaired group (i.e., Asperger’s and PDD-NOS); range = 16.6% − 100%

      

Pooled result not conducted

Smith et al. (2015)

Systematic Review

25

14 (56%)

Reduction ranged between 10% and 62% across studies; 8 studies (32%) reported ranges exceeding 40%

13

1%−31% AD

    

Pooled analysis not conducted

 

0%–44% Asperger’s

      

0%–50% PDD-NOS

      

Pooled analysis not conducted

Current follow-up study

Systematic review with meta-analysis

33

9 (27%)

20.8% (95% CI 16–27) pooled decrease across studies, p < 0.001

19

Pooled results: 10.1% AD, p < 0.001, 23.3% Asperger’s, p = 0.001, 46.1% PDD-NOS, p = 0.52

ASD autism spectrum disorder; CI confidence interval; AD autistic disorder; PDD-NOS pervasive developmental disorder-not otherwise specified

The number of studies included in the three previous systematic literature reviews which examined the impact of the DSM-5 diagnostic criteria on DSM-IV-TR ASD subtypes ranged from five to 13 studies. Across reviews, findings were consistent that the most affected subtype would be PDD-NOS, followed closely by Asperger’s Disorder, with AD being the least impacted. Comparing current study findings for estimated reductions in diagnoses by subtype with that of the first review, reductions are less for AD (10.1% vs. 22%) and Asperger’s (23.3% vs. 70%); while statistical significance was not achieved, the reduction for PDD-NOS was also less than previously reported (46.1% vs. 70%) (Kulage et al. 2014). Again, this trend may be reflected in the next release of ADDM Network data (Baio et al. 2018).

Social Communication Disorder

In the first review on this topic, 4 of 14 studies (29%) examined the impact of SCD and its potential to capture individuals with a DSM-IV-TR ASD diagnosis but who would not receive a DSM-5 ASD diagnosis (Kulage et al. 2014). Based on its intended purpose, it is surprising that five years later only five studies captured in the current review examined the potential impact of SCD; we expected to find a substantially higher number of studies exploring the impact of this new DSM-5 diagnosis. Importantly, when examining all nine studies that looked at SCD diagnoses, less than one-third (28.8%) of individuals who did not retain their ASD diagnosis under DSM-5 criteria would qualify for an SCD diagnosis. This is concerning and provides the only data combining results from multiple studies in the literature to date that SCD does not seem to be fulfilling its purpose as a “catch all” or alternative diagnosis for individuals who would have had an ASD diagnosis under DSM-IV-TR but not under DSM-5 criteria. Surprisingly, the PDD-NOS subtype—which was originally targeted by the SCD diagnosis – seems to be the subtype least likely to obtain an alternative SCD diagnosis (only 40% captured); however, across studies that examined DSM-IV-TR subtypes, the subtype sample sizes were small, limiting the scope of this finding. Discussion points in the studies which examined SCD emphasized two themes. Aligning with the results of this study, although SCD was originally described as an alternative diagnosis for individuals with symptoms of PDD-NOS but who would no longer have an autism diagnosis under DSM-5 criteria, it does not seem to be capturing a significant number of these individuals (Dickerson Mayes et al. 2013; Huerta et al. 2012; Mazurek et al. 2017; Ocakoglu et al. 2015; Wilson et al. 2013). Second, the few individuals who would receive SCD as an alternative diagnosis did not meet DSM-5 ASD criteria because of insufficient deficiencies in the RRB domain required for an ASD diagnosis (Huerta et al. 2012; Kim et al. 2014; Ohashi et al. 2015; Sumi et al. 2014; Taheri and Perry 2012).

Considering these findings, although limited, further research is clearly needed to evaluate the impact of SCD as a diagnosis and the degree to which it captures individuals who fail to meet DSM-5 ASD criteria, particularly across DSM-IV-TR subtypes and for individuals with significant impairment imposed by RRBs. Currently, the need for SCD to function as an alternative diagnosis for ASD is unclear; while some studies have indicated that an SCD diagnosis could serve as another means of obtaining required treatment and services (Greaves-Lord et al. 2013; Kim et al. 2014; Ohashi et al. 2015), others have questioned this possibility (Dickerson Mayes et al. 2013; Smith et al. 2015). The inherent overlap in diagnostic criteria for ASD and SCD poses challenges for its recognition and use as a distinct disorder from ASD (Visser and Tops 2017). It is essential to view SCD as an independent diagnosis and recognize where it overlaps with ASD before its usefulness can be ascertained and tailored treatments can be developed. Future studies which measure SCD prevalence beyond applying the diagnosis to individuals who do not meet DSM-5 ASD criteria are warranted (Swineford et al. 2014). Further complicating the applicability of the diagnosis, five years after DSM-5 publication research is still being conducted to design standardized screening and/or diagnostic instruments for SCD (Baird and Norbury 2016; Norbury 2014; Visser and Tops 2017; Yuan and Dollaghan 2018). Overall, these issues add to the “ongoing debate regarding the validity of SCD as a diagnostic entity” (Visser and Tops 2017). Indeed, examination of SCD as a diagnosis, relative to other developmental communication disorders, is in its infancy, leaving its impact unknown. Exploring whether SCD is a legitimate diagnosis independent of ASD, as well as its potential to serve as a gateway for eligibility for treatment and services, are important areas for future research.

Limitations

The findings of this systematic literature review and meta-analysis must be interpreted with some caution. Overall, risk of bias of the included studies was moderate with potential bias stemming from lack of blinding of raters to results of the references standard, DSM-IV-TR diagnosis, and failure to assess interrater agreement in classification of DSM-5 diagnoses. While we took measures to conduct a rigorous systematic review, it has some limitations. Heterogeneity greater than expected by chance alone was present in each meta-analytic model. Six variables were identified that explained some of the heterogeneity; however, it is likely that additional unidentified factors also contributed to heterogeneity both within and between studies but were not explored. Finally, importance of the findings on SCD, which are the products of two separate but related systematic reviews, is limited by the small sample sizes across studies.

Conclusions

The diagnosis of ASD and the potential impact of SCD for those who do not meet criteria for an ASD diagnosis using DSM-5 criteria is evolving. Findings of this systematic review and meta-analysis provide further insight regarding how DSM-5 is being used both nationally and internationally since the release of the new diagnostic criteria and point to areas of future research, particularly for SCD.

Notes

Acknowledgments

This study is a follow-up systematic literature review and meta-analysis to Kulage, K. M., Smaldone, A. M., & Cohn, E. G. (2014). How will DSM-5 affect autism diagnosis? A systematic literature review and meta-analysis. The Journal of Autism and Developmental Disorders, 44(8), 1918–1932,  https://doi.org/10.1007/s10803-014-2065-2.

Author Contributions

KMK conceived of the study, participated in the design and coordination of the study, and drafted and revised the manuscript. JG conducted the initial literature search and JU updated the literature search; both coordinated the study workflow in Covidence, drafted part of the methods section, and revised the manuscript. DR participated in the design and coordination of the study, drafted the introduction, and revised the manuscript. JMB provided clinical context in interpretation of study findings, drafted part of the results and discussion sections, and revised the manuscript. AMS conceived of the study, participated in the design and coordination of the study, performed analyses, and drafted and revised the manuscript. KMK, JG, JU, DR, and AMS participated in the literature screening, quality appraisal, and data extraction. All authors read and approved the final manuscript.

Funding

This study was not supported by any type of external or grant funding.

Compliance with Ethical Standards

Conflict of interest

All authors declare they have no conflicts of interest.

Ethical Approval

This manuscript does not contain any studies with human participants or animals performed by any of the authors.

Supplementary material

10803_2019_3967_MOESM1_ESM.docx (82 kb)
Supplementary material 1 (docx 82 KB)

References

  1. Adak, B., & Halder, S. (2017). Systematic review on prevalence for autism spectrum disorder with respect to gender and socio-economic status. Journal of Mental Disorders and Treatment, 3(1), 133.CrossRefGoogle Scholar
  2. American Psychiatric Association. (2000). Diagnostic and statistical manual of mental disordersfourth edition, text revison (DSM-IV-TR). Washington, DC: American Psychiatric Association.CrossRefGoogle Scholar
  3. American Psychiatric Association. (2013a). Diagnostic and statistical manual of mental disorders, fifth edition (DSM-5). Washington, DC: American Psychiatric Association.CrossRefGoogle Scholar
  4. American Psychiatric Association. (2013b). Social (pragmatic) communication disorder. Retrieved from https://www.psychiatry.org/File%20Library/Psychiatrists/Practice/DSM/APA_DSM-5-Social-Communication-Disorder.pdf.
  5. Baio, J., Wiggins, L., Christensen, D. L., Maenner, M. J., Daniels, J., Warren, Z., … Dowling, N. F. (2018). Prevalence of autism spectrum disorder among children aged 8 years—autism and developmental disabilities monitoring network, 11 Sites, United States, 2014. MMWR Surveillance Summary, 67(6), 1–23.  https://doi.org/10.15585/mmwr.ss6706a1.CrossRefGoogle Scholar
  6. Baird, G., & Norbury, C. F. (2016). Social (pragmatic) communication disorders and autism spectrum disorder. Archives of Disease in Childhood, 101(8), 745–751.  https://doi.org/10.1136/archdischild-2014-306944.CrossRefPubMedGoogle Scholar
  7. Barton, M. L., Robins, D. L., Jashar, D., Brennan, L., & Fein, D. (2013). Sensitivity and specificity of proposed DSM-5 criteria for autism spectrum disorder in toddlers. The Journal of Autism and Developmental Disorders, 43(5), 1184–1195.  https://doi.org/10.1007/s10803-013-1817-8.CrossRefPubMedGoogle Scholar
  8. Beighley, J. S., Matson, J. L., Rieske, R. D., Cervantes, P. E., Goldin, R., & Jang, J. (2014). Differences in stereotypic behavior in adults diagnosed with autism spectrum disorders using the DSM-IV-TR and the DSM-5. Journal of Developmental and Physical Disabilities, 26(2), 193–202.CrossRefGoogle Scholar
  9. Belardinelli, C., Raza, M., & Taneli, T. (2016). Comorbid behavioral problems and psychiatric disorders in autism spectrum disorders. Journal of Childhood & Developmental Disorders, 2(2), 1–9.CrossRefGoogle Scholar
  10. Bi, X. A., Liu, Y., Jiang, Q., Shu, Q., Sun, Q., & Dai, J. (2018). The diagnosis of autism spectrum disorder based on the random neural network cluster. Frontiers in Human Neuroscience, 12, 257.  https://doi.org/10.3389/fnhum.2018.00257.CrossRefPubMedPubMedCentralGoogle Scholar
  11. Candon, M. K., Barry, C. L., Epstein, A. J., Marcus, S. C., Kennedy-Hendricks, A., Xie, M., & Mandell, D. S. (2018). The differential effects of insurance mandates on health care spending for children’s autism spectrum disorder. Medical Care, 56(3), 228–232.  https://doi.org/10.1097/mlr.0000000000000863.CrossRefPubMedPubMedCentralGoogle Scholar
  12. Centers for Disease Control and Prevention. (2012). Prevalence of autism spectrum disorders–autism and developmental disabilities monitoring network, 14 sites, United States, 2008. MMWR Surveillance Summary, 61(3), 1–19.Google Scholar
  13. Christiansz, J. A., Gray, K. M., Taffe, J., & Tonge, B. J. (2016). Autism spectrum disorder in the DSM-5: Diagnostic sensitivity and specificity in early childhood. The Journal of Autism and Developmental Disorders, 46(6), 2054–2063.  https://doi.org/10.1007/s10803-016-2734-4.CrossRefPubMedGoogle Scholar
  14. Dawkins, T., Meyer, A. T., & Van Bourgondien, M. E. (2016). The relationship between the childhood autism rating scale: Second edition and clinical diagnosis utilizing the DSM-IV-TR and the DSM-5. The Journal of Autism and Developmental Disorders, 46(10), 3361–3368.  https://doi.org/10.1007/s10803-016-2860-z.CrossRefPubMedGoogle Scholar
  15. Dawson, G. (2013). Early intensive behavioral intervention appears beneficial for young children with autism spectrum disorders. The Journal of Pediatrics, 162(5), 1080–1081.  https://doi.org/10.1016/j.jpeds.2013.02.049.CrossRefPubMedGoogle Scholar
  16. Dickerson Mayes, S., Black, A., & Tierney, C. D. (2013). DSM-5 under-identified PDDNOS: Diagnostic agreement between the DSM-5, DSM-IV, and checklist for autism spectrum disorder. Research in Autism Spectrum Disorders, 7(2), 298–306.CrossRefGoogle Scholar
  17. Elsabbagh, M., Divan, G., Koh, Y. J., Kim, Y. S., Kauchali, S., Marcin, C., … Fombonne, E. (2012). Global prevalence of autism and other pervasive developmental disorders. Autism Research, 5(3), 160–179.  https://doi.org/10.1002/aur.239.
  18. Foley-Nicpon, M., Fosenburg, S. L., Wurster, K. G., & Assouline, S. G. (2017). Identifying high ability children with DSM-5 autism spectrum or social communication disorder: Performance on autism diagnostic instruments. The Journal of Autism and Developmental Disorders, 47(2), 460–471.  https://doi.org/10.1007/s10803-016-2973-4.CrossRefPubMedGoogle Scholar
  19. Fombonne, E., Quirke, S., & Hagen, A. (2009). Prevalence and interpretation of recent trends in rates of pervasive developmental disorders. McGill Journal of Medicine, 12(2), 73.PubMedPubMedCentralGoogle Scholar
  20. Ford, L. (2014). Co-occurrence of psychiatric disorders in individuals with autism spectrum disorders. In S. X. Ming & B. A. Pletcher (Eds.), Navigating the medical maze with a child with autism spectrum disorder (pp. 36–49). London: Jessica Kingsley Publishers.Google Scholar
  21. Goldman, S. (2013). Opinion: Sex, gender and the diagnosis of autism—a biosocial view of the male preponderance. Research in Autism Spectrum Disorders, 7(6), 675–679.  https://doi.org/10.1016/j.rasd.2013.02.006.CrossRefPubMedPubMedCentralGoogle Scholar
  22. Greaves-Lord, K., Eussen, M. L., Verhulst, F. C., Minderaa, R. B., Mandy, W., Hudziak, J. J., … Hartman, C. A. (2013). Empirically based phenotypic profiles of children with pervasive developmental disorders: Interpretation in the light of the DSM-5. The Journal of Autism and Developmental Disorders, 43(8), 1784–1797.  https://doi.org/10.1007/s10803-012-1724-4.
  23. Halladay, A. K., Bishop, S., Constantino, J. N., Daniels, A. M., Koenig, K., Palmer, K., … Szatmari, P. (2015). Sex and gender differences in autism spectrum disorder: Summarizing evidence gaps and identifying emerging areas of priority. Molecular Autism, 6, 36.  https://doi.org/10.1186/s13229-015-0019-y.
  24. Harstad, E. B., Fogler, J., Sideridis, G., Weas, S., Mauras, C., & Barbaresi, W. J. (2015). Comparing diagnostic outcomes of autism spectrum disorder using DSM-IV-TR and DSM-5 criteria. The Journal of Autism and Developmental Disorders, 45(5), 1437–1450.  https://doi.org/10.1007/s10803-014-2306-4.CrossRefPubMedGoogle Scholar
  25. Helles, A., Gillberg, C. I., Gillberg, C., & Billstedt, E. (2015). Asperger syndrome in males over two decades: Stability and predictors of diagnosis. Journal of Child Psychology and Psychiatry, 56(6), 711–718.  https://doi.org/10.1111/jcpp.12334.CrossRefPubMedGoogle Scholar
  26. Hicks, S. D., Rajan, A. T., Wagner, K. E., Barns, S., Carpenter, R. L., & Middleton, F. A. (2018). Validation of a salivary RNA test for childhood autism spectrum disorder. Frontiers in Genetics, 9, 534.  https://doi.org/10.3389/fgene.2018.00534.CrossRefPubMedPubMedCentralGoogle Scholar
  27. Higgins, J. P., Thompson, S. G., Deeks, J. J., & Altman, D. G. (2003). Measuring inconsistency in meta-analyses. BMJ, 327(7414), 557–560.  https://doi.org/10.1136/bmj.327.7414.557.CrossRefPubMedPubMedCentralGoogle Scholar
  28. Hiller, R. M., Young, R. L., & Weber, N. (2014). Sex differences in autism spectrum disorder based on DSM-5 criteria: evidence from clinician and teacher reporting. Journal of Abnormal Child Psychology, 42(8), 1381–1393.  https://doi.org/10.1007/s10802-014-9881-x.CrossRefPubMedGoogle Scholar
  29. Huerta, M., Bishop, S. L., Duncan, A., Hus, V., & Lord, C. (2012). Application of DSM-5 criteria for autism spectrum disorder to three samples of children with DSM-IV diagnoses of pervasive developmental disorders. The American Journal of Psychiatry, 169(10), 1056–1064.  https://doi.org/10.1176/appi.ajp.2012.12020276.CrossRefPubMedPubMedCentralGoogle Scholar
  30. Jacquemont, S., Coe, B. P., Hersch, M., Duyzend, M. H., Krumm, N., Bergmann, S., … Eichler, E. E. (2014). A higher mutational burden in females supports a “female protective model” in neurodevelopmental disorders. American Journal of Human Genetics, 94(3), 415–425.  https://doi.org/10.1016/j.ajhg.2014.02.001.
  31. Jashar, D. T., Brennan, L. A., Barton, M. L., & Fein, D. (2016). Cognitive and adaptive skills in toddlers who meet criteria for autism in DSM-IV but not DSM-5. The Journal of Autism and Developmental Disorders, 46(12), 3667–3677.  https://doi.org/10.1007/s10803-016-2901-7.CrossRefPubMedGoogle Scholar
  32. Kim, Y. S., Fombonne, E., Koh, Y. J., Kim, S. J., Cheon, K. A., & Leventhal, B. L. (2014). A comparison of DSM-IV pervasive developmental disorder and DSM-5 autism spectrum disorder prevalence in an epidemiologic sample. American Academy of Child & Adolescent Psychiatry, 53(5), 500–508.  https://doi.org/10.1016/j.jaac.2013.12.021.CrossRefGoogle Scholar
  33. Kogan, M. D., Vladutiu, C. J., Schieve, L. A., Ghandour, R. M., Blumberg, S. J., Zablotsky, B., … Lu, M. C. (2018). The prevalence of parent-reported autism spectrum disorder among US children. Pediatrics.  https://doi.org/10.1542/peds.2017-4161.
  34. Konst, M. J., Matson, J. L., Goldin, R., & Rieske, R. (2014). How does ASD symptomology correlate with ADHD presentations? Research in Developmental Disabilities, 35(9), 2252–2259.  https://doi.org/10.1016/j.ridd.2014.05.017.CrossRefPubMedGoogle Scholar
  35. Kulage, K. M., Smaldone, A. M., & Cohn, E. G. (2014). How will DSM-5 affect autism diagnosis? A systematic literature review and meta-analysis. The Journal of Autism and Developmental Disorders, 44(8), 1918–1932.  https://doi.org/10.1007/s10803-014-2065-2.CrossRefPubMedGoogle Scholar
  36. Lai, M. C., Lombardo, M. V., Auyeung, B., Chakrabarti, B., & Baron-Cohen, S. (2015). Sex/gender differences and autism: setting the scene for future research. Journal of the American Academy of Child and Adolescent Psychiatry, 54(1), 11–24.  https://doi.org/10.1016/j.jaac.2014.10.003.CrossRefPubMedPubMedCentralGoogle Scholar
  37. Levin, A. R., Varcin, K. J., O’Leary, H. M., Tager-Flusberg, H., & Nelson, C. A. (2017). EEG power at 3 months in infants at high familial risk for autism. Journal of Neurodevelopmental Disorders, 9(1), 34.  https://doi.org/10.1186/s11689-017-9214-9.CrossRefPubMedPubMedCentralGoogle Scholar
  38. Levy, S. E., Mandell, D. S., & Schultz, R. T. (2009). Autism. Lancet, 374(9701), 1627–1638.  https://doi.org/10.1016/s0140-6736(09)61376-3.CrossRefPubMedPubMedCentralGoogle Scholar
  39. Li, G., Liu, M., Sun, Q., Shen, D., & Wang, L. (2018). Early diagnosis of autism disease by multi-channel CNNs. Machine Learning in Medical Imaging, 11046, 303–309.  https://doi.org/10.1007/978-3-030-00919-9_35.CrossRefPubMedGoogle Scholar
  40. Lucas, N., Macaskill, P., Irwig, L., & Bogduk, N. (2010). The development of a quality appraisal tool for studies of diagnostic reliability (QAREL). Journal of Clinical Epidemiology, 63(8), 854–861.  https://doi.org/10.1016/j.jclinepi.2009.10.002.CrossRefPubMedGoogle Scholar
  41. Lucas, N., Macaskill, P., Irwig, L., Moran, R., Rickards, L., Turner, R., & Bogduk, N. (2013). The reliability of a quality appraisal tool for studies of diagnostic reliability (QAREL). BMC Medical Research Methodology, 13, 111.  https://doi.org/10.1186/1471-2288-13-111.CrossRefPubMedPubMedCentralGoogle Scholar
  42. Maenner, M. J., Rice, C. E., Arneson, C. L., Cunniff, C., Schieve, L. A., Carpenter, L. A., … Durkin, M. S. (2014). Potential impact of DSM-5 criteria on autism spectrum disorder prevalence estimates. JAMA Psychiatry, 71(3), 292–300.  https://doi.org/10.1001/jamapsychiatry.2013.3893.
  43. Magana, S., & Vanegas, S. B. (2017). Diagnostic utility of the ADI-R and DSM-5 in the assessment of latino children and adolescents. The Journal of Autism and Developmental Disorders, 47(5), 1278–1287.  https://doi.org/10.1007/s10803-017-3043-2.CrossRefPubMedGoogle Scholar
  44. Matson, J. L., Kozlowski, A. M., Hattier, M. A., Horovitz, M., & Sipes, M. (2012). DSM-IV vs DSM-5 diagnostic criteria for toddlers with autsim. Developmental Neurorehabilitation, 15(3), 185–190.  https://doi.org/10.3109/17518423.2012.672341.CrossRefPubMedGoogle Scholar
  45. Mattila, M. L., Kielinen, M., Linna, S. L., Jussila, K., Ebeling, H., Bloigu, R., … Moilanen, I. (2011). Autism spectrum disorders according to DSM-IV-TR and comparison with DSM-5 draft criteria: an epidemiological study. Journal of the American Academy of Child & Adolescent Psychiatry, 50(6), 583–592.e511.  https://doi.org/10.1016/j.jaac.2011.04.001.
  46. Mazurek, M., Lu, F., Symecko, H., Butter, E., Bing, N., Hundley, R., … Handen, B. (2017). A prospective study of the concordance of DSM-IV and DSM-5 diagnostic criteria for autism spectrum disorder. The Journal of Autism & Developmental Disorders, 47(9), 2783–2794.  https://doi.org/10.1007/s10803-017-3200-7.
  47. Moher, D., Liberati, A., Tetzlaff, J., & Altman, D. G. (2009). Preferred reporting items for systematic reviews and meta-analyses: the PRISMA statement. Journal of Clinical Epidemiology, 62(10), 1006–1012.  https://doi.org/10.1016/j.jclinepi.2009.06.005.CrossRefPubMedGoogle Scholar
  48. Mugzach, O., Peleg, M., Bagley, S. C., Guter, S. J., Cook, E. H., & Altman, R. B. (2015). An ontology for autism spectrum disorder (ASD) to infer ASD phenotypes from autism diagnostic interview-revised data. Journal of Biomedical Informatics, 56, 333–347.  https://doi.org/10.1016/j.jbi.2015.06.026.CrossRefPubMedPubMedCentralGoogle Scholar
  49. Muhle, R. A., Reed, H. E., Stratigos, K. A., & Veenstra-VanderWeele, J. (2018). The emerging clinical neuroscience of autism spectrum disorder: A review. JAMA Psychiatry, 75(5), 514–523.  https://doi.org/10.1001/jamapsychiatry.2017.4685.CrossRefPubMedGoogle Scholar
  50. Norbury, C. F. (2014). Practitioner review: Social (pragmatic) communication disorder conceptualization, evidence and clinical implications. Journal of Child Psychology and Psychiatry, 55(3), 204–216.  https://doi.org/10.1111/jcpp.12154.CrossRefPubMedGoogle Scholar
  51. Ocakoglu, F. T., Kayan, B., Kose, S., Ozbaran, N. B., Erermis, H. S., Bildik, T., & Aydin, C. (2015). Reassessment of pervasive development disorder-not otherwise specified outcome according to DSM-5 diagnostical criteria. Klinik Psikofarmakologi Bulteni, 25, S64–S65.Google Scholar
  52. Ohashi, K., Mizuno, Y., Miyachi, T., Asai, T., Imaeda, M., & Saitoh, S. (2015). Concordance of DSM-5 and DSM-IV-TR classifications for autism spectrum disorder. Pediatrics International, 57(6), 1097–1100.  https://doi.org/10.1111/ped.12704.CrossRefPubMedGoogle Scholar
  53. Orinstein, A. J., Helt, M., Troyb, E., Tyson, K. E., Barton, M. L., Eigsti, I. M., … Fein, D. A. (2014). Intervention for optimal outcome in children and adolescents with a history of autism. Journal of Developmental & Behavioral Pediatrics, 35(4), 247–256.  https://doi.org/10.1097/dbp.0000000000000037.
  54. Persaud, R. (1996). Misleading meta-analysis. “Fail safe N” is a useful mathematical measure of the stability of results. BMJ, 312(7023), 125.CrossRefGoogle Scholar
  55. Reichow, B., Hume, K., Barton, E. E., & Boyd, B. A. (2018). Early intensive behavioral intervention (EIBI) for young children with autism spectrum disorders (ASD). Cochrane Database of Systematic Reviews, 5, Cd009260.  https://doi.org/10.1002/14651858.CD009260.pub3.CrossRefPubMedGoogle Scholar
  56. Rice, C., & Autism and Developmental Disabilities Monitoring Network Surveillance Year 2000 Principal Investigators. (2007). Prevalence of autism spectrum disorders–autism and developmental disabilities monitoring network, six sites, United States, 2000. MMWR Surveillance Summary, 56(SS01), 1–11.Google Scholar
  57. Rieske, R. D., Matson, J. L., Beighley, J. S., Cervantes, P. E., Goldin, R. L., & Jang, J. (2015). Comorbid psychopathology rates in children diagnosed with autism spectrum disorders according to the DSM-IV-TR and the proposed DSM-5. Developmental Neurorehabilitation, 18(4), 218–223.  https://doi.org/10.3109/17518423.2013.790519.CrossRefPubMedGoogle Scholar
  58. Righi, G., Tierney, A. L., Tager-Flusberg, H., & Nelson, C. A. (2014). Functional connectivity in the first year of life in infants at risk for autism spectrum disorder: an EEG study. PLoS ONE, 9(8), e105176.  https://doi.org/10.1371/journal.pone.0105176.CrossRefPubMedPubMedCentralGoogle Scholar
  59. Rogers, S. (2016). Early start denver model. In R. G. Romanczyk & J. McEachin (Eds.), Comprehensive models of autism spectrum disorder treatment: Points of divergence and convergence (pp. 45–62). New York: Springer Publishing Company.CrossRefGoogle Scholar
  60. Romero, M., Aguilar, J. M., Del-Rey-Mejias, A., Mayoral, F., Rapado, M., Pecina, M., … Lara, J. P. (2016). Psychiatric comorbidities in autism spectrum disorder: A comparative study between DSM-IV-TR and DSM-5 diagnosis. International Journal of Clinical and Health Psychology, 16(3), 266–275.Google Scholar
  61. Salomone, E., Beranova, S., Bonnet-Brilhault, F., Briciet Lauritsen, M., Budisteanu, M., Buitelaar, J., … Charman, T. (2016). Use of early intervention for young children with autism spectrum disorder across Europe. Autism, 20(2), 233–249.  https://doi.org/10.1177/1362361315577218.
  62. Schieve, L. A., Rice, C., Boyle, C., Visser, S. N., & Blumberg, S. J. (2006). Mental health in the United States: Parental report of diagnosed autism in children aged 4–17 years–United States, 2003–2004. Morbidity and Mortality Weekly Report, 55(17), 481–486.Google Scholar
  63. Schreibman, L., Dawson, G., Stahmer, A. C., Landa, R., Rogers, S. J., McGee, G. G., … Halladay, A. (2015). Naturalistic developmental behavioral interventions: Empirically validated treatments for autism spectrum disorder. The Journal of Autism and Developmental Disorders, 45(8), 2411–2428.  https://doi.org/10.1007/s10803-015-2407-8.CrossRefPubMedGoogle Scholar
  64. Shen, M. D., Nordahl, C. W., Li, D. D., Lee, A., Angkustsiri, K., Emerson, R. W., … Amaral, D. G. (2018). Extra-axial cerebrospinal fluid in high-risk and normal-risk children with autism aged 2–4 years: a case-control study. Lancet Psychiatry, 5(11), 895–904.  https://doi.org/10.1016/s2215-0366(18)30294-3.CrossRefPubMedGoogle Scholar
  65. Signorelli, M. S., Petrosino, B., Magnano di San Lio, F., & Aguglia, E. (2015). Asperger syndrome in adults and DSM-5: What does the future hold? European Neuropsychopharmacology, 25(Suppl 2), S650.CrossRefGoogle Scholar
  66. Singh, J. S., & Bunyak, G. (2018). Autism disparities: A systematic review and meta-ethnography of qualitative research. Qualitative Health Research.  https://doi.org/10.1177/1049732318808245.CrossRefPubMedGoogle Scholar
  67. Smith, A. M., King, J. J., West, P. R., Ludwig, M. A., Donley, E. L. R., Burrier, R. E., & Amaral, D. G. (2018). Amino acid dysregulation metabotypes: Potential biomarkers for diagnosis and individualized treatment for subtypes of autism spectrum disorder. Biological Psychiatry.  https://doi.org/10.1016/j.biopsych.2018.08.016.CrossRefPubMedGoogle Scholar
  68. Smith, I. C., Reichow, B., & Volkmar, F. R. (2015). The effects of DSM-5 criteria on number of individuals diagnosed with autism spectrum disorder: A systematic review. The Journal of Autism and Developmental Disorders, 45(8), 2541–2552.  https://doi.org/10.1007/s10803-015-2423-8.CrossRefPubMedGoogle Scholar
  69. Soke, G. N., Maenner, M. J., Christensen, D., Kurzius-Spencer, M., & Schieve, L. A. (2018). Prevalence of co-occurring medical and behavioral conditions/symptoms among 4- and 8-year-old children with autism spectrum disorder in selected areas of the United States in 2010. The Journal of Autism and Developmental Disorders, 48(8), 2663–2676.  https://doi.org/10.1007/s10803-018-3521-1.CrossRefPubMedGoogle Scholar
  70. Solerdelcoll Arimany, M., Blazquez, A., Calvo, R., Varela, E., Puig, O., & Lazaro, L. (2017). Diagnostic stability of autism spectrum disorders with the DSM-5 diagnostic criteria. European Neuropsychopharmacology, 27, S1104.CrossRefGoogle Scholar
  71. Sturmey, P., & Dalfern, S. (2014). The effect of DSM5 autism diagnostic criteria on number of individuals diagnosed with autism spectrum disorders: A systematic review. Review Journal of Autism and Developmental Disorders, 1, 249–252.CrossRefGoogle Scholar
  72. Sumi, S., Miyachi, T., Ohya, K., Ohashi, K., & Saitoh, S. (2014). Application of the final DSM-5 criteria for young children with autism spectrum disorder. Autism-Open Access, 4(3), 135.Google Scholar
  73. Sung, M., Goh, T. J., Tan, B. L. J., Chan, J. S., & Liew, H. S. A. (2018). Comparison of DSM-IV-TR and DSM-5 criteria in diagnosing autism spectrum disorders in Singapore. The Journal of Autism and Developmental Disorders, 48(10), 3273–3281.  https://doi.org/10.1007/s10803-018-3594-x.CrossRefPubMedGoogle Scholar
  74. Swineford, L. B., Thurm, A., Baird, G., Wetherby, A. M., & Swedo, S. (2014). Social (pragmatic) communication disorder: A research review of this new DSM-5 diagnostic category. Journal of Neurodevelopmental Disorders. 6(1), 41.CrossRefGoogle Scholar
  75. Taheri, A., & Perry, A. (2012). Exploring the proposed DSM-5 criteria in a clinical sample. The Journal of Autism and Developmental Disorders, 42(9), 1810–1817.  https://doi.org/10.1007/s10803-012-1599-4.CrossRefPubMedGoogle Scholar
  76. Taheri, A., Perry, A., & Factor, D. C. (2014). A further examination of the DSM-5 autism spectrum disorder criteria in practice. Journal on Developmental Disabilities, 20(1), 116–121.Google Scholar
  77. Tartaglia, N. R., Wilson, R., Miller, J. S., Rafalko, J., Cordeiro, L., Davis, S., … Ross, J. (2017). Autism spectrum disorder in males with sex chromosome aneuploidy: XXY/Klinefelter syndrome, XYY, and XXYY. Journal of Developmental & Behavioral Pediatrics, 38(3), 197–207.  https://doi.org/10.1097/dbp.0000000000000429.
  78. Treffert, D. A. (1970). Epidemiology of infantile autism. Archives of General Psychiatry, 22(5), 431–438.CrossRefGoogle Scholar
  79. Turcotte, P., Mathew, M., Shea, L. L., Brusilovskiy, E., & Nonnemacher, S. L. (2016). Service needs across the lifespan for individuals with autism. The Journal of Autism and Developmental Disorders, 46(7), 2480–2489.  https://doi.org/10.1007/s10803-016-2787-4.CrossRefPubMedGoogle Scholar
  80. Turygin, N. C., Matson, J. L., Adams, H., & Belva, B. (2013). The effect of DSM-5 criteria on externalizing, internalizing, behavioral and adaptive symptoms in children diagnosed with autism. Developmental Neurorehabilitation, 16(4), 277–282.  https://doi.org/10.3109/17518423.2013.769281.CrossRefPubMedGoogle Scholar
  81. van Steensel, F. J., Bogels, S. M., & de Bruin, E. I. (2015). DSM-IV versus DSM-5 autism spectrum disorder and social anxiety disorder in childhood: Similarities and differences. Journal of Child and Family Studies, 24(9), 2752–2756.CrossRefGoogle Scholar
  82. Visser, S., & Tops, W. (2017). Social pragmatic communication disorder and autism spectrum disorder: Two of a kind? A narrative review. Annals of Psychiatry and Mental Health. 5(6).Google Scholar
  83. Volkmar, F. R., Szatmari, P., & Sparrow, S. S. (1993). Sex differences in pervasive developmental disorders. The Journal of Autism and Developmental Disorders, 23(4), 579–591.CrossRefGoogle Scholar
  84. Weitlauf, A. S., McPheeters, M. L., Peters, B., Sathe, N., Travis, R., Aiello, R., … Warren, Z. (2014). AHRQ comparative effectiveness reviews. In Therapies for children with autism spectrum disorder: Behavioral interventions update. Rockville, MD: Agency for Healthcare Research and Quality (US).Google Scholar
  85. Wheeler, A. C., Mussey, J., Villagomez, A., Bishop, E., Raspa, M., Edwards, A., … Bailey, D. B. Jr. (2015). DSM-5 changes and the prevalence of parent-reported autism spectrum symptoms in Fragile X syndrome. The Journal of Autism and Developmental Disorders, 45(3), 816–829.  https://doi.org/10.1007/s10803-014-2246-z.
  86. Wilson, C. E., Gillan, N., Spain, D., Robertson, D., Roberts, G., Murphy, C. M., … Murphy, D. G. (2013). Comparison of ICD-10R, DSM-IV-TR and DSM-5 in an adult autism spectrum disorder diagnostic clinic. The Journal of Autism and Developmental Disorders, 43(11), 2515–2525.  https://doi.org/10.1007/s10803-013-1799-6.
  87. Wing, L., & Gould, J. (1979). Severe impairments of social interaction and associated abnormalities in children: Epidemiology and classification. The Journal of Autism and Developmental Disorders, 9(1), 11–29.CrossRefGoogle Scholar
  88. Wong, C. M., & Koh, H. C. (2016). Brief report: Investigating the implications of applying the new DSM-5 criteria for diagnosing autism spectrum disorder in a preschool population in Singapore. The Journal of Autism and Developmental Disorders, 46(9), 3177–3182.  https://doi.org/10.1007/s10803-016-2841-2.CrossRefPubMedGoogle Scholar
  89. Xu, G., Strathearn, L., Liu, B., et al. (2018). Prevalence and treatment patterns of autism spectrum disorder in the United States, 2016. JAMA Pediatrics.  https://doi.org/10.1001/jamapediatrics.2018.4208.CrossRefPubMedPubMedCentralGoogle Scholar
  90. Yaylaci, F., & Miral, S. (2017). A comparison of DSM-IV-TR and DSM-5 diagnostic classifications in the clinical diagnosis of autistic spectrum disorder. The Journal of Autism and Developmental Disorders, 47(1), 101–109.  https://doi.org/10.1007/s10803-016-2937-8.CrossRefPubMedGoogle Scholar
  91. You, Y., Wu, B.-L., & Shen, Y. (2011). A pilot study on the diagnostic performance of DSM-IV and DSM-V for autism spectrum disorder. North American Journal of Medicine and Science, 4(3), 116–123.CrossRefGoogle Scholar
  92. Young, R. L., & Rodi, M. L. (2014). Redefining autism spectrum disorder using DSM-5: the implications of the proposed DSM-5 criteria for autism spectrum disorders. The Journal of Autism and Developmental Disorders, 44(4), 758–765.  https://doi.org/10.1007/s10803-013-1927-3.CrossRefPubMedGoogle Scholar
  93. Yuan, H., & Dollaghan, C. (2018). Measuring the diagnostic features of social (pragmatic) communication disorder: An exploratory study. American Journal of Speech-Language Pathology, 27(2), 647–656.  https://doi.org/10.1044/2018_ajslp-16-0219.CrossRefPubMedGoogle Scholar
  94. Zander, E., & Bolte, S. (2015). The new DSM-5 impairment criterion: A challenge to early autism spectrum disorder diagnosis? The Journal of Autism and Developmental Disorders, 45(11), 3634–3643.  https://doi.org/10.1007/s10803-015-2512-8.CrossRefPubMedGoogle Scholar
  95. Zhao, F., Zhang, H., Rekik, I., An, Z., & Shen, D. (2018). Diagnosis of autism spectrum disorders using multi-level high-order functional networks derived from resting-state functional MRI. Frontiers in Human Neuroscience, 12, 184.  https://doi.org/10.3389/fnhum.2018.00184.CrossRefPubMedPubMedCentralGoogle Scholar

Copyright information

© Springer Science+Business Media, LLC, part of Springer Nature 2019

Authors and Affiliations

  • Kristine M. Kulage
    • 1
    Email author
  • Johanna Goldberg
    • 2
  • John Usseglio
    • 3
  • Danielle Romero
    • 4
  • Jennifer M. Bain
    • 5
  • Arlene M. Smaldone
    • 1
  1. 1.Columbia University School of NursingNew YorkUSA
  2. 2.Memorial Sloan Kettering Cancer CenterNew YorkUSA
  3. 3.Augustus C. Long Health Sciences LibraryColumbia University Irving Medical CenterNew YorkUSA
  4. 4.Lucile Packard Children’s Hospital at StanfordPalo AltoUSA
  5. 5.Department of Neurology, Division of Child NeurologyColumbia University Vagelos College of Physicians and SurgeonsNew YorkUSA

Personalised recommendations