Gender differences and age patterns

Consistent with global ^{Reference Zhu, Zhang, Wang and Yang21} and regional ^{Reference Akomolafe, Abdallah, Mahmood, Elshoeibi, Al-Khulaifi and Mahmoud20} evidence, our results showed a persistently higher ASD burden among males compared with females across estimates of prevalence, incidence and YLDs. Male age-standardised rates were approximately two to three times higher than female rates throughout the study period. This pattern aligns with well-established gender differences in ASD diagnosis reported internationally, ^{Reference Napolitano, Schiavi, La Rosa, Rossi-Espagnet, Petrillo and Bottino22} and likely reflects a combination of biological susceptibility and potential underrecognition of ASD in females. ^{Reference Dougherty, Marrus, Maloney, Yip, Sandin and Turner23}

Age-specific analyses revealed that ASD burden was highest in early childhood and declined progressively with increasing age. The largest relative increases in prevalence and YLD rates between 1990 and 2023 were observed among children aged 2–4 and 5–9 years. This age-specific trend emphasises the importance of early childhood detection and intervention for ASD. These findings align with global burden estimates, indicating that ASD imposes a greater health burden in early life, with disability-adjusted life years peaking in childhood and declining with increasing age, highlighting the critical impact of ASD in younger populations. ^{Reference Santomauro, Erskine, Mantilla Herrera, Miller, Shadid and Hagins24} Such age-specific burden patterns have been reported in prior GBD analyses, where YLDs for ASD were highest among children and decreased across older age groups. ^{Reference Santomauro, Erskine, Mantilla Herrera, Miller, Shadid and Hagins24} In contrast, global and regional studies have suggested that modest declines or plateaus for age-specific ASD burden in older age groups may reflect differences between birth cohorts (i.e. people born in earlier periods who had different life experiences) and diagnostic practices that historically focused less on adults. ^{Reference Akomolafe, Abdallah, Mahmood, Elshoeibi, Al-Khulaifi and Mahmoud20,Reference Santomauro, Erskine, Mantilla Herrera, Miller, Shadid and Hagins24} They may also reflect survival with stable disability rather than progression to more severe health states. ^{Reference Rabba, Lampinen, Meldrum, Dwyer, South and Taylor25} Unfortunately, there are no age-stratified epidemiological studies from Jordan confirming this pattern. However, existing research in Jordan has focused primarily on childhood diagnoses, ^{Reference Zhu, Zhang, Wang and Yang21,Reference Hyassat, Al-Makahleh, Rahahleh and Al-Zyoud26} public knowledge, family experiences ^{Reference Shattnawi, Bani, Al-Natour, Al-Hammouri, Al-Azzam and Joseph27} and service gaps ^{Reference Masri, Al Suluh and Nasir28} rather than age-related prevalence or burden across the lifespan.

Regional context within the MENA

When viewed within the broader MENA context, the increasing burden of ASD observed in Jordan is not unique. Similar upward trends have been reported in recent GBD analyses of mental and neurodevelopmental disorders across the region. ^{Reference Ebrahimi Meimand, Amiri, Shobeiri, Malekpour, Saeedi Moghaddam and Ghanbari29} Many MENA countries have shown comparable patterns of ASD burden over time. ^{Reference Ebrahimi Meimand, Amiri, Shobeiri, Malekpour, Saeedi Moghaddam and Ghanbari29} Age-standardised prevalence and YLD rates were highest in Iran, Qatar and the United Arab Emirates in 2019, whereas lower rates were reported in Libya, Yemen and the Syrian Arab Republic. ^{Reference Ebrahimi Meimand, Amiri, Shobeiri, Malekpour, Saeedi Moghaddam and Ghanbari29} Although regional rates remained relatively stable from 1990 to 2019, the absolute number of ASD cases increased, likely reflecting improved case recognition and demographic growth across the region. ^{Reference Ebrahimi Meimand, Amiri, Shobeiri, Malekpour, Saeedi Moghaddam and Ghanbari29} However, the magnitude and pace of change varied considerably across the region, suggesting potential differences in health system infrastructures, diagnostic capacities, sociocultural perceptions of developmental disorders and data availability. ³⁰

Despite these similarities in regional patterns, detailed country-specific analyses of ASD remain scarce in the MENA region. Most available studies are limited to single time points, subnational samples or specific age groups, and few provided simultaneous estimates of prevalence, incidence and disability burden over extended periods. ^{Reference Shattnawi, Bani, Al-Natour, Al-Hammouri, Al-Azzam and Joseph27,Reference Al Jabery, Arabiat, Al Khamra, Betawi and Abdel Jabbar31} Nevertheless, the present study fills an important gap by offering a nationally focused, retrospective assessment of ASD burden in Jordan that is directly comparable with other GBD-based analyses.

National trends in ASD burden

At the national level, we observed that both all-ages counts and age-standardised rates of ASD prevalence and incidence increased by about threefold between 1990 and 2023. Although this pattern suggests a substantial rise in the recorded burden of ASD in Jordan, the interpretation of these trends requires caution, as multiple confounding factors likely contribute to this increase beyond a true rise in underlying risk.

First, part of the observed increase is likely driven by diagnostic expansion and improved case identification over time. The evolution of diagnostic criteria for ASD and greater clinical recognition, particularly following successive updates in classification systems, have broadened the diagnostic spectrum and facilitated identification of milder or previously unrecognised cases. ^{Reference Hirota and King32} In parallel, increased awareness among healthcare providers, educators and families in Jordan has likely contributed to higher help-seeking behaviour and referral rates, leading to improved detection. ^{Reference Shattnawi, Bani, Al-Natour, Al-Hammouri, Al-Azzam and Joseph27} Second, health system factors and service availability may have played a central role. Over the study period, Jordan has experienced gradual expansion in child developmental services, neurodevelopmental assessment capacity and special education services. These improvements likely enhanced diagnostic opportunities, particularly in urban settings, ^{Reference Shattnawi, Bani, Al-Natour, Al-Hammouri, Al-Azzam and Joseph27} and reduced underdetection that may have been more prominent in earlier decades.

Third, sociocultural changes and shifting perceptions of neurodevelopmental disorders may have influenced both recognition and reporting. ^{Reference AlSamhori, Kakish, Ellayyan, Mohammad, Hijazeen and Kheir33} Reduced stigma, increased parental awareness and greater acceptance of developmental screening in schools and primary care settings may have contributed to earlier identification of ASD cases that would previously have remained undiagnosed or misclassified.

In addition, demographic and epidemiological transitions in Jordan, including population growth and a sustained youthful age structure, could have increased the absolute number of individuals at risk of diagnosis. ^{Reference Tamimi, Al-Abbadi, Tamimi, Juweid, Ahmad and Tamimi34} Although age-standardised rates also increased, suggesting that demographic change alone does not fully explain the trend, population structure likely amplified the overall burden. It is important to mention that the methodological nature of GBD estimates, where ASD is modelled as a lifelong condition. Under this framework, incident cases accumulate over time, contributing to progressively increasing prevalence and YLDs. Therefore, even modest increases in modelled incidence can translate into substantial long-term increases in prevalence, particularly in the absence of recovery or remission assumptions.

Importantly, these findings should also be interpreted in light of the methodological limitations inherent to the GBD framework. GBD estimates are modelled rather than derived solely from nationally representative surveillance systems, and therefore depend heavily on the availability and quality of underlying epidemiological data. In regions such as the Middle East, including Jordan, limited population-based ASD registries, underdiagnosis, variability in access to specialist assessment and sparse longitudinal data may reduce estimate precision and increase uncertainty. In settings with incomplete primary data, the GBD framework incorporates statistical modelling techniques and covariates from comparable countries to generate estimates. Although the GDB approach may improve comparability across regions and years, it may also introduce uncertainty if local epidemiological patterns differ from those assumed by the model. Consequently, part of the observed temporal increase may reflect improvements in data capture and modelling inputs over time, rather than a true epidemiological rise.

Together, these findings suggest that the observed trends likely reflect a combination of true changes in recognised incidence, improved diagnostic capacity, health system development, sociocultural shifts and evolving data availability, rather than a single underlying aetiological driver. Unravelling the relative contribution of these factors remains challenging, but is essential for accurate interpretation of temporal trends in ASD burden in Jordan.

In the absence of country-level data, GBD 2023 estimates for Jordan were generated using standardised Bayesian meta-regression modelling (DisMod-MR 2.1), which synthesises information from available regional and global data sources, alongside relevant covariates such as the sociodemographic index, health system indicators and temporal trends in diagnostic practices. This approach enables internally consistent estimation across locations and time periods, but relies on borrowing data from neighbouring countries and comparable settings. As a result, observed temporal trends in ASD burden for Jordan should be interpreted as changes in modelled estimates rather than directly observed epidemiological trends. These patterns may reflect improvements in diagnostic recognition, service availability and global awareness of ASD, in addition to underlying demographic changes, rather than true increases in disease occurrence.

Implications for the Jordanian health system

The sustained increase in ASD burden has important implications for health, education and social care systems in Jordan. The rising prevalence and YLDs introduces a growing demand for comprehensive diagnostic services, early intervention programmes, special inclusive education and long-term support across the lifespan. ^{Reference Kim35} The concentration of burden in early childhood highlights the need for systematic national screening and early developmental services. In addition, the persistence of disability into adulthood highlights the importance of transition planning and adult support services. At the regional level, our findings emphasise the need for MENA-specific, ASD-related strategies. ^{Reference Smythe, Scherer, Nanyunja, Tann and Olusanya36} These strategies for addressing neurodevelopmental disorders might include workforce training, strengthening data infrastructure, integration of developmental health into primary care systems, and providing evidence-based rehabilitation and community integration services. ^{Reference Smythe, Scherer, Nanyunja, Tann and Olusanya36} GBD-based national analyses such as this study can support evidence-informed planning and facilitate comparisons across countries with similar demographic and health system characteristics.

From a rehabilitative services perspective, the increasing burden suggests a rising demand for multidisciplinary care, including child and adolescent mental health services, developmental paediatrics, psychology, speech and language therapy, occupational therapy and special education support. ^{Reference Alkahtani, Al Awaji, Al Bashir, Alahmari, Alwuraykah and Alotaibi37} However, current service provision in Jordan remains limited and unevenly distributed, with a strong concentration in urban areas and reliance on a small number of specialised centres. This imbalance may contribute to unmet needs, long waiting times and delayed access to intervention. ^{Reference Ronksley, Sanmartin, Quan, Ravani, Tonelli and Manns38}

Early detection and intervention remain critical priorities. In Jordan, developmental screening is not yet fully integrated into routine primary care or maternal–child health services, and access to standardised diagnostic pathways remains inconsistent. Strengthening early screening, improving referral systems and embedding ASD detection within primary healthcare could support earlier diagnosis and timely intervention during key neurodevelopmental periods, which are associated with better functional outcomes. ^{Reference Alkahtani, Al Awaji, Al Bashir, Alahmari, Alwuraykah and Alotaibi37}

Workforce capacity is a major structural constraint, with expected increases in ASD burden likely to further strain already limited numbers of developmental specialists, psychologists, speech and language therapists, occupational therapists and special education professionals. ^{Reference Alkahtani, Al Awaji, Al Bashir, Alahmari, Alwuraykah and Alotaibi37} Addressing this requires investment in pre-service and in-service training, expansion of multidisciplinary programmes and development of national clinical guidelines to ensure consistent care. ^{Reference Yang and Kim39} At the policy level, ASD should be more explicitly integrated into national health, education and social protection strategies, including strengthened disability-inclusive policies, expanded community-based rehabilitation and improved intersectoral coordination. ^{Reference Bertelli, Bianco, Salerno and Salvador-Carulla40} Establishing national ASD surveillance systems and strengthening routine data collection would support better planning. Although GBD estimates provide useful guidance for long-term planning, they should be interpreted in light of methodological limitations and used primarily to inform strategic resource allocation, rather than precise estimates of service need.

Strengths and limitations

The primary strength of this study lies in its use of standardised GBD 2023 estimates, enabling robust comparisons across time, gender and age groups with a transparent and reproducible methodology. However, a number of limitations should be acknowledged. GBD estimates are based on available data inputs and modelling assumptions, and in settings where primary epidemiological data are limited, uncertainty intervals remain wide. Jordan currently lacks nationally representative epidemiological studies, disease registries and surveillance systems for ASD. As highlighted by recent systematic reviews of ASD prevalence in the MENA region, no population-based studies from Jordan were identified, highlighting a substantial national evidence gap. ^{Reference Akomolafe, Abdallah, Mahmood, Elshoeibi, Al-Khulaifi and Mahmoud20} Consequently, the 2023 GBD study did not incorporate Jordan-specific primary epidemiological data when estimating ASD prevalence, incidence or disability. Country-specific data, however, might be underestimating ASD-related statistics, as national registries are not available. In addition, age-specific incidence estimates were not available for all age groups, restricting detailed age-pattern analyses for the incidence of ASD. Nevertheless, the overall trends observed in this comprehensive analysis are internally consistent and align with regional and global patterns. These estimates do not reflect directly observed case counts, but are derived from GBD modelling frameworks that account for evolving diagnostic definitions, improvements in global ASD recognition, demographic changes and limited diagnostic capacity in earlier decades, particularly in data-scarce settings such as Jordan. Future studies should encourage the establishment of national registries and testing evidence-based ASD interventions specifically applied in the country context. In addition, the integration of artificial intelligence and machine learning approaches is crucial in ASD diagnostic process. These suggested methodological techniques may overcome certain challenges in early detection of ASD, including limited specialist services, inequities in access to healthcare and inconsistencies in clinical evaluations across different clinical contexts.

Conclusion

This study describes changes in the burden of ASD in Jordan between 1990 and 2023, with increases observed in prevalence, incidence and disability-related metrics, as well as differences by age and gender. These findings should be interpreted in light of the descriptive and modelled nature of the data, which reflect underlying assumptions of the GBD framework and the availability of epidemiological inputs over time. Rather than claiming definitive causal or epidemiological changes, the observed trends likely reflect a combination of changes in diagnostic practices, health system capacity, data availability and modelled estimates. Nevertheless, this study provides useful descriptive evidence that may support health policy discussions, service planning and future research on neurodevelopmental disorders in Jordan and similar settings.