Comparative content validity of the Thai version of the Modified Checklist for Autism in Toddlers, Revised with Follow-Up (M-CHAT-R/F) and Thai Diagnostic Autism Scale (TDAS): enhancing early detection of autism spectrum disorder in high-risk children

Introduction

Autism spectrum disorder (ASD) is a neurodevelopmental (1) condition marked by persistent deficits in social communication and interaction, alongside restricted and repetitive patterns of behavior, interests, or activities. The global prevalence of ASD has shown a notable increase, with recent estimates indicating that approximately 1 in 44 children in the United States (2) are diagnosed with the condition. In Thailand, although comprehensive nationwide prevalence data remain limited, studies conducted in tertiary healthcare settings have reported a rise in ASD diagnoses (3), from 1 in 1,000 children in 2004 to 6 in 1,000 in 2015. Early identification and timely intervention are critical for improving long-term outcomes (4,5) for children with ASD. A substantial body of research has demonstrated that early, intensive behavioral interventions—particularly when implemented in early childhood—can significantly enhance social communication, adaptive behavior, and overall developmental outcomes (6,7) in children with ASD. Furthermore, early intervention has been shown to alleviate parental stress and improve the quality of life (8) for families as a whole. Despite the clear benefits of early diagnosis, significant challenges persist in the timely identification of ASD, particularly in resource-limited healthcare systems in developing countries (9). Barriers to early diagnosis (10,11) include a shortage of healthcare professionals with expertise in child development, cultural and linguistic differences, and inadequate access to healthcare services and appropriate diagnostic tools.

Currently, there are several internationally recognized tools such as Autism Diagnostic Observation Schedule (ADOS) and the Autism Diagnostic Interview-Revised (ADI-R) have been used to diagnose autism (9,10). However, their implementation in Thailand poses significant challenges. These tools are resource-intensive, require extensive clinician training, and necessitate sustained availability of licensed professionals. These limitations significantly constrain their accessibility in resource-limited Thai primary and secondary care settings (9,11).

The Modified Checklist for Autism in Toddlers, Revised with Follow-Up (M-CHAT-R/F), is one of the most widely utilized screening instruments (12) for ASD in children aged 16 to 30 months. This tool comprises a 20-item questionnaire, followed by a structured interview for cases where initial screening results are positive. The M-CHAT-R/F has been translated and culturally adapted for use in various countries/regions, with its validity assessed in diverse settings including Spain (13), Japan (14), Singapore (15), and Taiwan (16). While most studies have reported acceptable sensitivity and specificity, many have been conducted in small, specialized populations, limiting the generalizability of findings to broader, community-based or primary care settings.

In Thailand, the M-CHAT-R/F was translated into Thai (17) by Chaiudomsom et al., and its validity was preliminarily evaluated through a pilot study using a mobile application (18) format. However, the pilot study’s small sample size (n=30) and the lack of comparison with a standard diagnostic tool widely used in Thailand limited the robustness of its findings.

To address the need for culturally appropriate ASD screening tools, the Thai Diagnostic Autism Scale (TDAS) was developed in 2017 by Department of Mental Health (19) as a culturally and linguistically adapted tool based on Diagnostic and Statistical Manual of Mental Disorders, Fifth Edition (DSM-5) criteria (5). While early validation studies show high sensitivity (100%) and specificity (94.1%), the tool is diagnostic in nature and not optimized for population-wide or high-risk screening. It requires administration by specialists and is time-intensive (30–45 minutes), limiting its scalability in routine well-child visits. It includes both a parent interview and direct behavioral observation, together with the structure and purpose of multi-method tools. However, ADOS and ADI-R were not directly applied in this study due to two primary reasons: (I) uneasy access: in Thailand, ADOS is rarely used outside specialized tertiary centers due to high costs, the need for licensed administrators, and limited local availability of training programs; (II) government support: TDAS is the nationally endorsed diagnostic tool by the Thai Ministry of Public Health and is actively used by developmental pediatricians nationwide. Thus, using TDAS as the reference standard reflects current national practice and allows the findings to be directly translated into policy and clinical pathways. Though using TDAS as the reference standard may limit cross-national comparability. However, within Thailand’s healthcare framework, TDAS offers cultural validity and practical applicability, making it the most appropriate benchmark for validating a Thai M-CHAT-R/F.

Given the limited research on the performance of the Thai M-CHAT-R/F and its potential as a screening tool for primary and secondary healthcare settings, there is a clear need for further validation studies. This research seeks to compare the validity of the Thai M-CHAT-R/F with the TDAS in high-risk children with language delays. The findings from this study will provide critical insights into the feasibility of implementing the M-CHAT-R/F within the Thai healthcare system, contributing to improved early detection and intervention for children with ASD in Thailand.

Objective

To evaluate the validity, sensitivity, and specificity of the Thai version of M-CHAT-R/F as a screening tool for ASD by comparing its performance with the established diagnostic standard, the TDAS. We present this article in accordance with the STARD reporting checklist (available at https://pm.amegroups.com/article/view/10.21037/pm-25-4/rc).

Methods

Study design and participants

This prospective, cross-sectional study was conducted at Queen Sirikit National Institute of Child Health (QSNICH), the only government children’s hospital that provides primary–tertiary cares for children across Thailand. The samples were children who visited the Division of Developmental and Behavioral Pediatrics at QSNICH from January to December, 2022. The sample size was calculated using Cohen’s kappa agreement statistic, which is appropriate for validating agreement between two categorical classification tools—M-CHAT-R/F and TDAS. Based on a hypothesized kappa value of 0.8 (substantial agreement), with a null hypothesis of kappa =0.6, an alpha level of 0.05, and a power of 0.8, a minimum of 196 participants was required. We enrolled 230 participants aged 16–30 months with language delays were identified through routine developmental screening using (20) the Developmental Surveillance and Promotion Manual (DSPM) or the Developmental Assessment for Intervention Manual (DAIM). Exclusion criteria included: inability of parents or legal guardians to communicate in Thai, presence of active seizures, cerebral palsy, genetic disorders, or visual or hearing impairments. Anticipating a 5% (22 participants) was excluded from the study.

This method was selected as the study focused on agreement and diagnostic concordance, not differences in group means, which would typically require sample size calculations based on t-tests or Chi-squared analysis.

Study procedures

After obtaining informed consent, demographic data and relevant medical history using a standardized questionnaire, all participants were then screened using the Thai M-CHAT-R/F and were blinded to the evaluations and index tests. The screening was administered by trained research assistants. There were three research assistants who underwent structured training and pilot testing to ensure standardized delivery. The diagnostic assessment using TDAS was conducted independently by a board-certified developmental-behavioral pediatrician with over 10 years of experience in ASD diagnosis, ensuring consistency and clinical rigor. All participants, regardless of their M-CHAT-R/F results, were evaluated using the TDAS by a developmental pediatrician who was blinded to the M-CHAT-R/F results to prevent evaluation bias. If the data is incomplete, the participant will be contacted to complete the evaluations within 1–2 weeks.

Measures

• Thai M-CHAT-R/F (17) is a two-stage screening tool for ASD in children aged 16–30 months. It consists of a 20-item parent-report questionnaire (M-CHAT-R) and a follow-up interview (M-CHAT-R/F) for children who screen positive on the initial questionnaire.
• The TDAS is a diagnostic tool developed specifically for Thai children, based on DSM-5 criteria for ASD. It comprises two parts: a 13-item observational assessment of communication, social interaction, play, and repetitive behaviors, and a 17-item parent interview on the child’s development and behavior. The TDAS takes approximately 45–60 minutes to administer and has shown high sensitivity (100%) and specificity (94.1%) in previous studies (19). TDAS classification follows established guidelines: a total score ≥19 suggests ASD, while scores <19 indicate non-ASD diagnoses. This threshold has been validated in previous Thai studies demonstrating high sensitivity and specificity.

Statistical analysis

All the collected data for TDAS and M-CHAT-R/F assessments were from 208 participants. The data were analyzed using SPSS version 26.0. Descriptive statistics were used to summarize participant characteristics. The sensitivity, specificity, positive predictive value (PPV), and negative predictive value (NPV) of the Thai M-CHAT-R/F were calculated using the TDAS as the reference standard. Receiver operating characteristic (ROC) curve analysis was performed to determine the optimal cut-off score for the Thai M-CHAT-R/F. The statistical analyses used two-sided tests, with a P value <0.05 considered statistically significant. The agreement between M-CHAT-R/F and TDAS classifications was substantial, with a Cohen’s kappa coefficient of 0.849.

Ethical considerations

This study was conducted in accordance with the Declaration of Helsinki and its subsequent amendments. The study was approved by the institutional ethics committee of Queen Sirikit National Institute of Child Health dated August 21, 2023 (No. IRB00007346), and informed consent was taken from the guardians of all individual participants.

Results

Participant characteristics

Figure 1 represents the participants flow. A total of 230 children were enrolled through a convenience sample and went through several assessments. Children with no suspected language developmental delays of 22 persons were removed from the project, only 208 children participated in the study. The mean age was 24.91±4.01 months, with 135 (64.9%) males and 73 (35.1%) females, indicating a male-to-female ratio of approximately 1.8:1. Based on the TDAS evaluation, 87 children (41.8%) were diagnosed with ASD, while 121 (58.2%) received other diagnoses, including language delay (n=85) and global developmental delay (n=36) as shown in Table 1.

Figure 1 Flow of participants. DAIM, Developmental Assessment for Intervention Manual; DSPM, Developmental Surveillance and Promotion Manual; M-CHAT-R/F, Modified Checklist for Autism in Toddlers, Revised with Follow-Up; QSNICH, Queen Sirikit National Institute of Child Health; TDAS, Thai Diagnostic Autism Scale.

Children diagnosed with ASD had a significantly higher mean TDAS score (35.35±6.87) compared to those with other diagnoses (9.42±4.77). This highlights the TDAS’s effectiveness in differentiating ASD from other developmental disorders. This difference in scores reflects the severity of developmental symptoms in children with ASD, as TDAS is designed to assess behaviors associated with ASD. These findings underscore the use of the TDAS as a tool for supporting early and accurate diagnosis.

Table 2 provides a comprehensive overview of the demographic and clinical characteristics of 208 participants, divided into two diagnostic groups: children with ASD and those with other conditions (language and developmental delay). Both groups engaged in regular developmental activities to a similar extent, indicating that such activities are commonly implemented regardless of the child’s specific diagnosis. Key areas of significant difference between both groups include age and caregiver perceptions of developmental delays. Children with autism became slightly older and more likely to have their developmental delays detected early by their caregivers. Individual characteristics such as sex, family income, caregiver education, and screen time did not show significant differences between the groups, indicating that these factors are not major distinguishing variables in this study population.

The significant difference in caregiver awareness of developmental delays among children with ASD indicates the critical role of caregivers in early detection, suggesting that caregiver-reported concerns should be taken seriously in early screening processes. The older age of children in the ASD group demonstrates potential delays in diagnosis, emphasizing the need for earlier screening and timely intervention, particularly in high-risk groups.

Table 3 presents M-CHAT-R/F scores and corresponding risk levels, separated by diagnostic group. The table compares children diagnosed with ASD to those with other developmental conditions. The table highlights the distribution of participants across risk categories and provides key statistical comparisons between groups. In the low-risk group (scores 0–2), no children from the autism group were classified as low risk, while 25.8% of children with other conditions fell into this category. The P value =0.001 highlights a statistically significant difference, suggesting that M-CHAT-R/F effectively identifies children with ASD as moderate to high risk, underscoring its strong screening capabilities. For the medium-risk group (scores 3–7), only 8.7% of children with ASD were classified in this category, compared to 64.5% of children with other diagnoses. While the exact P value is not provided, the substantial difference suggests statistical significance, reflecting that many children with developmental delays but not ASD tend to fall into this medium-risk range. In the high-risk group (scores 8–20), 91.3% of children with ASD were classified as high risk, compared to only 9.7% of those with other diagnoses. This stark difference likely holds statistical significance, emphasizing the tool’s specificity for identifying ASD in high-risk populations. Additionally, the mean M-CHAT-R/F score was significantly higher for children with ASD (11.22±2.81) compared to those with other diagnoses (4.61±2.66), with a P value of 0.001, further validating the tool’s accuracy in distinguishing ASD from other developmental issues.

Table 4 shows a diagnostic performance of M-CHAT-R/F at different cut-off scores. A cut-off score of 8 offers an optimal balance between sensitivity (91.3%) and specificity (93.5%). Lowering the cut-off score increases sensitivity but reduces specificity, leading to a higher number of false positives as shown in Table 5.

Conversely, raising the cut-off score improves specificity but reduces sensitivity, potentially missing true ASD cases. At a cut-off score of 8, both PPV and NPV are balanced and high, suggesting that this cut-off score is effective in minimizing false positives and negatives, making it a robust threshold for clinical screening. Positive likelihood ratio (LR+): a value of 14.2 at the cut-off score of 8 suggests that children with a positive screen are 14.2 times more likely to have ASD than not. Negative likelihood ratio (LR−): a low value of 0.09 at a cut-off score of 8 indicates a low likelihood of having ASD if the test result is negative. The high LR+ and low LR− values at the cut-off score of 8 further support its use as the optimal threshold for ASD screening. Youden’s index: this index measures the effectiveness of a diagnostic test and is calculated as sensitivity + specificity − 1. The highest Youden’s index (0.849) occurs at the cut-off score of 8, indicating the best balance between sensitivity and specificity at this threshold. A Youden’s index of 0.849 at the cut-off score of 8 confirms that this is the most efficient score for balancing sensitivity and specificity, making it the optimal choice for clinical application.

Figure 2 shows ROC curve graph comparing the diagnostic accuracy of the M-CHAT-R and M-CHAT-R/F tools for ASD screening. The M-CHAT-R has an area under the curve (AUC) of 0.816 (95% CI: 0.784–0.995), indicating good diagnostic performance, with the ability to accurately distinguish ASD cases 81.6% of the time. In contrast, the M-CHAT-R/F shows superior performance with an AUC of 0.849 (95% CI: 0.886–1), reflecting higher diagnostic accuracy (AUC =0.849) than M-CHAT-R (AUC =0.816), and reduces false results. Thus, M-CHAT-R/F is preferred for pediatricians aiming for early ASD detection and intervention.

Figure 2 ROC curves for M-CHAT-R and M-CHAT-R/F. AUC, area under the curve; CI, confidence interval; M-CHAT-R, Modified Checklist for Autism in Toddlers Revised; M-CHAT-R/F, Modified Checklist for Autism in Toddlers, Revised with Follow-Up; ROC, receiver operating characteristic.

Discussion

This study assessed the validity of the Thai M-CHAT-R/F for children with language delays who are high risk of having ASD, demonstrating high sensitivity (91.3%) and specificity (93.5%) at the optimal cut-off score of 8, which achieved the highest Youden’s index (0.849). This finding is consistent with the original M-CHAT-R/F study by Robins et al. (12) recommendations, which proposed a score of 8 or higher signifies a high risk for ASD, necessitating referral for diagnostic assessment. The results in this study based on a high-risk cohort of children with language delays and a substantial sample size (n=208), supporting the effectiveness of this tool within the Thai context.

The slight difference in cut-off scores between studies may be attributed to variations in populations and methodologies. The prevalence of ASD (41.8%) was observed in the study population of high-risk children with language delays (20).

The mean age of children who went through early ASD diagnosis (26.3±2.96 months) was lower than many countries (21,22), enabling timely intervention, which has been shown to improve long-term outcomes (6-8). This study found a significant correlation between parental perception of developmental delays and ASD diagnosis. This highlights the importance of considering parental concerns in the screening process, as parents are often the first to notice developmental differences (23) in their children. The strong agreement between the Thai M-CHAT-R/F and TDAS (kappa =0.849) validated the screening tool within the Thai context. This is particularly important given that the TDAS was developed specifically for the Thai population, taking into account cultural and linguistic factors that may influence ASD presentation and detection. However, in Thailand, many caregivers—particularly in rural areas—may lack awareness of developmental milestones and view certain autistic behaviors (e.g., limited eye contact or delayed speech) as personality traits rather than developmental concerns. Parental education has been shown to influence help-seeking behaviors and engagement in early screening processes (24,25).

Since this study focused on high-risk population, it reflects clinical realities and resource efficiency than universal screening (24) and the use of the TDAS (19) as a reference enhanced diagnostic accuracy and local validity in Thai context. Moreover, a large sample size (n=208) supported robust statistical analysis. However, several limitations should be considered when interpreting the results. First, a tertiary care setting which limit the generalizability to primary care and community care. Second, a larger and more diverse population are needed to assess subgroup variations such as age, gender or socioeconomic status (SES). Such analyses could provide valuable insights for tailoring screening strategies to specific populations. Third, cultural and linguistic factors such as parental attitudes, deference to authority, may affect caregiver reporting and interpretation, and the detection of ASD symptoms (25). For example, Thai caregivers often exhibit high levels of deference to medical authority, which can lead to underreporting of behavioral concerns unless directly prompted. Additionally, there is a cultural tendency to interpret communication delays as “slow development” rather than signs of neurodevelopmental disorder. In many Thai communities, especially in rural areas, behaviors such as lack of eye contact or limited verbal expression are not immediately perceived as abnormal. These cultural perceptions may influence both caregiver responses and clinician interpretation. The TDAS was developed in direct response to these contextual nuances, incorporating culturally relevant behavior descriptions and parental interview items tailored to Thai familial dynamics. These cultural adaptations likely enhance the validity of TDAS within the Thai context. However, further research is needed to explore how cultural factors may affect the performance of both the M-CHAT-R/F and TDAS in different regions of Thailand. Lastly, the screening tools are not diagnostic, so positive results require comprehensive follow-up and evaluations to confirm diagnosis and create intervention plan.

Thai M-CHAT-R/F is suitable for targeted screening in high-risk groups and potentially scalable to primary or secondary care with proper support. Its ease of use, relatively short administration time (10–15 minutes), and strong psychometric properties make it an attractive option for widespread screening. However, successful implementation would require adequate training of healthcare providers, development of efficient referral systems, and ensuring access to appropriate diagnostic and intervention services.

The optimal cut-off score of 8 using TDAS aligns with original M-CHAT-R/F (12) and outperforms other international versions in some aspects. A study from Chaudhomsom et al. (17) reported an optimal cut-off of 6 using DSM-5, a discrepancy likely attributable to differences in study populations (ASD prevalence of 13.3%) and sample size (n=30). The higher cut-off score of 8 identified in our study may help mitigate false positives, which is particularly crucial in high-risk populations to prevent overburdening limited diagnostic resources. At this cut-off, the PPV was 91.3%, and the NPV was 93.5%, both notably higher than those reported in other studies. Adjusting the cut-off to 7 would enhance sensitivity to 95.7% but reduce specificity to 83.9%, whereas increasing it to 9 would improve specificity to 96.8% but reduce sensitivity to 82.6%. The selection of an appropriate cut-off should be guided by the specific objectives and available resources of the screening program. Previous studies in low- and middle-income countries have demonstrated that the M-CHAT-R/F can be effectively administered by non-specialists in routine care settings, provided they receive appropriate training and support (11).

Given the tool’s simplicity, brief administration time, and cultural acceptability in our cohort, we propose that its implementation in lower-tier Thai health services is both feasible and warrants further exploration.

Screening for ASD involves complexities in identifying the optimal target population and timing. Two main approaches are debated: universal and targeted screening. The American Academy of Pediatrics (AAP) recommends universal screening for all children at 18 and 24 months using standardized instruments (4) like the M-CHAT-R/F. Universal screening aims for early detection and timely intervention, improving long-term developmental outcomes (26). Its benefits include early identification (27), reduced disparities in diagnosis and intervention, and increased awareness (28) of ASD. However, challenges include high false positive (29,30) rates, significant resource demands (31), and potential over-diagnosis.

Targeted screening focuses on high-risk groups, such as children with delayed language development. This approach allows for more efficient use of resources (32,33) and may reduce unnecessary anxiety, making it suitable for settings with limited healthcare infrastructure. However, it risks missing diagnoses in children without apparent risk factors or those with subtle symptoms, potentially delaying intervention. Evidence from multiple studies (32-35) indicates universal screening may be cost-effective in the long term by reducing lifetime care costs for individuals with ASD. However, these findings may not apply universally due to variations in healthcare systems, cost structures, and resource availability.

The feasibility of integrating the Thai M-CHAT-R/F into routine clinical practice is crucial for its widespread use in Thailand’s healthcare system. Several factors support its viability. The M-CHAT-R/F is time-efficient (5–10 minutes for the questionnaire and 10–15 minutes for the follow-up interview), making it ideal for busy primary care settings with limited time (36) for developmental screening. Its simple questionnaire format enables healthcare professionals, including nurses and community health workers, to administer it after proper training, even in areas with limited access to specialists (37). The tool also requires no specialized equipment, making it cost-effective and easily implementable in both urban and rural settings (38). Our findings show that the Thai version of the M-CHAT-R/F is culturally acceptable to both parents and healthcare providers, which is a key factor (39) for successful adoption.

The M-CHAT-R/F can be integrated into existing screening programs like the DSPM or the DAIM, providing a more comprehensive approach without adding significant burden to the healthcare system. Effective implementation will require proper training to ensure accurate administration (27) and interpretation of the tool. Additionally, access to diagnostic and intervention services for children who screen positive is essential, with adequate follow-up resources, particularly in rural or underserved areas, posing a potential challenge requiring strategic planning and resource allocation (29). Adaptations of M-CHAT-R/F are needed to address Thailand’s cultural and ethnic (9,31) diversity.

The high validity of the Thai M-CHAT-R/F in high-risk populations suggests it could be an effective tool for targeted ASD screening in resource-limited settings. Future research should focus on integrating this tool into existing screening programs, improving healthcare provider training, and exploring parental education strategies (27,28), while assessing the healthcare system’s capacity to provide timely diagnostic evaluations and early interventions is also important for enhancing screening (29,31) outcomes.

Policymakers should consider the balance between early detection and resource management when deciding on cut-off scores and implementation scope. For pediatricians, using a cut-off score of 8 can improve diagnostic accuracy when screening children for ASD, ensuring that fewer cases are missed while avoiding over-referrals due to false positives. Strategic investments are needed to ensure rural and underserved communities can benefit from early ASD detection.

Conclusions

Thai M-CHAT-R/F is a valid and reliable screening tool for ASD in high-risk Thai children with language delays. Its high sensitivity and specificity, cut-off score of 8 coupled with its ease of use, make it a promising instrument for improving early detection of ASD in Thailand. As the results of this study demonstrate excellent psychometric properties of the Thai M-CHAT-R/F among high-risk children in a specialized setting, its application in broader healthcare contexts remains a recommendation, not a direct conclusion. The purpose of this study was to highlight the potential for broader use based on its ease of administration, brief completion time, and cultural acceptability—not to imply proven generalizability. Implementation of the Thai M-CHAT-R/F in primary and secondary healthcare settings could facilitate earlier diagnosis and intervention for children with ASD, potentially leading to improve developmental outcomes and quality of life for affected children and their families.

Acknowledgments

We extend our sincere gratitude to the following individuals and institutions for their invaluable support. We deeply appreciate Dr. Diana Robins, Director of the AJ Drexel Autism Institute, USA, for her contributions. We are also grateful to Assoc. Prof. Dr. Kusonlaporn Chaiyudom and her team from the Department of Psychiatry, Faculty of Medicine, Khon Kaen University, for permission to use the Thai version of the Modified Checklist for Autism in Toddlers, Revised, and Follow-up (M-CHAT-R/F). Our appreciation also extends to Dr. Duangkamol Tangviriyapaiboon, Deputy Medical Director of the Rajanagarindra Institute of Child Development, for authorizing the use of the Thai Diagnostic Autism Scale (TDAS). We especially acknowledge the Department of Medical Services for funding this research. The contributions of all mentioned were indispensable to the success of this work, for which we are profoundly grateful.

Footnote

Reporting Checklist: The authors have completed the STARD reporting checklist. Available at https://pm.amegroups.com/article/view/10.21037/pm-25-4/rc

Data Sharing Statement: Available at https://pm.amegroups.com/article/view/10.21037/pm-25-4/dss

Peer Review File: Available at https://pm.amegroups.com/article/view/10.21037/pm-25-4/prf

Funding: None.

Conflicts of Interest: All authors have completed the ICMJE uniform disclosure form (available at https://pm.amegroups.com/article/view/10.21037/pm-25-4/coif). The authors have no conflicts of interest to declare.

Ethical Statement: The authors are accountable for all aspects of the work in ensuring that questions related to the accuracy or integrity of any part of the work are appropriately investigated and resolved. This study was conducted in accordance with the Declaration of Helsinki and its subsequent amendments. The study was approved by the institutional ethics committee of Queen Sirikit National Institute of Child Health dated August 21, 2023 (No. IRB00007346) and informed consent was taken from the guardians of all individual participants.

Open Access Statement: This is an Open Access article distributed in accordance with the Creative Commons Attribution-NonCommercial-NoDerivs 4.0 International License (CC BY-NC-ND 4.0), which permits the non-commercial replication and distribution of the article with the strict proviso that no changes or edits are made and the original work is properly cited (including links to both the formal publication through the relevant DOI and the license). See: https://creativecommons.org/licenses/by-nc-nd/4.0/.

References

1. American Psychiatric Association. Diagnostic and statistical manual of mental disorders. 5th ed. Arlington, VA: American Psychiatric Publishing; 2013. Available online: https://doi.org/10.1176/appi.books.9780890425596
2. Maenner MJ, Shaw KA, Bakian AV, et al. Prevalence and Characteristics of Autism Spectrum Disorder Among Children Aged 8 Years - Autism and Developmental Disabilities Monitoring Network, 11 Sites, United States, 2018. MMWR Surveill Summ 2021;70:1-16. [Crossref] [PubMed]
3. Srisinghasongkram P, Pruksananonda C, Chonchaiya W. Two-Step Screening of the Modified Checklist for Autism in Toddlers in Thai Children with Language Delay and Typically Developing Children. J Autism Dev Disord 2016;46:3317-29. [Crossref] [PubMed]
4. Zwaigenbaum L, Bauman ML, Stone WL, et al. Early Identification of Autism Spectrum Disorder: Recommendations for Practice and Research. Pediatrics 2015;136:S10-40. [Crossref] [PubMed]
5. Bradshaw J, Steiner AM, Gengoux G, et al. Feasibility and effectiveness of very early intervention for infants at-risk for autism spectrum disorder: a systematic review. J Autism Dev Disord 2015;45:778-94. [Crossref] [PubMed]
6. Dawson G, Rogers S, Munson J, et al. Randomized, controlled trial of an intervention for toddlers with autism: the Early Start Denver Model. Pediatrics 2010;125:e17-23. [Crossref] [PubMed]
7. Estes A, Munson J, Rogers SJ, et al. Long-Term Outcomes of Early Intervention in 6-Year-Old Children With Autism Spectrum Disorder. J Am Acad Child Adolesc Psychiatry 2015;54:580-7. [Crossref] [PubMed]
8. Karst JS, Van Hecke AV. Parent and family impact of autism spectrum disorders: a review and proposed model for intervention evaluation. Clin Child Fam Psychol Rev 2012;15:247-77. [Crossref] [PubMed]
9. Samms-Vaughan ME. The status of early identification and early intervention in autism spectrum disorders in lower- and middle-income countries. Int J Speech Lang Pathol 2014;16:30-5. [Crossref] [PubMed]
10. Durkin MS, Elsabbagh M, Barbaro J, et al. Autism screening and diagnosis in low resource settings: Challenges and opportunities to enhance research and services worldwide. Autism Res 2015;8:473-6. [Crossref] [PubMed]
11. de Vries PJ. Thinking globally to meet local needs: autism spectrum disorders in Africa and other low-resource environments. Curr Opin Neurol 2016;29:130-6. [Crossref] [PubMed]
12. Robins DL, Casagrande K, Barton M, et al. Validation of the modified checklist for Autism in toddlers, revised with follow-up (M-CHAT-R/F). Pediatrics 2014;133:37-45. [Crossref] [PubMed]
13. Canal-Bedia R, García-Primo P, Martín-Cilleros MV, et al. Modified checklist for autism in toddlers: cross-cultural adaptation and validation in Spain. J Autism Dev Disord 2011;41:1342-51. [Crossref] [PubMed]
14. Kamio Y, Inada N, Koyama T, et al. Effectiveness of using the Modified Checklist for Autism in Toddlers in two-stage screening of autism spectrum disorder at the 18-month health check-up in Japan. J Autism Dev Disord 2014;44:194-203. [Crossref] [PubMed]
15. Koh HC, Lim SH, Chan GJ, et al. The clinical utility of the modified checklist for autism in toddlers with high risk 18-48 month old children in Singapore. J Autism Dev Disord 2014;44:405-16. [Crossref] [PubMed]
16. Beacham C, Reid M, Bradshaw J, et al. Screening for Autism Spectrum Disorder: Profiles of Children Who Are Missed. J Dev Behav Pediatr 2018;39:673-82. [Crossref] [PubMed]
17. Chaiudomsom K, Patjanasoontorn N, Suphakunpinyo C, et al. The Validity of the Modified Checklist for Autism in Toddlers, Revised with Follow-Up (M-CHAT-R/F) Thai Version as the Autism Screening Application: A Pilot Study. J Med Assoc Thai 2022;105:517-23.
18. Jacoby P, Epstein A, Kim R, et al. Reliability of the Quality of Life Inventory-Disability Measure in Children with Intellectual Disability. Journal of Developmental & Behavioral Pediatrics ;41:534-9. [Crossref] [PubMed]
19. Tangviriyapaiboon D, Sirithongthaworn S, Apikomonkon H, et al. Development and psychometric evaluation of a Thai Diagnostic Autism Scale for the early diagnosis of Autism Spectrum Disorder. Autism Res 2022;15:317-27. [Crossref] [PubMed]
20. Sirithongthaworn S. Development of a manual for monitoring and promoting early childhood development. J Psychiatr Assoc Thailand 2018;63:3-12.
21. Daniels AM, Mandell DS. Explaining differences in age at autism spectrum disorder diagnosis: a critical review. Autism 2014;18:583-97. [Crossref] [PubMed]
22. Sheldrick RC, Maye MP, Carter AS. Age at First Identification of Autism Spectrum Disorder: An Analysis of Two US Surveys. J Am Acad Child Adolesc Psychiatry 2017;56:313-20. [Crossref] [PubMed]
23. Ozonoff S, Young GS, Steinfeld MB, et al. How early do parent concerns predict later autism diagnosis? J Dev Behav Pediatr 2009;30:367-75. [Crossref] [PubMed]
24. Siu ALUS Preventive Services Task Force (USPSTF). Screening for Autism Spectrum Disorder in Young Children: US Preventive Services Task Force Recommendation Statement. JAMA 2016;315:691-6. [Crossref] [PubMed]
25. Mandell DS, Novak M. The role of culture in families’ treatment decisions for children with autism spectrum disorders. Ment Retard Dev Disabil Res Rev 2005;11:110-5. [Crossref] [PubMed]
26. Tariq Q, Daniels J, Schwartz JN, et al. Mobile detection of autism through machine learning on home video: A development and prospective validation study. PLoS Med 2018;15:e1002705. [Crossref] [PubMed]
27. Bordini D, Lowenthal R, Gadelha A, et al. Impact of training in autism for primary care providers: a pilot study. Braz J Psychiatry 2015;37:63-6. [Crossref] [PubMed]
28. Tekola B, Baheretibeb Y, Roth I, et al. Challenges and opportunities to improve autism services in low-income countries: lessons from a situational analysis in Ethiopia. Glob Ment Health (Camb) 2016;3:e21. [Crossref] [PubMed]
29. Divan G, Vajaratkar V, Desai MU, et al. Challenges, coping strategies, and unmet needs of families with a child with autism spectrum disorder in Goa, India. Autism Res 2012;5:190-200. [Crossref] [PubMed]
30. Thabtah F, Peebles D. Early Autism Screening: A Comprehensive Review. Int J Environ Res Public Health 2019;16:3502. [Crossref] [PubMed]
31. Zachor D, Yang JW, Itzchak EB, et al. Cross-cultural differences in comorbid symptoms of children with autism spectrum disorders: an international examination between Israel, South Korea, the United Kingdom and the United States of America. Dev Neurorehabil 2011;14:215-20. [Crossref] [PubMed]
32. Penner M, Rayar M, Bashir N, et al. Cost-Effectiveness Analysis Comparing Pre-diagnosis Autism Spectrum Disorder (ASD)-Targeted Intervention with Ontario’s Autism Intervention Program. J Autism Dev Disord 2015;45:2833-47. [Crossref] [PubMed]
33. Salomone E, Charman T, McConachie H, et al. Child’s verbal ability and gender are associated with age at diagnosis in a sample of young children with ASD in Europe. Child Care Health Dev 2016;42:141-5. [Crossref] [PubMed]
34. Hazlett HC, Gu H, Munsell BC, et al. Early brain development in infants at high risk for autism spectrum disorder. Nature 2017;542:348-51. [Crossref] [PubMed]
35. Hahler EM, Elsabbagh M. Autism: A Global Perspective. Curr Dev Disord Rep 2015;2:58-64.
36. Khowaja MK, Hazzard AP, Robins DL. Sociodemographic Barriers to Early Detection of Autism: Screening and Evaluation Using the M-CHAT, M-CHAT-R, and Follow-Up. J Autism Dev Disord 2015;45:1797-808. [Crossref] [PubMed]
37. Guthrie W, Wallis K, Bennett A, et al. Accuracy of Autism Screening in a Large Pediatric Network. Pediatrics 2019;144:e20183963. [Crossref] [PubMed]
38. Marlow M, Servili C, Tomlinson M. A review of screening tools for the identification of autism spectrum disorders and developmental delay in infants and young children: recommendations for use in low- and middle-income countries. Autism Res 2019;12:176-99. [Crossref] [PubMed]
39. Sritharan B, Koola MM. Barriers faced by immigrant families of children with autism: A program to address the challenges. Asian J Psychiatr 2019;39:53-7. [Crossref] [PubMed]