Comparative cross-sectional analysis of the Griffiths III and the Developmental Autism Early Screening (DAES) test in children with neurodevelopmental disorders

Abstract

Background

Global Developmental Delay (DD) and Autism Spectrum Disorder (ASD), are neurodevelopmental disorders (NDDs) characterized by early impairments in communication, cognition, and socio-emotional functioning. Early identification remains challenging due to symptom heterogeneity and overlaps across conditions. We aimed to investigate possible differences in the early developmental profiles among children aged 18 to 48 months at risk for NDDs using the Griffiths Mental Development Scales (Griffiths III) and the Developmental Autism Early Screening (DAES), an observational and interactive Level 2 screening tool.

Methods

Significant differences across the developmental domains assessed with the Griffiths III and the DAES were examined by comparing children referred for risk of ASD or DD with a control group of typically developing children (CTRL). A total of 310 participants were enrolled. Sensitivity, specificity, positive predictive value (PPV), and negative predictive value (NPV) of the DAES total score were calculated. Spearman’s rank correlation was used to assess the association between the DAES total score and the ADOS-2 Calibrated Severity Score (CSS).

Results

Significant differences emerged between groups across the Griffiths III subscales, with the most pronounced impairments observed in the Language and Communication (p < 0.00001) and Personal-Social-Emotional (p < 0.00001) domains. DAES total scores significantly differentiated the groups, with the highest scores recorded in the ASD cohort with respect to DD (p = 0.0004) and CTRL (p = 0.00001). Notably, higher DAES scores were associated with greater ASD symptom severity as measured by the ADOS-2 CSS (p = 0.013).

Conclusion

The study highlights the clinical value of using Griffiths III and the DAES to support the early identification of children at risk for ASD compared with those with developmental delay (DD) and typical development. Integrating these measures into routine clinical assessment may improve the early detection of atypical developmental trajectories and guide individualized intervention planning during critical periods of neurodevelopment.

Background

Neurodevelopmental disorders (NDDs) are conditions that emerge during development and are characterized by impairments in personal, social, academic, or occupational functioning [1]. Their prevalence among individuals under 18 years of age varies widely, and they occur more frequently in males. Comorbidities are common, both across different NDD subtypes and with other psychiatric conditions [2, 3]. Among NDDs, global Developmental Delay (DD) and Autism Spectrum Disorder (ASD) have major impacts on developmental trajectories, affecting communication, cognition, and adaptive functioning [4, 5].

Developmental delay (DD) is characterized by significant delays in at least two developmental domains, including motor, language, cognitive, social, or adaptive functioning. Its global prevalence is estimated at 1%-3%, although rates vary depending on diagnostic criteria, population characteristics, and access to developmental services [6]. Early recognition and etiological assessment are essential to enable timely intervention and improve long-term developmental outcomes [7]. ASD is a heterogeneous condition marked by difficulties in communication, social interaction, and restricted or repetitive behaviors [1]. Prevalence continues to rise, reaching 32.2 per 1,000 children aged 8 years in the U.S., with a 3.4:1 male-to-female ratio [8]. In Italy, prevalence is 1.3% among children aged 7–9 years (male-to-female ratio 4.4:1) [9]. Diagnosis relies on behavioral assessments, developmental history, clinical judgment, and standardized tools, as no biological markers exist [10–12]. Symptoms typically emerge within the first two years, often with delays or regression in social and communication skills [1], and early signs may appear around 12 months [13]. Early diagnosis is vital as timely intervention improves outcomes [14, 15]. Early identification remains challenging due to overlap with other developmental conditions and typical variability [16].

The American Academy of Pediatrics recommends ASD screening at 18 and 24 months using developmental monitoring and standardized tools [17, 11]. Screening instruments focus on early communication markers such as joint attention, social interaction, eye contact, and play [18].

ASD screening instruments are typically classified into Level 1 (general population) and Level 2 (children at elevated risk or who fail Level 1) [19]. Level 1 tools include the M-CHAT and M-CHAT-R/F [20, 21], Q-CHAT [22, 23], ESAT [24, 25], and CESDD [26]. Level 2 tools comprise STAT [27], BISCUIT-Part 1 [28], ADEC [29], SORF [30], RITA-T [31], and ESAC [32]. The STAT and RITA-T currently have the strongest empirical support among Level 2 screening tools for children aged ≤ 36 months, demonstrating their ability to improve the efficiency of the diagnostic process for ASD [33].

The Griffiths Mental Development Scales (Griffiths III) [34] represent a consistent tool for measuring the child developmental profile based on feasibility and cross-cultural validity. Increasing evidence pointed to the possible use of the Griffiths III in enhancing recognition of some developmental patterns across neurodevelopmental disorders, such as ASD [35–37]. Neurodevelopmental profiling with the Griffiths III has also helped identify early predictors of later verbal abilities in large ASD cohorts [38]. Building on this evidence, Cirnigliaro et al. [39] developed the Developmental Autism Early Screening (DAES), an observational and interactive Level 2 screening tool designed to detect early atypical developmental trajectories in children at risk for ASD and to distinguish them from peers with DD or typical development.

The DAES includes 36 items derived from the Griffith III that are most predictive of ASD risk. These items are organized within the B (Language and Communication) and D (Personal-Social-Emotional) subscales, domains typically affected in ASD [39]. They cover a wide range of developmental functions, including auditory attention, communicative intent, expressive and receptive language, reciprocity, social interaction, play, social cognition, joint attention, self-awareness, and emotional understanding [39].

Clinically, the DAES [39] can be used with children already assessed with the Griffiths III to evaluate ASD risk between 18 and 48 months of age (developmental age 12–36 months). Its brief administration time and lack of need for specialized personnel make it a practical level 2 screening option for toddlers showing atypical behaviors and/or developmental delays.

The present study was undertaken to evaluate possible differences on the DAES in children aged 18–48 months at risk for NDDs. We sought to verify the discriminative ability of this instrument in distinguishing children with a true risk of ASD, with respect to children with DD or typical development. Based on the results of this analysis, it is expected that the DAES could complement existing level 2 screening tools across various settings.

Methods

Participants

We conducted a cross-sectional study to analyze significant differences across various developmental domains by comparing children referred for ASD or DD risk with a control group of typically developing children (CTRL). Participants with chronological age (CA) ranging from 12 to 48 months, were enrolled according with inclusion criteria: (a) developmental age (DA) between 12 and 36 months on the Griffiths III A-subscale; (b) ASD group: diagnosis of ASD based on the Autism Diagnostic Interview-R (ADI-R) [40] and the Autism Diagnostic Observation Schedule–Second Edition (ADOS-2) [41], according to DSM-5-TR criteria; (c) DD group: diagnosis of global DD established by using the Griffiths III-III, according to DSM-5-TR criteria; (d) CTRL group: a General Developmental Quotient (GDQ) greater than 85 on the Griffiths III. Exclusion criteria included bilingualism, the presence of genetic syndromes with ASD or DD, and the diagnosis of associated neurological disorders such as epilepsy or hearing and visual impairments.

Participants were recruited between January 2024 and September 2025 at the Child Neurology and Psychiatry Section, Department of Clinical and Experimental Medicine, University of Catania. A total of 310 children, were enrolled and divided into the three above mentioned groups.

Written informed consent from both parents was acquired. The current study was part of an overall larger study aimed at identifying markers, predictors and developmental trajectories of ASD. The larger overall study was approved by the local ethics committee at the University Hospital Referral Centre for ASD with number n° 759. All procedures performed in the present study were in accordance with the 1964 Declaration of Helsinki and its later amendments (2013).

Standardized measures

All participants were assessed using the Griffiths III [34], which provides a comprehensive measure of overall developmental profiles, including cognitive, language, motor, and socio-emotional domains, from birth to 72 months. ASD symptoms were assessed using the gold-standard Autism Diagnostic Observation Schedule, Second Edition (ADOS-2). According to participants’ age and language level, the appropriate module was administered. ADOS-2 scores were converted into Calibrated Severity Scores (CSS; range 1–10), with higher scores indicating greater autism severity [42, 43].

Moreover, all participants underwent the DAES [39] test to identify specific indicators of ASD risk in children younger than 48 months with a developmental age of 12 to 36 months.

DAES administration

The DAES was administered in clinical settings to children “clinically referred” for concerns related to ASD or DD, all of whom had previously completed the Griffiths III assessment. The DAES consists of 36 items derived from the Griffiths III B and D subscales, selected for their strong predictive value in distinguishing children with ASD or DD from CTRL with typical development [39]. DAES was scored after the Griffiths III administration and did not require a separate assessment session. Administration within each subscale was discontinued after six consecutive failed items. For each item not passed (i.e., skill not achieved), a score of + 1 or + 2 was assigned.

According with DAES princeps study, items that significantly differentiated children with ASD from both DD and CTRL groups were assigned a score of + 2, whereas items that differentiated ASD children only from the CTRL group were assigned a score of + 1 [39]. Higher total scores indicate greater risk for ASD, with a DAES total score ≥ 15 reflecting a moderate-to-high risk level [39].

Statistical analysis

One-way ANOVA was used to compare quantitative variables, including chronological age (CA), developmental age derived from the Griffith III A subscale (DA), and Griffiths III total and subscale (A-E) developmental quotients (DQs), across the three participant groups (ASD, DD, and CTRL). When significant group effects were detected, post hoc multiple comparisons were performed using Tukey’s HSD test to identify which group differences accounted for the observed effects.

For the validation of the DAES, total DAES scores were compared across the three groups using ANOVA followed by pairwise post hoc analyses. Diagnostic accuracy metrics, including sensitivity, specificity, positive predictive value (PPV), and negative predictive value (NPV), were also calculated for the DAES total score. 95% confidence intervals for sensitivity, specificity, PPV, and NPV were calculated using binomial proportion confidence intervals.

The association between the DAES total score and the ADOS-2 Calibrated Severity Score (CSS) was examined using Spearman’s rank correlation, with the strength of the relationship quantified using Spearman’s rho (R).

To account for multiple testing across developmental outcomes, a Benjamini–Hochberg false discovery rate (FDR) correction was applied to the p-values of the main ANOVA analyses.

The significance threshold was set at α ≤ 0.05. All statistical analyses were conducted using SPSS version 27 (IBM Corp., Armonk, NY, USA).

Results

This cross-sectional study considered 310 children including 93 with ASD-risk, 112 with DD-risk and 105 CTRL (Table 1). The latter group comprised healthy children which underwent clinical follow-up for transient conditions such as neonatal jaundice, suspected maternal infection, fever or prematurity.

Table 1.

Demographic features of participants

Participants	ASD	DD	CTRL
Number (n)	93	112	105
Male (n, %)	73 (78%)	79 (71%)	71 (68%)
Female (n, %)	20 (22%)	33 (29%)	34 (32%)
CA (months) Mean ± SD	32.96 ± 8.45	32.69 ± 8.71	23.79 ± 8.11
DA (months) Mean ± SD	19.55 ± 5.72	22.47 ± 8.42	22.58 ± 7.55

ASD: Autism Spectrum Disorder; DD: Developmental Delay; CTRL: Control Group; CA: Chronological Age; DA: Developmental Age; SD: Standard Deviation

Chronological age (CA) was comparable between the ASD and DD groups, whereas the CTRL group showed a significantly lower CA.

The ASD group displayed a lower mean developmental age (DA) (19.55 ± 5.72 months) than the DD (22.47 ± 8.42 months) and CTRL groups (22.58 ± 7.55 months). A significant overall group effect in the ANOVA analyses was observed, F(2, 307) = 3.29, p = 0.039, η² = 0.02. Post hoc Tukey tests did not reveal any significant DA differences comparing the ASD group to both the DD group (p = 0.054) and the CTRL group (p = 0.068). This indicates that the children in each group had a similar non-verbal cognitive level.

Global DQs on the Griffith III across ASD, DD, and CTRL groups was significantly different (F = 48.88; p < 0.00001). Post-hoc pairwise comparisons showed that DQ scores did not differ significantly between children with ASD and DD, whereas both groups scored significantly lower than the neurotypical CTRL group. This pattern aligns with expectations, confirming that developmental delays are present in ASD and DD groups relative to typically developing peers (Table 2).

Table 2.

Comparison of global DQs on the Griffiths III among groups (Tukey HSD test)

Pairwise comparisons	Mean (Difference)	Standard Error	p
ASD: DD	7.67	0.96	p = 0.2066
ASD: CTRL	39.94	1.11	p < 0.00001**
DD: CTRL	32.27	1.03	p = 0.00001**

ASD: Autism Spectrum Disorder; DD: Developmental Delay; CTRL: Control Group.**Statistically significant values (p ≤ 0.001)

Then, we focused on identifying differences among the ASD, DD and CTRL groups on the Griffiths III subscales. Comparisons were therefore conducted between the mean DQs of each Griffiths III subscale (A-E), analyzed separately (Fig. 1). A significant overall difference in B-subscale scores (Language and Communication) was found (F = 159.73, p < 0.00001). Post Hoc comparisons showed that both the ASD group and the DD group scored significantly lower than the CTRL group (Table 3).

Fig. 1.

Comparison of Griffiths III DQ subscales scores across the three groups (ANOVA test). ASD: Autism Spectrum Disorder; DD: Developmental Delay; CTRL: Control Group. Statistically significant values. A: Griffiths III subscale A; B: Griffiths III subscale B; C: Griffiths III subscale C; D: Griffiths III subscale D; E: Griffiths III subscale E; GDQ: Griffiths III General Developmental Quotient. Error bars represent standard error of the mean (SEM). ** p < 0.001

Table 3.

Comparison of DQs on the Griffiths III subscales B and D among groups (Tukey HSD test)

Pairwise comparisons	Mean (Difference)	Standard Error	p
B subscale DQ
ASD: DD	5.95	1.56	0.00494*
ASD: CTRL	31.75	1.31	< 0.00001**
DD: CTRL	25.8	1.43	< 0.00001**
D subscale DQ
ASD: DD	11.99	2.18	< 0.00001**
ASD: CTRL	48.99	1.89	< 0.00001**
DD: CTRL	36.99	2.21	< 0.00001**

ASD: Autism Spectrum Disorder; DD: Developmental Delay; CTRL: Control Group. * p < 0.05; ** p < 0.01

A highly significant difference among all three groups (F = 318.57; p < 0.00001) was also found in the D subscale (Personal-Social-Emotional), (F = 318.57; p < 0.00001), confirmed by Post Hoc tests. The DD group scored higher than the ASD group, while the CTRL group showed the highest performance overall (Table 3). After FDR correction, significant effects were confirmed for GDQ and subscale analyses. The largest gap occurred between ASD and CTRL, indicating substantially lower socio-emotional performance in children with ASD. Overall, the Griffiths III D subscale showed good discriminatory capacity across the study groups.

DAES validation analyses

The DAES was administered to all 310 children across the ASD, DD, and CTRL groups. Mean total DAES scores were highest in the ASD group (16.36 ± 6.9), followed by the DD group (8.87 ± 6.5) and the CTRL group (5.68 ± 3.9).

A significant difference among the three groups (F = 84.01; p < 0.00001) was observed, indicating that DAES scores varied meaningfully across diagnostic categories. Post hoc comparisons confirmed significant pairwise differences between all groups (Table 4).

Table 4.

DAES total scores comparisons among groups (Tukey HSD test)

Pairwise comparisons	Mean (Difference)	Standard Error	p
ASD: DD	7.49	0.94	p = 0.00040**
ASD: CTRL	10.68	0.81	p = 0.00001**
DD: CTRL	3.19	0.73	p = 0.00039**

ASD: Autism Spectrum Disorder; DD: Developmental Delay; CTRL: Control Group; CA: Chronological Age; DA: Developmental Age; Statistically significant values (** p ≤ 0.01) 0,812962756

The discriminative ability of the DAES to identify children with ASD was examined across three comparisons: ASD versus CTRL, ASD versus DD, and ASD versus a combined non-ASD group (DD + CTRL) (Table 5).

Table 5.

DAES discriminative abilities among groups

Comparison	Sensitivity (95% CI)	Specificity (95% CI)	PPV (95% CI)	NPV (95% CI)
ASD vs. CTRL	68.8% (58.8–77.3)	94.3% (88.1–97.4)	91.4% (82.5–96.0)	77.3% (69.4–83.7)
ASD vs. DD	68.8% (58.8–77.3)	77.0% (68.4–83.8)	71.1% (61.0–79.5)	75.0% (66.4–82.0)
ASD vs. DD + CTRL	68.8% (58.8–77.3)	85.3% (80.0–89.4)	66.7% (56.8–75.3)	86.5% (81.3–90.4)

ASD: Autism Spectrum Disorder; DD: Developmental Delay; CTRL: Control Group; CI: Confidence Interval; PPV: Positive Predictive Value; NPV: Negative Predictive Value

A total cut-off score of 12.5 was applied, in line with the DAES principal study [39], which showed that this threshold reliably distinguishes ASD from DD and neurotypical development.

When evaluating the DAES ability to distinguish ASD from the CTRL group, sensitivity was 68.8% and specificity 94.3%, with a positive predictive value (PPV) 91.4% and a negative predictive value (NPV) 77.3%. These metrics indicate that a positive DAES result is highly indicative of ASD, whereas a negative result is less reliable for confidently ruling it out. In the ASD vs. DD comparison, the DAES showed moderate accuracy (68.8% sensitivity, 77.0% specificity, 71.1% PPV and 75.0% NPV), correctly identifying about 70% of children with ASD and correctly classifying roughly three-quarters of those with DD as non-ASD.

When ASD was compared with the combined non-ASD group (DD + CTRL), there was an increase of DAES performance with 86.5% of NPV, while showing 68.8% sensitivity, 85.3% specificity and PPV 66.7%. DAES may be more effective in excluding non-ASD conditions in the broader non-ASD population, although its sensitivity remained moderate across all comparisons. Nonetheless, the occurrence of both false positives and false negatives in the ASD vs. DD comparison underscores the need to interpret DAES scores within a comprehensive, multi-method clinical assessment rather than relying on the tool as a standalone diagnostic measure.

At the end, relationships among different measures were computed to support validity of the DAES by searching possible correlations between DAES and ADOS-2 scores.

We observed that more than 80% of children with ASD who obtained a DAES total score above 15 (n = 48), indicating a moderate-to-high risk of ASD [39], also showed a “moderate-to-high” level of autism symptom severity according to the ADOS-2 evaluation (n = 39). In line with these findings, a linear regression analysis was conducted to examine the relationship between the DAES total score and the ADOS-2 total score. To allow comparisons across different ADOS-2 modules, Calibrated Severity Scores (CSS) were used in the correlation analyses. A significant positive association emerged between DAES total scores and ADOS-2 CSS (R = 0.27, p = 0.013), indicating that higher DAES scores are associated with greater ASD symptom severity (Fig. 2).

Fig. 2.

Linear regression analysis between ADOS-2 CSS and DAES Total Score. DAES: Developmental Autism Early Screening; ADOS-2: Autism Diagnostic Observation Schedule, 2nd edition; CSS: Calibrated Severity Score

These findings support the validity of the DAES as a standardized measure for quantifying autism-related behaviors in early childhood. They further underscore the DAES utility and reliability as a component of early ASD screening protocols.

Discussion

In the present study, we aimed to describe the developmental profiles of children with ASD, DD, and typical development (CTRL), and to evaluate the discriminative ability of the DAES as a Level 2 screening instrument [39]. To this end, we conducted a cross-sectional analysis of developmental profiles in an independent referral sample of 310 children using the Griffiths III and the DAES screening tool.

Previous research has emphasized the importance of assessing early developmental profiles in young children with an increased risk of ASD using standardized direct measures such as the Griffiths III [44]. This enables a comprehensive description of the child’s functional profile and provides important information on the potential co-occurrence of language, psychomotor, or intellectual difficulties. However, available evidence indicates that children with ASD tend to present widespread developmental impairments, with more difficulties in the Language and Communication (subscale B) and Personal-Social-Emotional (subscale D) domains [39, 44]. These domains have consistently emerged as the most affected when comparing ASD children with both typically developing peers and children with other developmental disorders. Lower scores in these areas have been associated with greater autism symptom severity, as reflected by negative correlations between autistic traits and performance on subscales B and D [44]. Interestingly, these same domains have also been shown to be the most responsive to early intervention. This highlights their relevance not only for early detection, but also for monitoring developmental change over time [45, 46]. Moreover, a substantial proportion of children with DD may meet diagnostic criteria for ASD, often exhibiting the most pronounced impairments in language and personal–social domains [5]. Examining a sample that includes both ASD and DD groups can therefore help identify distinctive developmental patterns, enhance differential diagnosis, guide individualized intervention planning, and support predictions of long-term functional outcomes.

In line with previous research [39, 47, 48], our results revealed significant differences in the Griffiths III DQs across different subscales among three groups (ASD, DD and CTRL), with special regard to Language and Communication (B) and Personal-Social-Emotional (D) subscales. The analysis of the B subscale revealed a marked reduction in linguistic and communicative abilities in the ASD group compared with both the DD and CTRL groups. These difficulties are evident in children limited attempts to initiate social interactions, narrate stories or personal experiences, and in the presence of atypical language patterns, including immediate echolalia or repetitive and stereotyped speech. In contrast, children with DD showed intermediate performance, suggesting that while developmental delay can affect communication skills, the qualitative differences observed in ASD are more pronounced and specific to the disorder [47, 48].

Regarding the D subscale, the results revealed a strong and statistically significant gap among the three groups. The most pronounced difference was observed between the ASD and CTRL groups, with the ASD group exhibiting the lowest scores. These findings are consistent with difficulties in emotional reciprocity, social interaction, and adaptive behaviours, core features of ASD [35, 39]. Previous studies have also emphasized that deficits in language and personal-social-emotional domains are strongly correlated with the severity of ASD [37, 38, 49].

Building upon these findings, the subsequent analysis focused on whether DAES total scores and risk levels were consistent with the developmental patterns previously observed and whether it could effectively differentiate among ASD, DD and neurotypical children. Consistent with former findings [39], in the present study the ASD group showed the highest DAES total scores, significantly differing from both the DD and CTRL groups. This pattern further underscores and validates the utility and reliability of the DAES as a Level 2 screening instrument for the early identification of ASD risk. Using a cut-off score of 12.5 points [39], the test yielded increased negative predictive values when comparing ASD with the combined non-ASD group (DD + CTRL). Based on present data, it is plausible that the DAES might be used for clinical screening purposes, being highly accurate when distinguishing ASD from a broader non-ASD population.

For further validation, we next examined the consistency of DAES scores and alignment with ADOS-2 assessments in the ASD group. Notably, DAES scores were significantly correlated with ADOS-2 CSS. Specifically, more than 80% of children with DAES total scores above 15, indicating moderate-to-severe risk [39], also showed a moderate-to-high level of autism risk symptoms on the ADOS-2 score. In line with this, linear regression analyses revealed a significant positive association between DAES total scores and ADOS-2 total, indicating that higher DAES scores correspond to greater ASD symptom severity. These findings support the convergent validity of the DAES and its potential role in quantifying autism-related behaviors in young children.

When interpreting these results, it is important to consider that participants’ ages ranged from 18 to 48 months, which prevented the inclusion and analysis of younger children at risk for ASD. Given that the DAES is a Level 2 screening tool, it is also conceivable that the referral age for children with neurodevelopmental concerns is typically higher than the age at first developmental surveillance. Future studies should aim to include younger children, particularly those under 18 months who may already exhibit early signs of ASD, as well as larger and more diverse samples with longitudinal follow-up. Such designs would help clarify how early language and socio-emotional scores relate to later adaptive and cognitive functioning. An additional limitation of this research is that most of the children with ASD in the study had a gap between their CA and DA. In this regard, we assured that non-significant differences might occur among DA in the study groups. Finally, we highlight, DAES lower performance to differentiate between ASD and diverse neurodevelopmental disorders, such as DD. This is consistent with level 2 screening challenges in distinguishing ASD from other neurodevelopmental conditions at certain ages [19].

In contrast, DAES key advantage is its immediate applicability in clinical contexts for “clinically referred” children, allowing for the evaluation of ASD risk between 18 and 48 months of chronological age. The DAES brief administration time and minimal training requirements prompt its use as a screening option for toddlers presenting atypical behavioural features or developmental delays.

Overall, this study, conducted in an independent referral sample, provides further validation of the DAES as a tool for stratifying neurodevelopmental risk and differentiating children with ASD, those with DD, and neurotypical peers. The implementation of standardized developmental assessments represents a valuable approach for characterizing NDDs at early ages, supporting the early identification of atypical developmental trajectories and facilitating the timely recognition of children at risk for autism during the crucial first years of life [35]. Early intervention during this developmental window offers the greatest potential to influence long-term outcomes [45, 46, 49].

Conclusion

In conclusion, our findings support the use of the DAES as an instrument for identification of ASD risk. Significant correlations between the DAES and ADOS-2 help validate the potential usefulness of the DAES as a rapid early screening measure of ASD. The concurrent application of the Griffiths III and the DAES may facilitate the timely detection of atypical developmental trajectories and assist clinicians in planning personalized interventions targeting communicative and socio-emotional domains, ultimately promoting more effective learning strategies and developmental outcomes of children at risk for ASD.

Abbreviations

NDDs

Neurodevelopmental Disorders

DAES

Developmental Autism Early Screening

ASD

Autism spectrum disorder

DD

Global developmental delay

CTRL

Neurotypical control group

GRIFFITHS III

Griffiths Mental Development Scales

ADI-R

Autism Diagnostic Interview-R

ADOS-2

Autism Diagnostic Observation Schedule–Second Edition

GDQ

General Developmental Quotient

DQ

Developmental Quotient

DA

Developmental age

CA

Chronological age

SA

Social affect

RRB

Restricted and repetitive behaviour

CSS

Calibrated severity score

PPV

Positive predictive value

NPV

Negative predictive value

SEM

Standard error of the mean

CI

Confidence Interval

Author contributions

M.R.: Writing-Original Draft, Writing-Review and Editing, Conceptualization, Data Curation, Investigation, Methodology, Formal Analysis. L.C.: Writing-Original Draft, Writing-Review and Editing, Conceptualization, Data Curation, Investigation, Methodology, Formal Analysis. G.C.: Writing-Original Draft, Formal Analysis, Data Curation. V.M.: Writing-Review and Editing, Investigation, Formal Analysis, Data Curation. L.A.: Writing-Review and Editing, Investigation. C.M.V.: Writing-Review and Editing, Supervision, Validation.R.R.: Writing-Review and Editing, Supervision, Validation. R.B.: Writing-Review and Editing, Supervision, Data curation, Conceptualization, Methodology, Validation. All authors have read and agreed to the published version of the manuscript.

Funding

This research received no external funding.

Data availability

The dataset generated during the current study is available from the corresponding author on reasonable request.

Declarations

Ethics approval and consent to participate

The current study was part of an overall larger study aimed at identifying markers, predictors and developmental trajectories of ASD. The larger overall study was approved by the local ethics committee at the University Hospital Referral Centre for ASD with number n° 759. All procedures performed in the present study were in accordance with the 1964 Declaration of Helsinki and its later amendments (2013).

Consent for publication

Not applicable.

Competing interests

The authors declare no competing interests.