Long-term psychiatric and neurodevelopmental profiles of children with global developmental delay

Abstract

BACKGROUND

Global developmental delay (GDD) is a significant neurodevelopmental condition affecting approximately 1%-3% of children worldwide. Despite its prevalence, limited data are available on the long-term psychiatric and neurodevelopmental trajectories of affected children.

AIM

To characterize the 2-year longitudinal profiles of children diagnosed with GDD, with a focus on psychiatric comorbidities and neurodevelopmental outcomes.

METHODS

This retrospective cohort study included 100 children diagnosed with GDD at the Pediatric Neurodevelopment Center of the Children’s Hospital of Shanxi Province. Participants were enrolled between January 2021 and December 2021 and followed for 24 months, with follow-up completed by December 2023. Comprehensive assessments - including the Bayley Scales of Infant and Toddler Development, Fourth Edition, and the Griffiths Mental Development Scales, Third Edition - were conducted at baseline and at 24-month follow-up. Psychiatric comorbidities were evaluated using the Diagnostic and Statistical Manual of Mental Disorders, Fifth Edition criteria and standardized diagnostic tools.

RESULTS

Among the 100 children (mean age at baseline: 36.2 ± 12.4 months, 62% male), 73% exhibited persistent developmental delays at the 2-year follow-up. Psychiatric comorbidities were present in 68% of participants, with attention-deficit/hyperactivity disorder (42%), autism spectrum disorder (31%), and anxiety disorders (24%) being the most prevalent. Children with multiple psychiatric comorbidities demonstrated significantly poorer developmental outcomes (mean developmental quotient: 62.3 ± 8.7) compared with those without comorbidities (78.4 ± 10.2, P < 0.001). Early intervention was associated with improved outcomes, particularly in language and social domains.

CONCLUSION

Children with GDD exhibit high rates of psychiatric comorbidities, which significantly affect their long-term neurodevelopmental trajectories. Early identification of comorbid conditions and the implementation of targeted intervention strategies are essential for optimizing outcomes in this vulnerable population.

Key Words: Global developmental delay; Psychiatric comorbidity; Neurodevelopmental outcomes; Longitudinal study; Early intervention

Core Tip: Global developmental delay (GDD) affects up to 3% of children and is frequently complicated by psychiatric comorbidities, which may worsen long-term outcomes. In this 2-year retrospective cohort study of 100 children with GDD, 68% developed psychiatric disorders, most commonly attention deficit/hyperactivity disorder, autism spectrum disorder, and anxiety. Children with multiple psychiatric comorbidities demonstrated significantly poorer developmental quotients compared to those without comorbidities. Early intervention, particularly in language and social domains, improved developmental outcomes. These findings highlight the importance of early psychiatric screening and integrated intervention strategies to optimize trajectories in children with GDD.

INTRODUCTION

Global developmental delay (GDD) represents a heterogeneous group of neurodevelopmental conditions characterized by significant delays in two or more developmental domains - including motor skills, language, cognition, social interaction, and adaptive functioning - in children under 5 years of age[1]. GDD affects approximately 1%-3% of children globally, with recent evidence suggesting even higher prevalence rates in resource-limited settings, reaching up to 18.8%[2]. Despite the substantial burden on affected children, their families, and healthcare systems, comprehensive longitudinal data on psychiatric and neurodevelopmental outcomes remain limited.

The etiological landscape of GDD is complex and diverse, encompassing both genetic and environmental factors. Advances in genetic testing have revealed that approximately 50% of GDD cases can be attributed to genetic causes, including chromosomal abnormalities, copy number variations, and single-gene mutations[3]. The implementation of chromosomal microarray analysis and whole-exome sequencing has significantly improved diagnostic yields, with pathogenic variants identified in 33.5% of cases in recent studies[4]. Nonetheless, a considerable proportion of children lack a definitive etiologic diagnosis, underscoring the need for comprehensive longitudinal studies to better characterize developmental trajectories and outcomes.

Recently, the relationship between GDD and psychiatric comorbidities has gained increasing attention. Children with neurodevelopmental disorders exhibit markedly higher rates of mental health conditions than their typically developing peers, with the prevalence of psychiatric comorbidities ranging from 40% to 70%[5]. Attention deficit/hyperactivity disorder (ADHD), autism spectrum disorder (ASD), anxiety disorders, and mood disorders are among the most commonly observed comorbidities[6]. These psychiatric conditions not only complicate clinical presentation but also significantly influence functional outcomes, quality of life, and treatment responsiveness.

Recent epidemiological data indicate that the prevalence of mental, behavioral, and developmental disorders among children has risen substantially over the past decade, particularly with increases in anxiety, depression, and developmental delays[7]. This trend highlights the importance of comprehensive assessment and long-term monitoring of children with GDD to identify emerging psychiatric concerns and implement timely interventions. The co-occurrence of multiple neurodevelopmental and psychiatric conditions has been linked to more severe functional impairment, greater healthcare utilization, and increased caregiver stress[8].

Developmental trajectories of children with GDD are highly variable and influenced by multiple factors, including severity of delay, presence of comorbidities, access to early intervention services, and socioeconomic contexts[9]. Longitudinal studies have demonstrated that while some children with early developmental delays achieve age-appropriate milestones with appropriate intervention, others continue to exhibit persistent deficits into school age and beyond[10]. The predictive accuracy of early developmental assessments for later cognitive and adaptive functioning remains an active area of investigation, with evidence suggesting that comprehensive evaluations conducted at 2-3 years of age may provide valuable prognostic insights[11].

Early intervention programs are critical in the management of GDD and have demonstrated effectiveness in improving developmental outcomes across multiple domains[12]. Parent-implemented early intervention programs, in particular, have shown promising results in enhancing cognitive, language, and social-emotional development in children with GDD[13]. However, the optimal timing, intensity, and specific program components remain subjects of ongoing research. Furthermore, the impact of psychiatric comorbidities on intervention responsiveness and the need for integrated mental health services within developmental intervention programs require further exploration.

The assessment and diagnosis of GDD present unique challenges that require multidisciplinary expertise and comprehensive evaluation protocols. Current guidelines recommend a tiered approach, including detailed medical history, physical examination, developmental testing, genetic evaluation, neuroimaging when indicated, and screening for metabolic disorders[14]. The Bayley Scales of Infant and Toddler Development, Fourth Edition (Bayley-IV) and the Griffiths Mental Development Scales, Third Edition (Griffiths-III) represent gold-standard tools for developmental assessment and provide detailed profiles across multiple domains[15]. However, interpretation of results must consider cultural factors, language differences, and sensory or motor impairments that may affect test performance.

Despite increasing recognition of the importance of long-term follow-up, few studies have systematically examined psychiatric and neurodevelopmental outcomes in children with GDD over extended periods. Most existing research has relied on cross-sectional designs or short-term follow-up, limiting understanding of developmental trajectories and the emergence of psychiatric comorbidities. This knowledge gap hinders the development of culturally appropriate assessment tools, intervention strategies, and health care policies for children with GDD. The present study aimed to address these gaps by providing comprehensive longitudinal data on the psychiatric and neurodevelopmental profiles of children with GDD over a 2-year follow-up period, with important implications for clinical practice, healthcare policy, and the development of targeted intervention strategies for children with GDD and their families.

MATERIALS AND METHODS

Study design and setting

This retrospective cohort study was conducted at the Children’s Hospital of Shanxi Province, which serves both urban and suburban populations. Children were enrolled between January and December 2021, with baseline assessments performed at enrollment and 24-month follow-up evaluations completed by December 2023. The study protocol was approved by the institutional review board, and all procedures adhered to the Declaration of Helsinki and applicable local ethical guidelines. Written informed consent was obtained from all parents or legal guardians prior to participation, and assent was obtained from children who were developmentally capable of providing it. The study setting was selected for its comprehensive, multidisciplinary approach to neurodevelopmental assessment and intervention. The center employs standardized protocols for developmental evaluation, diagnostic assessment, and follow-up care, ensuring consistency in data collection and clinical management throughout the study period. All assessments were conducted in child-friendly environments designed to optimize task performance and minimize anxiety, with flexibility for multiple sessions to accommodate individual needs and attention spans. The electronic health record system of the center supported systematic data collection and facilitated continuous tracking of participants throughout the study period.

Participants and eligibility criteria

The study included 100 children diagnosed with GDD according to established diagnostic criteria. Eligibility criteria were carefully defined to ensure a representative sample while maintaining diagnostic precision. Inclusion criteria were: (1) Age between 12 months and 60 months at the initial assessment, allowing for comprehensive developmental evaluation during critical early childhood periods; (2) Confirmed diagnosis of GDD, defined as significant delays of ≥ 2 standard deviations (SDs) below the mean in two or more developmental domains based on standardized assessment tools; (3) Availability of complete baseline assessment data, including developmental, medical, and family history information; (4) Commitment to attend follow-up assessments at 6, 12, 18, and 24 months post-baseline; and (5) Absence of known genetic syndromes or metabolic disorders at study entry that could confound interpretation of developmental trajectories.

Exclusion criteria were established to minimize confounding factors while maintaining external validity. Children were excluded if they had: (1) Identified genetic syndromes or chromosomal abnormalities diagnosed prior to enrollment, including Down syndrome, Fragile X syndrome, or other established genetic conditions known to affect development; (2) Severe sensory impairments (profound hearing loss or legal blindness) that would significantly compromise the validity of developmental assessments; (3) A history of acquired brain injury - such as traumatic brain injury, stroke, or central nervous system infections - occurring after the neonatal period; (4) Active medical conditions requiring intensive treatment that could affect developmental assessment, such as uncontrolled seizure disorders or ongoing chemotherapy; (5) Insufficient proficiency in the assessment language among both the child and primary caregiver to ensure valid assessment outcomes; or (6) Plans to relocate outside the hospital’s catchment area during the study period. Recruitment was conducted through multiple pathways to enhance sample representativeness. Primary referral sources included pediatric clinics, early intervention programs, child development centers, and community health services. Consecutive sampling was employed, and all eligible children who presented during the recruitment period were invited to participate. To minimize selection bias, recruitment materials were provided in multiple languages, and support for transportation to assessment appointments was provided when needed. The final cohort of 100 participants represented a diverse demographic profile, with broad representation across socioeconomic strata and geographic areas.

Developmental assessment procedures

Comprehensive developmental assessments were conducted using standardized psychometrically robust instruments administered by trained professionals with expertise in pediatric neuropsychology and developmental pediatrics. The primary assessment battery included the Bayley-IV for children under 42 months and the Griffiths-III for older participants or when more detailed domain-specific information was required. These instruments were selected based on their strong psychometric properties, extensive normative datasets, and widespread use in both clinical and research settings, thereby facilitating comparison with existing literature. The Bayley-IV assessment protocol evaluated five core developmental domains: Cognitive, language (receptive and expressive), motor (fine and gross), social-emotional, and adaptive behavior. Administration followed standardized procedures, with accommodations provided as needed for children with physical disabilities or attention difficulties. These included scheduled breaks, multiple testing sessions, and environmental modifications to optimize engagement and performance. Raw scores were converted into scaled scores, composite scores, and developmental age equivalents based on age-appropriate normative data. To ensure measurement consistency, the assessment team completed annual reliability training, maintaining inter-rater reliability > 0.90 across all domains. Quality assurance procedures included video review of 10% of assessments by senior clinicians and regular calibration meetings to resolve scoring discrepancies. The Griffiths-III provided detailed developmental profiling across six subscales: Foundations of learning, language and communication, eye and hand coordination, personal-social-emotional, gross motor, and practical reasoning. This instrument was particularly valuable for children approaching school age and for those requiring a more nuanced profiling of specific developmental domains. Administration procedures were adapted to individual needs while preserving standardization integrity. Cultural considerations were incorporated through the use of culturally appropriate materials and interpreter support when necessary. Assessments were conducted in the child’s primary language whenever possible, with bilingual assessors available for commonly spoken regional languages. Complementary assessments included the Vineland Adaptive Behavior Scales, Third Edition, administered through a semi-structured parent interview to evaluate adaptive functioning across communication, daily living skills, socialization, and motor domains. The Childhood Autism Rating Scale, Second Edition, was administered when ASD features were suspected, providing standardized observation-based ratings across 15 functional areas. Language assessments incorporated both formal standardized measures and natural language sampling, including analyses of mean length of utterance, vocabulary diversity, and pragmatic language skills. Motor assessments included standardized evaluations of fine and gross motor functioning, with specific attention to motor planning, coordination, and functional motor abilities in daily activities.

Psychiatric evaluation methods

Psychiatric comorbidities were systematically evaluated using a multi-method, multi-informant approach that incorporated clinical interviews, standardized rating scales, and observational assessments. The psychiatric evaluation protocol was designed to identify both categorical diagnoses consistent with Diagnostic and Statistical Manual of Mental Disorders, Fifth Edition criteria and dimensional symptom profiles that, while not meeting full diagnostic thresholds, could still meaningfully impact functioning. All psychiatric assessments were conducted by board-certified child and adolescent psychiatrists or licensed clinical psychologists with specialized training in developmental psychopathology.

The diagnostic interview process utilized the Mini-International Neuropsychiatric Interview for Children and Adolescents, a structured diagnostic tool covering major psychiatric disorders in children and adolescents. Interviews were administered to parents or primary caregivers, with supplementary information obtained from teachers and other relevant informants when available. Clinical interviews incorporated developmental history, family psychiatric history, psychosocial stressors, and the functional impact of symptoms across home, school, and community settings. Particular consideration was given to the developmental appropriateness of behaviors and the need to distinguish psychiatric symptoms from manifestations of developmental delay. Standardized rating scales were used to quantify symptom severity and monitor changes over time. The Conners Third Edition Parent and Teacher Rating Scales were employed to assess ADHD symptoms, associated behavioral features, and functional impairment. The Child Behavior Checklist and Teacher Report Form offered broadband assessments of emotional and behavioral problems across internalizing and externalizing domains. Anxiety symptoms were assessed using the Screen for Child Anxiety Related Disorders, with parent and child versions administered when developmentally appropriate. The Children’s Depression Rating Scale-Revised was used to assess depressive symptoms in verbal children using observational measures for nonverbal or minimally verbal participants.

Observational assessments were conducted across multiple settings and contexts to enhance the ecological validity of psychiatric diagnoses. Structured observations during developmental testing sessions provided information on attention, activity levels, emotional regulation, and social interaction patterns. Play-based assessments offered opportunities to observe symbolic plays, emotional themes, and interpersonal dynamics. When feasible, school observations were conducted to assess behavior in naturalistic educational settings. Video recordings of assessment sessions, obtained with appropriate consent, enabled detailed review and facilitated consensus diagnosis in complex cases. All psychiatric diagnoses were established through consensus conferences involving at least two clinicians. These conferences emphasized the careful differentiation of psychiatric symptoms from developmentally mediated behaviors and ensured that diagnoses were developmentally appropriate, clinically significant, and supported by convergent evidence from multiple sources.

Early intervention documentation

Documentation of early intervention services was systematically collected from multiple sources to ensure comprehensive capture of all therapeutic services received by participants. Information was obtained through parental interviews, review of medical records, coordination with early intervention providers, and examination of individualized family service plans or individualized education programs. The documentation protocol captured both the quantity and quality of intervention services, acknowledging that intervention intensity, modality, and approach may significantly influence developmental outcomes. Intervention variables documented included: (1) Types of services received - such as speech-language therapy, occupational therapy, physical therapy, special education, behavioral intervention, and developmental therapy; (2) Frequency and duration of each service, recorded as hours per week and total duration over the study period; (3) Service delivery model, distinguishing among individual therapy, group therapy, home-based services, center-based programs, and inclusive educational settings; (4) Intervention approach and theoretical framework, including applied behavior analysis, developmental approaches, sensory integration therapy, and eclectic models; (5) Parent involvement and training components, such as parent education sessions, home program implementation, and parent-mediated intervention strategies; and (6) Timing of intervention initiation relative to initial diagnosis and developmental level at the start of intervention. Quality indicators of early intervention services were assessed through review of provider qualifications, adherence to evidence-based practices, documentation of goal attainment, and coordination among service providers. When available, fidelity monitoring data - including adherence to treatment protocols and competence - were collected from intervention programs. Barriers to service access were also documented, including waitlist duration, insurance coverage limitations, geographic accessibility, and family factors that affected service utilization. Changes in intervention services throughout the study period were tracked, including additions, discontinuations, and modifications made in response to each child’s developmental progress or emerging needs.

Statistical analysis

Statistical analyses were conducted using SPSS version 28.0 and R 4.2.0. Statistical significance set at P < 0.05, with Bonferroni correction applied for multiple comparisons. Missing data were addressed using multiple imputation procedures, followed by sensitivity analyses to assess the robustness of results. Descriptive statistics included means ± SD for normally distributed variables, medians and interquartile ranges for non-normally distributed variables, and frequencies and percentages for categorical variables. Group differences were compared using t-tests and χ² tests, with Cohen’s d and Cramer’s V reported as effect sizes. Linear mixed-effects models were used to examine developmental trajectories over the 24-month follow-up period, with time specified as both a fixed and a random effect. Model predictors included psychiatric comorbidity status, intervention intensity, and their interaction, with adjustment for baseline development level, age, sex, socioeconomic status (SES), and maternal education. Logistic regression analyses were performed to identify predictors of persistent developmental delays at 24 months, incorporating baseline developmental status, psychiatric comorbidities, family history, intervention intensity, and sociodemographic characteristics. Variable selection was guided by stepwise procedures and clinical judgment. Model performance was assessed using the area under the receiver operating characteristic curve, and bootstrap validation was applied to calculate the sensitivity, specificity, and predictive values at optimal cutoff points.

RESULTS

Baseline characteristics

The study cohort comprised 100 children with confirmed GDD, with baseline assessments conducted at a mean age of 36.2 ± 12.4 months (range: 12-60 months). The sample demonstrated a male predominance, with 62 males (62%) and 38 females (38%), consistent with the sex distribution commonly reported in neurodevelopmental disorders (Table 1). SES, assessed using parental education and household income, showed a broad distribution: 32 children (32%) were from low-income households (< 30000 annually), 43 (43%) from middle-income households (30000-75000), and 25 (25%) from high-income households (> 75000). Geographic distribution indicated that 58 children (58%) resided in urban areas, 32 (32%) in suburban areas, and 10 (10%) in rural communities.

Table 1 Baseline demographic characteristics (n = 100).

Characteristic	n (%)	mean ± SD
Age at baseline, months		36.2 ± 12.4
12-24 months	24 (24)
25-36 months	28 (28)
37-48 months	32 (32)
49-60 months	16 (16)
Sex
Male	62 (62)
Female	38 (38)
Socioeconomic status
Low income (< 30000)	32 (32)
Middle income (30000-75000)	43 (43)
High income (> 75000)	25 (25)
Geographic distribution
Urban	58 (58)
Suburban	32 (32)
Rural	10 (10)

SD: Standard deviation.

Developmental profiles at baseline

Comprehensive baseline developmental assessments revealed significant delays across multiple domains (Table 2). The mean global developmental quotient (DQ) was 65.4 ± 11.3, indicating moderate to severe developmental delays within the cohort. Domain-specific analyses showed the most pronounced impairments in language development (mean ± SD: 58.7 ± 13.2), followed by the cognitive (mean ± SD: 62.3 ± 10.8), social-emotional (mean ± SD: 64.5 ± 12.1), fine motor (mean ± SD: 68.2 ± 11.6), and gross motor (mean ± SD: 71.4 ± 10.3) domains. Notably, 43% of the children demonstrated an uneven developmental profile, characterized by discrepancies greater than 15 points between their highest and lowest domain scores.

Table 2 Baseline developmental assessment scores by domain (n = 100), n (%).

Domain	DQ (mean ± SD)	Range	< 70 (delayed)	70-85 (borderline)	> 85 (average)
Global	65.4 ± 11.3	42-89	73 (73)	21 (21)	6 (6)
Cognitive	62.3 ± 10.8	40-85	81 (81)	16 (16)	3 (3)
Language	58.7 ± 13.2	35-88	86 (86)	11 (11)	3 (3)
Social-emotional	64.5 ± 12.1	38-90	76 (76)	18 (18)	6 (6)
Fine motor	68.2 ± 11.6	42-92	68 (68)	24 (24)	8 (8)
Gross motor	71.4 ± 10.3	45-95	62 (62)	28 (28)	10 (10)
Adaptive behavior	61.8 ± 12.7	36-87	79 (79)	17 (17)	4 (4)

DQ: Developmental quotient; SD: Standard deviation.

Psychiatric comorbidity prevalence

At baseline, 68% of the children met criteria for at least one psychiatric comorbidity according to the Diagnostic and Statistical Manual of Mental Disorders, Fifth Edition. Among these, ADHD was the most prevalent, affecting 42% of the cohort. The combined presentation was the most common (24%), followed by the predominantly inattentive (11%) and predominantly hyperactive-impulsive (7%) presentations. ASD was diagnosed in 31% of participants, of whom 18% met criteria for level 2 (requiring substantial support) and 13% for level 1 (requiring support). Anxiety disorders were present in 24% of the children, including specific phobia (12%), separation anxiety disorder (8%), and generalized anxiety disorder (4%). Language disorders beyond those attributable to GDD were identified in 22% of the participants. Intellectual disability was confirmed in 38% of children who had reached an age appropriate for formal intelligence quotient testing.

Longitudinal developmental outcomes

At the 24-month follow-up, 73 children (73%) continued to exhibit significant developmental delays, although the pattern and severity of these delays varied considerably across individuals (Table 3). The mean global DQ at follow-up was 70.2 ± 13.1, reflecting a statistically significant but modest improvement from baseline [mean change = 4.8 points, 95% confidence interval (CI): 3.2-6.4, P < 0.001]. Domain-specific gains were most pronounced in gross motor skills (mean change = 7.3 points) and fine motor skills (mean change = 6.1 points), whereas language development showed the least improvement (mean change = 3.2 points). Notably, 27 children (27%) demonstrated sufficient developmental progress to no longer meet criteria for GDD, although most continued to experience specific learning or developmental challenges.

Table 3 Developmental outcomes at 24-month follow-up, mean ± SD (n = 100).

Domain	Baseline	24-month	Mean change (95%CI)	P value	Effect size (d)
Global DQ	65.4 ± 11.3	70.2 ± 13.1	4.8 (3.2-6.4)	< 0.001	0.39
Cognitive	62.3 ± 10.8	66.8 ± 12.4	4.5 (2.9-6.1)	< 0.001	0.39
Language	58.7 ± 13.2	61.9 ± 14.6	3.2 (1.8-4.6)	< 0.001	0.23
Social-emotional	64.5 ± 12.1	69.3 ± 13.8	4.8 (3.1-6.5)	< 0.001	0.37
Fine motor	68.2 ± 11.6	74.3 ± 12.9	6.1 (4.3-7.9)	< 0.001	0.50
Gross motor	71.4 ± 10.3	78.7 ± 11.2	7.3 (5.4-9.2)	< 0.001	0.68
Adaptive behavior	61.8 ± 12.7	67.2 ± 14.1	5.4 (3.7-7.1)	< 0.001	0.40

DQ: Developmental quotient; SD: Standard deviation; CI: Confidence interval; d: Cohen’s d.

Evolution of psychiatric comorbidities

The prevalence of psychiatric comorbidities increased over the follow-up period, with 76% of the children meeting criteria for at least one psychiatric disorder at 24 months, compared with 68% at baseline. New psychiatric diagnoses emerged in six children (18% of the 32 children without identified comorbidities at baseline). The most notable increases were observed for anxiety disorders (from 24% to 32%) and depressive symptoms (from 3% to 8%). The prevalence of ADHD remained relatively stable (42%-44%), although a shift toward more severe presentations was noted. The persistence of psychiatric diagnoses was high, with 89% of children who had a comorbidity at baseline continuing to meet diagnostic criteria at follow-up. Multiple psychiatric comorbidities were present in 38% of children at follow-up and were associated with poorer functional outcomes across all developmental domains.

Impact of psychiatric comorbidities on developmental trajectories

Children with psychiatric comorbidities demonstrated significantly different developmental trajectories compared with those without comorbidities (Table 4). Linear mixed-effects modeling revealed a significant interaction between psychiatric comorbidity status and time [F (2194) = 18.67, P < 0.001], indicating divergent developmental pathways. Children without psychiatric comorbidities showed greater developmental gains (mean ± SD: 8.2 ± 4.3 points) than those with a single comorbidity (mean ± SD: 4.6 ± 3. 8 points) or multiple comorbidities (mean ± SD: 2.1 ± 3.2 points). The adverse effects of psychiatric comorbidities were most pronounced in the social-emotional and language domains, with minimal impact on motor development.

Table 4 Developmental outcomes by psychiatric comorbidity status.

Comorbidity group	n	Baseline DQ (mean ± SD)	24-month DQ (mean ± SD)	mean ± SD	F-statistic	P value
No comorbidity	32	71.2 ± 9.8	79.4 ± 10.3	8.2 ± 4.3	18.67	< 0.001
Single comorbidity	30	65.8 ± 10.4	70.4 ± 11.2	4.6 ± 3.8
Multiple comorbidities	38	60.3 ± 11.2	62.4 ± 13.1	2.1 ± 3.2

DQ: Developmental quotient; SD: Standard deviation.

Early intervention effects

Participation in early intervention was significantly associated with developmental outcomes (Table 5). Children who received intensive early intervention (> 10 hours per week) demonstrated greater developmental gains (mean ± SD: 7.2 ± 4.1 points) than those who received standard intervention (5-10 hours per week; mean ± SD: 4.3 ± 3.6 points) or minimal intervention (< 5 hours per week; mean ± SD: 2.8 ± 3.1 points). The effect of intervention intensity was moderated by baseline developmental level and the presence of psychiatric comorbidities. In multivariate analysis, after controlling for baseline factors, each additional hour of weekly intervention was associated with a 0.38-point increase in DQ at follow-up (95%CI: 0.21-0.55, P < 0.001). Parent-implemented intervention components showed particularly strong benefits for language and social-emotional development.

Table 5 Developmental outcomes by early intervention intensity (n = 100).

Intervention intensity	n	Baseline DQ (mean ± SD)	24-month DQ (mean ± SD)	mean ± SD	P value
Intensive (> 10 hours/week)	34	64.8 ± 11.2	72.0 ± 12.8	7.2 ± 4.1	< 0.001
Standard (5-10 hours/week)	42	65.6 ± 10.9	69.9 ± 13.2	4.3 ± 3.6
Minimal (< 5 hours/week)	24	66.2 ± 12.1	69.0 ± 13.6	2.8 ± 3.1

DQ: Developmental quotient; SD: Standard deviation.

Predictors of persistent developmental delay

Logistic regression analysis identified several baseline factors that were significantly associated with persistent developmental delay at the 24-month follow-up (Table 6). The strongest predictors included a baseline global DQ < 55 [odds ratio (OR) = 4.82, 95%CI: 2.13-10.91], a diagnosis of ASD (OR = 3.67, 95%CI: 1.68-8.02), the presence of multiple psychiatric comorbidities (OR = 3.21, 95%CI: 1.44-7.15), and limited access to early intervention services (OR = 2.89, 95%CI: 1.31-6.38). Protective factors included higher maternal education (OR = 0.68 per additional year of education, 95%CI: 0.52-0.89), initiation of early intervention before 24 months of age (OR = 0.45, 95%CI: 0.23-0.88), and the absence of a language disorder (OR = 0.51, 95%CI: 0.27-0.96). The final model demonstrated good predictive performance, with an area under the receiver operating characteristic curve of 0.81 (95%CI: 0.74-0.88).

Table 6 Predictors of persistent developmental delay at 24-month follow-up.

Predictor variable	Odds ratio	95%CI	P value
Baseline DQ < 55	4.82	2.13-10.91	< 0.001
Autism spectrum disorder	3.67	1.68-8.02	0.001
Multiple psychiatric comorbidities	3.21	1.44-7.15	0.004
Limited early intervention (< 5 hours/week)	2.89	1.31-6.38	0.008
Male sex	1.92	0.89-4.14	0.095
Low SES	1.78	0.82-3.86	0.144
Maternal education (per year)	0.68	0.52-0.89	0.005
Early intervention before 24 months	0.45	0.23-0.88	0.020
Absence of language disorder	0.51	0.27-0.96	0.037

CI: Confidence interval; DQ: Developmental quotient; SES: Socioeconomic status.

Subgroup analyses

Subgroup analyses revealed significant heterogeneity in outcomes based on specific psychiatric comorbidity profiles. Children with ASD and comorbid ADHD demonstrated the poorest outcomes (mean ± SD: 58.3 ± 9.7), whereas those with ADHD alone showed more favorable developmental trajectories (mean ± SD: 72.4 ± 11.2). Sex differences were also observed: Females exhibited greater gains in language development (5.1 points vs 2.6 points, P = 0.023) despite having similar baseline scores. Age at diagnosis significantly influenced outcomes, with children diagnosed before 24 months demonstrating a stronger response to intervention (mean change = 6.8 points vs 3.9 points, P = 0.008). Socioeconomic factors played a substantial role in service utilization, as high-SES families accessed 2.3 times more intervention hours than low-SES families, partially mediating the association between SES and outcomes.

DISCUSSION

This comprehensive 2-year longitudinal study provides critical insights into the psychiatric and neurodevelopmental trajectories of children with GDD, revealing complex patterns of comorbidity, variable developmental outcomes, and differential responses to intervention. The high prevalence of psychiatric comorbidities observed in our cohort - affecting 68% of participants at baseline and increasing to 76% at follow-up - underscores the neurobiological overlap between GDD and co-occurring mental health conditions. This finding aligns with emerging conceptualizations of neurodevelopmental disorders as interconnected rather than discrete entities[16]. Our results are consistent with those of recent meta-analyses demonstrating that children with developmental delays face substantially elevated risks of psychiatric disorders throughout childhood and adolescence, emphasizing the need for enhanced screening protocols and integrated service delivery models[17].

The predominance of ADHD as the most common psychiatric comorbidity (42% of the cohort) reflects broader epidemiological patterns and highlights the shared neurobiological substrates underlying attentional regulation and global development. The relative stability of ADHD diagnoses over the 2-year follow-up period, accompanied by a shift toward more severe presentations, suggests that attentional difficulties in children with GDD may represent a persistent neurodevelopmental phenotype rather than a transient developmental feature. This finding has important clinical implications because early identification and treatment of ADHD symptoms in children with developmental delays have been shown to improve attention, behavior, and broader developmental outcomes[18]. The high rate of co-occurrence with ASD further emphasizes the importance of comprehensive assessment protocols capable of identifying multiple overlapping conditions and disentangling their specific contributions to functional impairment.

The modest but statistically significant improvements in DQ observed over the study period (mean gain of 4.8 points) reflect both the chronic nature of GDD and the potential for developmental progress with appropriate support. Differential improvement across domains, with motor skills showing the greatest gains and language development lagging, suggests that intervention effects may be domain-specific, with certain developmental areas being more amenable to environmental modification. This pattern is consistent with neuroplasticity research indicating that motor systems may retain broader windows of plasticity than language networks, particularly after the critical period for language acquisition[19]. The persistence of developmental delays in 73% of the cohort at follow-up, despite ongoing intervention, highlights the need for realistic expectations and long-term support planning for families.

The emergence of new psychiatric diagnoses among previously unaffected children and the increasing prevalence of anxiety and depressive symptoms underscore the dynamic nature of psychiatric comorbidities in children with GDD. These trends may reflect the developmental unfolding of genetic vulnerabilities, increased environmental demands, and the cumulative emotional toll of chronic functional impairment. The particularly notable rise in anxiety disorders - from 24% to 32% - may be attributable to increased awareness of developmental differences, growing academic and social expectations, and stress associated with chronic developmental challenges. These findings support the need for ongoing psychiatric monitoring and preventive mental health interventions aimed at children with GDD who are at elevated risk for emerging psychopathology[20].

One of the most clinically significant findings is the strong interaction between psychiatric comorbidity burden and developmental trajectory. Children with multiple psychiatric comorbidities exhibited markedly attenuated developmental gains compared to those without comorbidities, suggesting that psychiatric symptoms may impair learning processes, reduce intervention engagement, and impede the neuroplastic mechanisms underlying developmental progress. The dose-response relationship observed between comorbidity burden and outcome severity supports a cumulative risk model in which each additional psychiatric condition compounds functional impairment and limits compensatory mechanisms. The particularly strong negative impact on social-emotional and language development likely reflects the interconnected nature of psychiatric functioning with emotional regulation and communication skills, as emotional dysregulation and social challenges can significantly impair language acquisition and social communication[21].

The demonstrated effectiveness of intensive early intervention, particularly when initiated before 24 months of age, offers empirical support for the current early identification and intervention policies. The dose-response relationship between intervention intensity and developmental gains - with each additional hour of weekly intervention contributing meaningfully to DQ improvement - suggests that many children with GDD may require more intensive services than are currently available. The observed benefits of parent-implemented intervention components for language and social-emotional development align with attachment theory and research emphasizing the central role of parent-child interaction quality in shaping developmental outcomes[22]. These results support the expansion of parental training and coaching models within early intervention systems as a cost-effective approach to enhance intervention intensity.

The identification of specific predictors of persistent developmental delay offers valuable prognostic insights for clinicians and families. The strong predictive value of a baseline DQ below 55 highlights the severity of initial presentation as the most robust indicator of long-term outcomes, consistent with prior longitudinal studies of intellectual disability[23]. However, the modifiable nature of several identified risk and protective factors, including early intervention access and timing, offers opportunities for targeted interventions. The protective effect of maternal education, independent of SES, may reflect an enhanced capacity to navigate service systems, implement home-based strategies, and advocate for children’s needs, highlighting the potential value of caregiver education and empowerment programs[24].

The observed sex differences in developmental trajectories - specifically, greater language gains in females despite similar baseline scores - warrant further investigation into potential sex-specific mechanisms and intervention approaches. These differences may reflect known sex-based variations in language development trajectories, the differential effects of sex hormones on neuroplasticity, or sociocultural influences on engagement and responsiveness to intervention. The superior outcomes among children diagnosed before 24 months of age highlight the critical importance of early detection and timely intervention and support recent initiatives to lower the age for developmental screening and expand universal screening protocols[25].

This study has some limitations that should be considered when interpreting these findings. The retrospective design, while allowing for naturalistic observation, limits causal inference regarding intervention effects and may introduce selection bias. The 24-month follow-up period provides valuable short-term insights but cannot capture long-term developmental trajectories into school age and adolescence. Reliance on clinical diagnoses and standardized assessments, although enhancing diagnostic validity, may not fully capture functional impairments and quality-of-life outcomes, which are increasingly recognized as essential metrics. Finally, although the cohort was diverse, the single-center design may limit generalizability to other regions or healthcare systems with differing service delivery models.

CONCLUSION

Children with GDD frequently experience psychiatric comorbidities such as ADHD, autism, and anxiety, all of which can significantly exacerbate developmental challenges over time. Early intensive intervention, particularly before the age of 2 years, improves outcomes, but access to these services remains uneven due to socioeconomic barriers. To optimize developmental trajectories for this vulnerable population, healthcare systems must adopt integrated care models that address both developmental and mental health needs concurrently, while ensuring equitable access to early intervention services for families, regardless of economic status.