Autism spectrum disorder (ASD) is neurodevelopmental pathology that manifest as deficits in certain fundamental areas.
The diagnostic requirements for ASD have just two criteria, namely A) deficits in communication and social interaction and B) restricted and repetitive behaviors and interests, as shown by the current Diagnostic and Statistical Manual of Mental Disorders, Fifth Edition (DSM-5) diagnostic scheme.
Considering data from the last decades, several studies have been published to support the hypothesis of an increased number of ASD cases. Prevalence differs in some populational studies from North America, indicating 1 out of 68 children in 2012 [1] or 1 out of 59 in 2014 [2]. The most recent published study indicated that ASD prevalence was 13.4 per 1000 children aged 4 years in 2010, 15.3 in 2012, and 17.0 in 2014. It has been proposed that the average age of diagnosis is three or four years of age, or even later for children of low socioeconomic status or without a previous family history of ASD. The period between initial suspicion of ASD and final diagnosis can be a stressful and confusing time for families, and early identification and intervention can address this issue [3]. Nonetheless, perhaps the most crucial reason that justifies the need to detect the first symptoms is that it allows for intensive early intervention that leads to a better long-term prognosis [4]. In fact, in retrospective reports, most parents of children with ASD recall concerns regarding their child’s development after the first year of life [5].
Although screening for ASD is recommended at 18 months using M-CHAT, according to studies that have been conducted with high-risk siblings, sufficient evidence has been found to affirm that there are differential behavioral markers of ASD between 12 and 18 months of age [6]. Additionally, the M-CHAT instrument performs with a low-to-moderate accuracy in children with developmental concerns [7].
For effective screening, both specific screening tools (designed to identify children with risk signs of ASD) and wide-range screening tools can be used. With this second approach, children with any developmental difficulties will be detected in a first examination, and specific ASD tools will be used in the follow-up on this group. More accurate screening tools are essential for proper early detection. The best tools are those with higher values for sensitivity and positive predictive values (PPV). Sensitivity refers to the proportion of children correctly identified as “high-risk”, and PPV is defined as the proportion of children with positive screening who finally receive a diagnosis [7].
Table 1 shows the level 1 screening instruments applied from 12 months of age that met the abovementioned inclusion criteria. Screening instruments that can be applied before 18 months are shaded.
Table 1. Screening instruments for a population under two years of age.
| Instrument | Administration Age (Months) | Used Sample (Children) | Sensitivity | Specificity | Positive Predictive Value (PPV) | Negative Predictive Value (NPV) | Reference |
|---|
][44]. The screening takes approximately 30 min to complete.
FYI: (First Year Inventory): This is a screening parent-report questionnaire [22][24]. It is a 63-item questionnaire developed as a general population-screening tool to identify 12-month-old infants who might be at risk for autism spectrum disorder (ASD) or a related developmental disorder. The screening takes approximately 30 min to complete.
ITC/CSBS-DP (Infant-Toddler Checklist/Communication and Symbolic Behavior Scales Developmental Profile): This is a 24-item questionnaire used as a screening tool for children from 9 months old and is a part of the evolutionary profile of the CSBS-DP scales. It was originally developed as a measure of language detection and evaluates social and communication behaviors [43][45]. It takes 10 min to complete.
PREAUT-Grid (Program of Research and Studies on AUTISM): A 10-item test to be administered by the pediatrician from observing how the child interacts with the pediatrician and his mother [35][37]. Evaluates children from 4 to 24 months old.
APSI (Autism Parent Screen for Infants): Evaluates children from 6 to 24 months old [8][10]. It is a 26-item forced-choice (yes, sometimes, no) parent-report questionnaire designed to monitor early signs of ASD in infants aged 6–24 months and takes approximately 10–15 min to complete.
CESDD (Checklist for Early Signs of Developmental Disorders): A 12-item checklist completed by nursery staff for children from 3 to 39 months [17][19]. It takes 10 min to complete.
SACS (Social Attention and Communication Study): A 15-item tool completed by clinicians to assess children from 8 to 24 months old [37][39]. It takes 10 min to complete.
STAT (Screening Tool for Autism in Toddlers and Young Children): A tool for children from 12 to 23 months old [38][40]. It consists of 12 activities assessing play, communication, and imitation skills and takes 15–20 min to administer by a trained clinician.
Table 2 includes the screening instruments found in this search that were valid to be used in children after 2 years.
Table 2. Screening instruments used in children older than 2 years.
Supplementary Table S1. These points need to be taken into account: (A) The samples included in the studies were largely heterogeneous from the general population, including children from different groups with ASD, OPD, or typical development. (B) Several studies were not using a control group. (C) There were limitations of sample size in several studies, which makes it hard to conclude the predictive ability of the screening instruments in larger populations. (D) Different patient ages in the groups were found that may affect the results. (E) Some studies indicated that the scale was performed by the parents alone and other studies by a child psychologist during a home visit or at the clinic.
| Instrument | Administration Age (Years) | Sample Used | Sensitivity | Specificity | Positive Predictive Value (PPV) | Negative Predictive Value (NPV) | Reference | ||||||||
|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|
| APSI | 6–24 | 204 high-risk, 79 low-risk controls | 0.67 | 0.87 | 6–8 months: 0.43 | ||||||||||
| AQ-10 Child | 4–11 | 432 ASD | 21–24 months: 0.79 | 0.87–0.99 | [ | 0.958 | 0.97][ | 0.9410] | |||||||
| --- | [ | 36 | ] | [ | 38 | ] | BeDevel | 18–42 | 155 (75 ASD, 55 NT, 25 OPD) | ||||||
| ASRS | 2–18 | 0.83 | 37 ASD, 30 OPD |
Parents: 0.64 Teachers: 0.520.81 |
Parents: 63.30 0.80 |
Teachers: 0.71 | Parents: 0.680.83 | Teachers: 0.61[ |
Parents: 0.59 Teachers: 0.629][11] |
||||||
| [ | 44 | ] | [ | 46 | ] | BDI-II | 0–95 | 604 ASD, | |||||||
| BASC-2 (PRS-P) | 1064 NT | 2–50.94 | 0.31 | 224 positive ASD screening | 0.76 | 0.73--- | --- | [ | 0.8610 | 0.57][12] | |||||
| [ | 45 | ] | [ | 47 | ] | BISCUIT | |||||||||
| CASD | 17–37 | 1007 risk of ASD |
CASD SF | 3–17 | 469 ASD,0.93 | 0.86 |
138 OPD | CASD: 0.86 CASD SF: 0.95… |
CASD: 100 … |
CASD SF: 0.96 | …[11][13] | ||||
| … | [ | 46 | ] | [ | 48 | ] | BITSEA | 11–48 | 223 ASD, 289 non-ASD |
||||||
| CAST | Up to 24 months: 0.91 | 4–12 | 1496 general population | Complete version: 0.83 Reduced version: 0.85Up to 24 months: 0.80 |
--- | --- | Complete version: 0.92 Reduced version: 0.91[12][14] |
||||||||
| Complete version: 0.63 | Reduced version: 0.61 | --- | [ | 47 | ] | [49] | CASI (for Indi) | 18-120 | 405 (75 ID, 83 ASD, 87 OPD, 160 NT) | 0.89 | 0.89 | 0.67 | 0.96 | [13][15] | |
| CBCL 1.5-5 | 18–71 | 226 ASD, 163 OPD |
0.74 | 0.53 | --- | --- | [14][16] | ||||||||
| CBCL 1.5-5 (Withdrawn PDP Scale) |
18–59 | 101 ASD, 117 NT |
Withdrawn: 0.89 PDP: 0.85 |
Withdrawn: 0.92 PDP: 0.90 |
Withdrawn: 0.90 PDP: 0.88 |
Withdrawn: 0.90 PDP: 0.87 |
[15][17] | ||||||||
| 18–59 | 80 ASD, 103 OPD |
Withdrawn: 0.88 PDP: 0.83 |
Withdrawn: 0.63 PDP: 0.60 |
Withdrawn: 0.65 PDP: 0.62 |
Withdrawn: 0.87 PDP: 0.82 |
[16][18] | |||||||||
| CESDD | 3–39 | 357 risk of ASD or language problems | 0.92 and 0.90 | 0.73 and 0.68 | 0.19 and 0.32 | 0.99 and 0.98 | [17][19] | ||||||||
| CS-TSA | 18–60 | 27 ASD, 41 OPD, 64 NT |
Section 1: 0.89 Section 2: 0.78 |
Section 1: 0.68 Section 2: 0.79 |
--- | --- | [18][20] | ||||||||
| DBC-ES | 18–48 | n = 142 ASD or PDD n = 65 OPD |
0.83 | 0.48 | 0.78 | 0.56 | [19][21] | ||||||||
| Screening questionnaire developed for Taiwan | 18–24 | 18 ASD, 59 NT |
1 | 0.96 | |||||||||||
| 32 | |||||||||||||||
| ] | |||||||||||||||
| EDUTEA | 3–12 | 2660 general population | 0.87 | 0.91 | 0.86 | 0.99 | [48][50] | ||||||||
| M-CHAT and JA-OBS | 2–3 | 3999 general population | 0.90 | --- | 0.96 | --- | [49][51] | ||||||||
| OERA | 3–10 | 76 ASD, 23 non-ASD |
0.92 | 0.91 | 0.92 | --- | [50][52] | 0.31 | 0.99 | [51]0.90 | 1 | [20][22] | |||
| [ | 53 | ] | ECSA (Brief) | 18–60 | 69 ASD | 0.89 | 0.85 | --- | --- | [21][23] | |||||
| FYI | 12 | 38 ASD, 15 NO-ASD, developmental delay 40 NT |
0.92 | 0.78 | 0.74 | 0.93 | [22][24] | ||||||||
| ITC/CSBS-DP | 6–24 | 5385 general population | 0.93 | 0.89 | 12–24 months: +0.70 | 12–24 months: +0.87 | [23][25] | ||||||||
| M-CHAT | 16–30 | Sample 1: 2480 High- and low-risk Sample 2: 2055 low-risk |
Both samples: 1 | Both samples: 0.98 | M1: 0.35 M2: 0.19 |
1 | [24][26] | ||||||||
| 122 risk of ASD, 106 NT |
0.86 | 0.80 | 0.88 | --- | [25][27] | ||||||||||
| 966 born very premature | 0.52 | 0.84 | 0.20 | 0.96 | [26][28] | ||||||||||
| PAAS | 18–48 | 105 general population | |||||||||||||
| SCQ | 3–7 | 219 premature children | 0.91 | 0.86 | 109 language delay 732 NT |
0.90 | 0.99 | 0.96 | 0.99 | [27][29] | |||||
| M-CHAT/F | 16/30 | 341 positive for M-CHAT | 0.55 | 0.79 | 0.78 | --- | [28][30] | ||||||||
| M-CHAT-R/F | 16–30 | 15,612 general population | 0.96 | 0.86 | 0.47 | 0.99 | [29][31] | ||||||||
| M-CHAT-JV/F | 16–30 | 1851 general population | 0.47 | 0.98 | 0.45 | --- | [30][0.88 | 0.93 | 0.95 | 0.84 | [31][33] | ||||
| PDQ-1 | 18–36 | 42 ASD, 38 OPD, 100 NT |
0.85 | 0.99 | 0.88 | 0.99 | [32][34] | ||||||||
| POEMS | 1–24 | 108 high-risk | 0.74 | 0.73 | [33][35] | ||||||||||
| POSI | 16–48 | 232 children (16–36 months) | 0.83 | 0.75 | --- | --- | [34][36] | ||||||||
| PREAUT Grid | 4–24 | 4755 general population |
P-4: 0.16 a 0.21 P-9: 0.30 a 0.41 P-24: 0.40–0.41 |
P-4: 0.99 P-9: 0.99 P-24: 0.99 |
P-4: 0.25 a 0.26 P-9: 0.20 a 0.36 P-24: 0.26 a 0.27 |
P-4: 0.99 P-9: 0.99 P-24: 0.99 |
[35][37] | ||||||||
| Q-CHAT-10 | 18–24 | 126 ASD | 0.91 | 0.89 | 0.58 | --- | [36][38] | ||||||||
| SACS | 12–24 | 20,770 general population | 0.69 a 0.83 | 0.99 | 0.81 | --- | [37][39] | ||||||||
| STAT | 12–23 | 71 siblings with ASD | 0.95 | 0.73 | 0.56 | 0.97 | [38][40] | ||||||||
| SORF | 18–24 | n = 84 ASD, n = 82 OPD, n = 62 NT |
0.80 | 0.78 | 0.81 | 0.78 | [39][41] | ||||||||
| TIDOS | 18–60 | n = 86 ASD n = 76 OPD n = 97 general population |
0.95 | 0.85 | 0.91 | 0.90 | [40][42] | ||||||||
| YACHT-18 | 18 | 2814 general population |
0.60 ASD, 0.82 other developmental disorders |
0.86 for developmental disorders | --- | --- | [41][43] |
ASD: Autism Spectrum disorder. APSI: Autism Parent Screen for Infants. BDI-II: Battelle Developmental Inventory, second edition. BISCUIT: Baby and Infant Screen for Children with autism Traits. BITSEA: Brief Infant-Toddler Social and Emotional Assessment. CBCL: Child Behavior Checklist. PDP: Pervasive Developmental Problems. CESDD: Checklist for Early Signs of Developmental Disorders. CASI: Chandigarh Autism Screening Instrument. CS-TSA: “Chestionarul de Screening Pentru Tulburări de Spectru Autist”. DBC-ES: Developmental Behavior Checklist: Early Screen. ECSA: Brief Early Childhood Screening Assessment. FYI: First Year Inventory. ID: Intellectual Disability. ITC/CSBS-DP: Infant-Toddler Checklist/Communication and Symbolic Behavior Scales Developmental Profile. M-CHAT: Modified Checklist for Autism in Toddlers. M-CHAT-R: Modified Checklist for Autism in Toddlers, Revised. M-CHAT-R/F: Modified Checklist for Autism in Toddlers, Revised with Follow-Up. M-CHAT-JV/F: Japanese Version of the Modified Checklist for Autism in Toddlers, Revised with Follow-Up. NT: Normotypic. OPD: Other Pervasive Disorders. PAAS: Pictorial Autism Assessment Schedule. PDQ-1: Psychological Development Questionnaire-1. POEMS: Parent Observation of Early Markers Scale. POSI: Parent’s Observational Screen of Social Interactions. PPV: Positive predictive value. NPV: Negative predictive value. PREAUT Grid: “Programme de Recherches et d’Etudes sur l’Autisme”. P-4, P-9, P-24: PREAUT 4, 9 or 24 months. Q-CHAT-10: Quantitative Checklist for Autism in Toddlers. SACS: Social Attention and Communication Study. STAT: Screening Tool for Autism in Toddlers and Young Children. TIDOS: Three-Item Direct Observation Screen. YACHT-18: Young Autism and Other Developmental Disorders Checkup Tool. Shaded: Screening instruments that can be applied before 18 months.
Among all the instruments identified in this work, the following instruments stand out for their predictive value and their easy and early application from 12 months of age:
BDI-II (Battelle development inventory, second edition): Evaluates children from 12 months to 8 years [10][12]. It helps measure a child’s progress along this developmental continuum by both global domains and discrete skill sets. The screening tool takes approximately 10–30 min to complete.
BITSEA (Brief Infant-Toddler Social and Emotional Assessment): Evaluates children from 11 to 48 months old. The 42-item questionnaire aims to identify children with social-emotional problems, behavior, or deficits in social and emotional competence [42
AQ-10: Short Autism Spectrum Quotient. ASD: Autism spectrum disorder. ASRS: Autism Spectrum Rating Scales. BASC-2: Behavior Assessment System for Children, Second Edition. PRS-P Scale: Parent Rating Scales—Preschool. CASD: Checklist for Autism Spectrum Disorder. CASD-SF: Checklist for Autism Spectrum Disorder, Short Form. CAST: Childhood Autism Spectrum Test. M-CHAT: Modified Checklist for Autism in Toddlers. JA-OBS: Joint Attention Observation of Toddlers. OPD: Other Pervasive Disorders. OREA: Structured Observation for Autism Screening (acronym in Portuguese). PPV: Positive predictive value. NPV: Negative predictive value. SCQ: Social Communication
In order to analyze the risk of bias in the eligible studies, an assessment was done, according to the Cochrane Handbook for Systematic Reviews of Interventions Version 5.1.0. Details are shown in