Abstract
Objectives
To assess the trends of autism spectrum disorders (ASD) in China and globally, to extract the epidemiological characteristics and to predict the burden in 2030. Data on the prevalence and disability-adjusted life years (DALYs) for ASD from 1990 to 2021 were collected from the Global Burden of Disease (GBD) database.
Methods
The Join-point regression model was used to analyse changes in the disease burden of ASD over time, and the Auto-regressive Integrated Moving Average (ARIMA) model was employed to predict the trend by 2030.
Results
From 1990 to 2021, the age-standardised prevalence and age-standardised DALYs rate of ASD in China were lower than the global average level. However, the average annual percentage change (AAPC) in these rates was 0.22% for prevalence and 0.23% for DALYs, which were higher than the global average levels. It is forecasted that the disease burden of ASD in China showed an uptrend by 2030, corresponding to the age-standardised prevalence of 660.28 per 100 000 and age-standardised DALYs rate of 126.66 per 100 000. In comparison, the global levels are 793.01 per 100 000 and 147.43 per 100 000 by 2030, respectively.
Conclusions
Our findings demonstrate a strong positive correlation between the age-standardised prevalence and DALYs rate of ASD and sociodemographic index (SDI) across most countries. The health impact of ASD has been on an upward trajectory over the past three decades. ASD significantly contributes to health loss, particularly among preschool boys. The disease burden of ASD in China is likely to increase progressively in the future, highlighting the urgency of implementing targeted prevention and control measures.
Keywords: Adolescent Health, Health Policy, Epidemiology
WHAT IS ALREADY KNOWN ON THIS TOPIC
Autism spectrum disorder (ASD) has been recognised as an increasing global public health concern over the past three decades. However, a systematic characterisation of its long-term trends in China especially in comparison with global rates and future projections has remained limited.
WHAT THIS STUDY ADDS
By analysing Global Burden of Disease data, this study reveals that from 1990 to 2021,China’s age-standardised prevalence and disability-adjusted life years (DALYs) rate in average annual percentage change for ASD higher than the global average, and are projected to continue rising through 2030.
HOW THIS STUDY MIGHT AFFECT RESEARCH, PRACTICE OR POLICY
These results underscore a growing ASD disease burden in China, highlighting the need for targeted prevention and control strategies, particularly among high-risk groups such as preschool-aged boys and providing critical evidence to guide public health resource planning.
Introduction
Autism spectrum disorder (ASD) is a complex neurodevelopmental disorder, which is characterised by social communication impairments, restricted interests or range of activities, and repetitive and stereotyped behaviours.1 ASD definition was based on the clinical threshold as established by the Diagnostic and Statistical Manual of Mental Disorders (DSM), the International Classification of Disease criteria, the Chinese Classification of Mental Disorders or diagnosed by a clinician using established tools.
Globally, approximately 1 in 100 children are diagnosed with ASD, a condition that profoundly impacts not only affected individuals but also caregivers and broader societal structures.2 This rising prevalence reflects, in part, substantial revisions to diagnostic frameworks over recent decades—notably, DSM-5’s consolidation of all pervasive developmental disorders into a single ASD category, which significantly broadened diagnostic scope. These evolving criteria underscore the dynamic nature of ASD diagnosis and emphasise the necessity of accounting for such changes when interpreting epidemiological data. Based on the estimates of the Autism and Developmental Disabilities Monitoring Network of the Centers for Disease Control and Prevention in the USA in 2023, about 1 in every 36 children (2.7%) is diagnosed with ASD. This estimate has increased by 89% compared with 1 in 68 in 2010 and by 317% compared with 1 in 150 in 2000 (https://www.cdc.gov/ncbddd/autism/data.html). The results of a 2022 study showed that the prevalence of ASD in Asia, the Americas, Europe, Africa and Australia was 0.4% (95% CI 0.1 to 1), 1% (95% CI 0.8 to 1.1), 0.5% (95% CI 0.2 to 1), 1% (95% CI 0.3 to 3.1) and 1.7% (95% CI 0.5 to 6.1), respectively.3 The epidemiological survey based on a large sample population in China showed that the prevalence of ASD was 0.7%.4 The increasing prevalence of ASD highlights the critical need for comprehensive understanding and efficient regulation of the disorder, given its multifaceted implications that extend beyond the individual and familial levels to encompass broader societal and healthcare systems.5
The Global Burden of Disease (GBD) Study, led by the Institute for Health Metrics and Evaluation at the University of Washington, uses a multilevel, Bayesian statistical modelling framework to estimate disease burden globally. This approach integrates data from various sources, including vital registration systems, surveys and expert opinions, to produce robust, comparable estimates of mortality, morbidity and risk factors across different populations and time periods.6 It quantifies the health burden caused by diseases, injuries and risk factors, introducing the key metric of disability-adjusted life years (DALYs). DALYs combine years of life lost (YLLs) due to premature death and years lived with disability (YLDs), offering a more complete picture of disease burden than traditional measures like mortality and incidence rates, which do not capture non-fatal consequences or the severity and duration of disability. It enables a more thorough understanding of the total impact of diseases on population health, considering both fatal and non-fatal outcomes and highlights the role of risk factors in shaping global health.7 ASD is classified as a non-fatal condition that generally does not result in premature mortality, thus having no impact on the YLLs component of the DALY metric. Instead, the primary effect of ASD is reflected in the YLD component, which accounts for the disability associated with the disorder.
Based on the GBD database, this study employs rigorous analytical methodologies to systematically characterise the evolving patterns of ASD-related health burden, dissecting age-specific distributions, gender disparities, sociodemographic index (SDI) correlations and geographic variations in DALYs, with a view to elucidate current epidemiological profiles and emerging trends of ASD burden in China and globally. The findings provide an evidence base for developing targeted intervention strategies, optimising policy frameworks, while advancing societal awareness and inclusion to ultimately improve health outcomes and quality of life for this population.
Materials and methods
Patient and public involvement statement
We declare that this study solely involves the analysis of GBD data, without the inclusion of any patients or public participation.
Data acquisition and sources
The GBD Study 2021 uses Bayesian hierarchical models to synthesise data from disparate sources, accounting for uncertainty in each data point. It provides an unparalleled lens into the health implications of diseases and injuries across 204 countries and territories. For this study, we meticulously extracted data pertaining to DALYs and prevalence attributable to ASD from the GBD study database, accessible at GBD Results Tool (http://ghdx.healthdata.org). The prevalent cases, prevalence, age-standardised prevalence, DALYs, DALYs rate and age-standardised DALYs rate of ASD in China and globally from 1990 to 2021 were selected for statistical analysis. One DALY represents the loss of the equivalent of 1 year of full health. DALYs for a disease or health condition are the sum of the YLLs due to premature mortality and the years lived with a disability due to prevalent cases of the disease or health condition in a population, and the age-standardised DALYs rate is a weighted average of the age-specific DALY rate per 100 000 cases, where the weights are the proportions of cases in the corresponding age groups.8 The SDI values for the respective countries and regions in 2021 were sourced from the GBD study database. The SDI is a composite measure of a country’s level of development index that takes into account per capita income, educational attainment and the total fertility rate in a country.7 We used the Strengthening the Reporting of Observational Studies in Epidemiology (STROBE) reporting guideline to draft this manuscript, and the STROBE reporting checklist when editing.
Statistical methodology
In this study, Spearman’s rank correlation analysis was employed to systematically examine the associations between ASD burden metrics and the national SDI. The Join-point regression model was selected for its ability to identify significant changes in trends over time, allowing us to pinpoint specific periods where the disease burden of ASD experienced notable shifts. The Join-point Regression Program (V.4.9.0.0) was utilised to analyse the trend of ASD disease burden in China and globally from 1990 to 2021 and calculate the average annual percentage change (AAPC) with a significance level of α=0.05.9 The auto regression integrated moving average (ARIMA) model was chosen to predict the trend of ASD disease burden in China and globally from 2022 to 2030 due to its wide application in time series forecasting. Difference operation was performed on the time series data until it is transformed into a stationary sequence. The model with the minimum Akaike information criterion (AIC) was selected as the optimal prediction model. Ljung-Box test evaluated the model fit by testing residual auto-correlation, and white noise residuals indicated well-fitted model. This process was implemented through the R-forecast package, with a significance level of α=0.05.10
Results
The global and regional incidence of autism in 1990 and 2021 and AAPC of ASR from 1990 to 2021 are shown in table 1. All SDI regions and GBD regions are separated clearly. From 1990 to 2021, the incidence of ASD is showing an upward trend. At the national level, there are significant differences in the prevalence, and DALYs of ASD globally.
Table 1. Disease burden of ASD in China and globally in 1990 and 2021.
| Prevalent cases | Age-standardised prevalence | DALYs | Age-standardised DALYs rate | |
|---|---|---|---|---|
| China | ||||
| 1990 | 7 347 264 | 612.61 | 1 392 966 | 115.81 |
| 2021 | 9 078 613 | 655.75 | 1 700 739 | 124.19 |
| AAPC | 0.68 | 0.22 | 0.64 | 0.23 |
| P value | <0.05 | <0.05 | <0.05 | <0.05 |
| Global | ||||
| 1990 | 41 929 996 | 773.25 | 7 868 387 | 144.51 |
| 2021 | 61 823 540 | 788.34 | 11 544 038 | 147.56 |
| AAPC | 1.26 | 0.07 | 1.24 | 0.07 |
| P value | <0.05 | <0.05 | <0.05 | <0.05 |
Age-standardised prevalence and age-standardised DALYs rate per 100 000.
AAPC, average annual percentage change; ASD, autism spectrum disorders; DALYs, disability-adjusted life years.
From 1990 to 2021, the age-standardised prevalence and age-standardised DALYs rate of ASD in China were lower than the global average level. However, the AAPC in these rates was 0.22% for prevalence and 0.23% for DALYs, which were higher than the global average levels. It is forecasted that the disease burden of ASD in China showed an uptrend by 2030, corresponding to the age-standardised prevalence of 660.28 per 100 000 and age-standardised DALY rate of 126.66 per 100 000. In comparison, the global levels are 793.01 per 100 000 and 147.43 per 100 000 by 2030, respectively. Moreover, regarding the range of the SDI quintiles, the High SDI quintile in 2021 had the highest incidence, prevalence and DALY counts (see online supplemental appendix 1).
The disease burden of ASD in China and globally in 1990 and 2021
Global maps were generated from a data visualisation tool available from the Global Health Data Exchange (GHDx) repository (https://vizhub.healthdata.org/gbd-compare/), see online supplemental appendix 2. There are significant geographical differences in the burden distribution of ASD. In certain countries in Europe and North America, some nations in southern South America, Japan and South Korea among East Asian countries, as well as Australia and New Zealand among Oceania countries, the age-standardised prevalence and the age-standardised DALY rate are relatively high, indicating a relatively heavier disease burden of ASD.
Our analysis of the age-standardised prevalence and DALY rate of ASD across 204 countries revealed a strong positive correlation with the SDI shown in figure 1. Specifically, the Pearson correlation coefficient (r) between age-standardised ASD prevalence and SDI exceeded 0.8 in 143 countries, while the correlation between age-standardised DALY rate and SDI surpassed 0.8 in 147 countries. Notably, China exhibited an exceptionally strong correlation, with r=0.986 for age-standardised prevalence and r=0.987 for age-standardised DALY rate.
Figure 1. The correlation distribution between disease burden of ASD and SDI in 204 countries globally. (a) The correlation distribution between age-standardised prevalence and SDI; (b) the correlation distribution between age-standardised DALYs rate and SDI. ASD, autism spectrum disorder; DALYs, disability-adjusted life years; SDI, sociodemographic index.
Prevalent cases, DALYs, age-standardised prevalence and age-standardised DALY rate of ASD in China and globally in 1990 and 2021 are reported in table 1. In 2021, ASD was associated with 9 078 613 prevalent cases and 1 700 739 DALYs in China, the prevalent cases and DALYs increased by an average of 0.68% and 0.64% from 1990 to 2021. ASD was associated with 655.75 per 100 000 age-standardised prevalence and 124.19 per 100 000 age-standardised DALYs rate, the age-standardised prevalence and age-standardised DALY rate increased by an average of 0.22% and 0.23% from 1990 to 2021 in China. Also in 2021, ASD was associated with 61 823 540 prevalent cases and 11 544 038 DALYs globally, the prevalent cases and DALYs increased by an average of 1.26% and 1.24% from 1990 to 2021. ASD was associated with 788.34 per 100 000 age-standardised prevalence and 147.56 per 100 000 age-standardised DALY rate globally, the age-standardised prevalence and age-standardised DALY rate increased by an average of 0.07% and 0.07% from 1990 to 2021.
The disease burden of ASD by different sexes in China and globally
Prevalent cases, DALYs, age-standardised prevalence and age-standardised DALYs rate of ASD by different sexes in China and globally are reported in table 2. It can be found from the table that all metrics of the disease burden of ASD in China and globally showed an average increase with positive AAPCs from 1990 to 2021 for both males and females.
Table 2. Disease burden of ASD by different sexes in China and globally from 1990 to 2021.
| China | Global | |||||||
|---|---|---|---|---|---|---|---|---|
| Prevalent cases | Age-standardised prevalence | DALYs | Age-standardised DALYs rate | Prevalent cases | Age-standardised prevalence | DALYs | Age-standardised DALYs rate | |
| Male | ||||||||
| 1990 | 5 522 941 | 892.25 | 1 049 190 | 168.89 | 28 580 940 | 1046.01 | 5 379 577 | 195.93 |
| 2021 | 6 934 846 | 967.95 | 1 301 567 | 183.4 | 42 133 879 | 1064.71 | 7 896 838 | 199.8 |
| AAPC | 0.73 | 0.27 | 0.7 | 0.27 | 1.26 | 0.06 | 1.25 | 0.06 |
| P value | <0.05 | <0.05 | <0.05 | <0.05 | <0.05 | <0.05 | <0.05 | <0.05 |
| Female | ||||||||
| 1990 | 1 824 323 | 314.72 | 343 777 | 59.17 | 13 349 055 | 497.32 | 2 488 810 | 92.39 |
| 2021 | 2 143 767 | 318.17 | 399 172 | 60.01 | 19 689 661 | 508.08 | 3 647 200 | 94.45 |
| AAPC | 0.53 | 0.11 | 0.5 | 0.12 | 1.26 | 0.07 | 1.24 | 0.07 |
| P value | <0.05 | <0.05 | <0.05 | <0.05 | <0.05 | <0.05 | <0.05 | <0.05 |
Age-standardised prevalence and age-standardised DALYs rate per 100 000.
AAPC, average annual percentage change; ASD, autism spectrum disorder; DALYs, disability-adjusted life years.
Both prevalent cases and DALYs remained more than three times higher in males than females in China during the last 3 decades, while the ratio of these two burden metrics globally was greater than two times. The male to female ratio of age-standardised prevalence and age-standardised DALY rate were progressively increased in China from 2000 to 2021, while the ratio of these two burden metrics globally was progressively decreased from 2003 to 2021. Ratios of males and females are shown in figure 2.
Figure 2. Sex ratio trends of disease burden of ASD from 1990 to 2021. (a) Male to female ratio of prevalent cases and age-standardised prevalence of China; (b) male to female ratio of DALYs and age-standardised DALYs rate of China; (c) male to female ratio of prevalent cases and age-standardised prevalence globally; (d) male to female ratio of DALYs and age-standardised DALYs rate globally. ASD, autism spectrum disorder; DALYs, disability-adjusted life years.
The disease burden of ASD in different age groups in China and globally
In China, both the prevalence and DALYs rate of ASD decreased with ageing in 2021. The prevalence and DALY rate were the highest in the 0–4 age group, which were 713.66 per 100 000 and 139.36 per 100 000, respectively. In the 80+ age group, the prevalence and DALY rate were the lowest, which were 320.50 per 100 000 and 53.12 per 100 000, respectively. Globally, both the prevalence and DALY rate of ASD decreased with ageing in 2021. The prevalence and DALY rate were the highest in the 0–4 age group, which were 877.73 per 100 000 and 169.20 per 100 000, respectively. In the 80+ age group, the prevalence and DALY rate were the lowest, which were 362.05 per 100 000 and 59.47 per 100 000, respectively. In China, the prevalence and DALY rate of ASD were lower than the global average level in every age group. See table 3.
Table 3. Disease burden of ASD in different age groups in China and globally.
| Age group | 1990 | 2021 | ||||||
|---|---|---|---|---|---|---|---|---|
| Prevalence | DALYs rate | Prevalence | DALYs rate | |||||
| China | Global | China | Global | China | Global | China | Global | |
| 0–4 | 686.77 | 851.86 | 133.11 | 163.06 | 713.66 | 877.73 | 139.36 | 169.20 |
| 5–9 | 669.08 | 830.92 | 129.71 | 159.60 | 702.86 | 853.14 | 136.77 | 164.44 |
| 10–14 | 652.19 | 819.26 | 125.74 | 156.76 | 696.17 | 840.95 | 134.76 | 161.21 |
| 15–19 | 641.82 | 805.19 | 123.37 | 153.30 | 691.70 | 831.01 | 133.37 | 158.35 |
| 20–24 | 638.74 | 794.66 | 122.10 | 150.47 | 685.70 | 820.52 | 131.70 | 155.45 |
| 25–29 | 637.94 | 798.33 | 121.40 | 150.18 | 679.52 | 810.57 | 129.74 | 152.64 |
| 30–34 | 637.47 | 803.87 | 120.73 | 150.45 | 671.37 | 798.00 | 127.50 | 149.55 |
| 35–39 | 624.95 | 792.2 | 117.73 | 147.47 | 664.30 | 793.98 | 125.19 | 147.81 |
| 40–44 | 619.16 | 796.31 | 115.68 | 147.16 | 653.51 | 788.52 | 122.63 | 145.83 |
| 45–49 | 607.03 | 780.39 | 112.71 | 143.12 | 637.95 | 770.99 | 118.56 | 141.57 |
| 50–54 | 596.15 | 762.81 | 109.84 | 138.77 | 628.38 | 751.82 | 115.78 | 136.95 |
| 55–59 | 578.03 | 737.95 | 105.30 | 132.78 | 613.65 | 733.04 | 111.72 | 131.98 |
| 60–64 | 548.39 | 706.6 | 98.71 | 125.56 | 600.17 | 718.97 | 107.85 | 127.63 |
| 65–69 | 492.01 | 635.42 | 87.23 | 111.16 | 560.81 | 662.30 | 99.03 | 115.74 |
| 70–74 | 415.88 | 541.04 | 72.33 | 92.97 | 509.62 | 602.51 | 88.45 | 103.45 |
| 75–79 | 330.59 | 442.61 | 56.68 | 74.70 | 442.26 | 509.25 | 75.40 | 85.86 |
| 80+ | 222.19 | 294.38 | 37.34 | 48.57 | 320.50 | 362.05 | 53.12 | 59.47 |
Prevalence and DALYs rate per 100 000.
ASD, autism spectrum disorder; DALYs, disability-adjusted life years.
Forecasting the disease burden of ASD in China and globally by 2030
Based on the time series of the age-standardised prevalence and age-standardised DALY rate of ASD, the auto.arima function automatically selected the models with the optimal AIC. The stationarity test results are available in online supplemental appendix 3. It is found that all sequences belong to white noise sequences with the p values of each sequence significantly greater than 0.05, so that here is no autocorrelation or partial autocorrelation. The model is established with high credibility. The Ljung-Box test results are available is online supplemental appendix 4. As the projection shown in figure 3, it is forecasted that the disease burden of ASD in China showed an uptrend by 2030, corresponding to the age-standardised prevalence of 660.28 per 100 000 and age-standardised DALY rate of 126.66 per 100 000. In comparison, the global age-standardised prevalence and age-standardised DALY rate of ASD are 793.01 per 100 000 and 147.43 per 100 000 by 2030, respectively. The forecasting results are summarised in table 4.
Figure 3. Trends in the disease burden of ASD in China and globally by 2030. (a) Age-standardised prevalence of ASD in China; (b) age-standardised DALYs rate of ASD in China; (c) age-standardised prevalence of ASD globally; (d) age-standardised DALYs rate of ASD globally. Grey bands represent 95% CIs. Age-standardised prevalence and age-standardised DALYs rate per 100 000 population. ASD, autism spectrum disorder; DALYs, disability-adjusted life years.
Table 4. Forecasting results of the disease burden of ASD in China and globally in 2030.
| Burden metrics | Male | Female | Both | |
|---|---|---|---|---|
| China | DALYs | 1 369 502 (1 358 148, 1 380 855) | 419 669 (416 558, 422 781) | 1 790 093 (1 774 011, 1 806 174) |
| Prevalent cases | 7 317 701 (7 254 391, 7 381 012) | 2 264 028 (2 228 139, 2 299 917) | 9 581 263 (9 498 051, 9 664 476) | |
| Age-standardised DALYs rate | 188.03 (184.64, 191.43) | 59.44 (55.84, 63.03) | 126.66 (123.44, 129.88) | |
| Age-standardised prevalence | 984.13 (967.20, 1001.06) | 304.93 (279.32, 330.54) | 660.28 (645.21, 675.35) | |
| Global | DALYs | 8 627 656 (8 584 894, 8 670 418) | 3 983 507 (3 957 821, 4 009 192) | 12 611 163 (12 547 066, 12 675 260) |
| Prevalent cases | 46 068 602 (45 869 758, 46 267 447) | 21 134 541 (20 937 132, 21 331 950) | 67 599 085 (67 294 453, 67 903 716) | |
| Age-standardised DALYs rate | 200.32 (198.73, 201.91) | 94.67 (93.11, 96.24) | 147.43 (145.92, 148.93) | |
| age-standardised prevalence | 1067.27 (1061.19, 1073.36) | 508.14 (499.39, 516.89) | 793.01 (790.55, 795.47) |
Data in parentheses are 95% CIs. Age-standardised prevalence and age-standardised DALYs rate per 100 000.
ASD, autism spectrum disorder; DALYs, disability-adjusted life years.
Discussion
Our study leveraged the GBD database, which integrates diverse global data sources, to elucidate the evolving epidemiological landscape of ASD, highlighting critical disparities in disease burden across regions, sexes, age groups and SDIs, with a focused lens on China. One key finding was the substantial increase in disease burden of ASD in China from 1990 to 2030.
Current disease burden of ASD
The Joinpoint regression model, used to analyse trends from 1990 to 2021, indicated significant variations in the global distribution of ASD’s disease burden. Joinpoint regression assumes that the data can be divided into distinct segments, each with its own linear trend. These trends may reflect the impact of factors such as social-economic conditions or interventions.11 Our study found that both the age-standardised prevalence and age-standardised DALY rate of ASD in China were lower than the global averages. Our findings demonstrate a strong positive correlation between the age-standardised prevalence and DALY rate of ASD and SDI across most countries. This suggests that higher SDI nations, which typically have better healthcare infrastructure, greater diagnostic capacity and increased awareness of neurodevelopmental disorders, may report higher ASD detection rates. The strong positive correlation observed in China further supports the idea that socioeconomic development is closely linked to ASD diagnosis and reporting. Over the past three decades, the increasing global population has driven an upward trend in the number of ASD prevalence and DALYs in China and globally, with population structure significantly influencing the burden. It has to be emphasised that data collection in China has several limitations, such as small sample sizes, limited follow-up data and the lack of contemporary diagnostic tools, all contributing to the under-reporting of ASD cases compared with Western countries.12 Moreover, China’s relatively late initiation of ASD epidemiological research, combined with insufficient support in manpower, technology and funding, limits the depth and scale of ASD studies.
Sex and age differences in disease burden of ASD
Several hypotheses explain the male bias observed in ASD prevalence, such as under-recognition of ASD in females,13 differences in brain structure and genetic load,14 and potential female protective effects against autistic behaviour.15 These elements encompassing diagnostic biases, biological vulnerability and possible protective mechanisms collectively shape the observed gender disparities in ASD prevalence and burden. Our findings showed that the global male age-standardised prevalence and age-standardised DALY rate were 2.1 times higher than female. This differs from the more commonly reported male-to-female prevalence ratio of 3–4, as we account for population age structures. The male to female sex ratio in China has shown a slight widening over the past two decades. This trend coincides with the period from 2000 to 2021, which was marked by rapid modernisation and expansion of the national healthcare system, particularly in mental health services and primary care networks. However, significant regional disparities in diagnostic capacity persisted, leading to the under-recognition of ASD among females in certain areas. Early public awareness campaigns predominantly focused on the typical symptom presentations in males. Furthermore, in some Chinese families, traditional cultural norms have contributed to relatively less developmental attention being directed towards female children. This study also found that the age group of 0–4 years had the highest prevalence in 2021, consistent with the findings of the recent Chinese study.16 The disease burden of ASD in each age group in China was lower than the global average. The use of screening instruments such as the Clancy Autism Behavior Scale, which often rely on parental reports, may contribute to the under-reporting of ASD, leading to lower prevalence estimates.17
The increase in disease burden of ASD over time
Several factors contribute to the growing burden of ASD over time. First, advancements in medical technology and professional knowledge have significantly enhanced the understanding, identification and diagnosis of ASD. The development of eye-tracking technology has improved early autism diagnosis, achieving 91% sensitivity and 87% specificity, thereby reducing wait times for evaluations and enabling timely intervention.18 Second, changes in diagnostic criteria have expanded the diagnostic scope of ASD, driven by continuous updates and revisions. The fifth edition of the DSM merged all of the diagnoses that were under the pervasive developmental disorders category into a single disorder termed ASD.19 Third, environmental factors, including social stress, environmental pollution and gestational infections, have been associated with an increased prevalence and disease burden of ASD. Exposure to environmental pollutants, such as air pollution and toxic metals, has been associated with an increased risk of ASD.20 Fourth, increased awareness and understanding of ASD among patients and their families have led to earlier recognition of symptoms and a higher rate of diagnoses. A 175% increase in ASD diagnoses in the USA, from 2011 to 2022, is attributed to heightened advocacy, education and changes in screening practices.21
Forecasting of ASD disease burden
The ARIMA model is a widely used model that predicts future values of a time series by utilising the past values of the same series.22 It forecasted that the age-standardised prevalence and age-standardised DALY rate of ASD in China would increase from 2022 to 2030, and the global age-standardised DALY rate of ASD would decrease from 2022. We generated prediction intervals to provide a range of possible outcomes for the disease burden of ASD in 2030. These intervals account for the variability in the data and the potential impact of external factors such as advances in medical technology, changes in public health policies and shifts in population structure. While our model suggests an increasing trend in the disease burden of ASD, it is important to note that the actual future burden may vary within these intervals.
Globally, the relatively stable age suggests that while the overall burden of ASD will continue to be significant, the rate of increase may slow down. However, this does not mean that the global community can afford to be complacent. On the one hand, the Chinese government has implemented several initiatives to enhance rehabilitation assistance for children with ASD, particularly focusing on those aged 0 to 6 years, establishing outpatient and inpatient services.23 On the other hand, China’s health expenditure remains below the global average, with health spending accounting for only 5.38% of GDP in 2021 compared with the global average of 10.34% (https://apps.who.int/nha/database). There has been a huge gap in global investment in mental disorders, and the growth rate of global overall expenditure has remained stable in recent years.24 Furthermore, China’s fertility rate is declining, contributing to an ageing population. The Chinese population is projected to age rapidly from 2021 to 2050.25 This demographic shift will likely lead to a significant increase in the disease burden of ASD due to the rising YLDs.
Policy implications
Through comparative and predictive analysis of ASD prevalence and DALYs between China and the global population, we provide actionable policy recommendations. We recommend prioritising early screening and diagnostic capacity-building in resource-limited regions through standardised tools and provider training; expanding access to culturally appropriate rehabilitation support; launching public awareness campaigns to reduce stigma via educational programmes in schools, community centres and healthcare settings; and advancing research on ASD aetiology and modifiable risk factors to develop targeted interventions.
With the improvement of national development levels will promote the upgrading of autism health policies from ‘spot assistance’ to ‘full-cycle, inclusive and rule-of-law-based’ approaches through multiple dimensions, including policy coverage, resource allocation, service systems, safeguard mechanisms and social support, significantly enhancing the health and quality of life of individuals with autism.
These evidence-based recommendations aim to guide public health planning, inform policy development and support the design of interventions that respond to the evolving needs of individuals with ASD and their families. Addressing the limitations of this study while building on its strengths will be critical for developing comprehensive, evidence-based approaches to effectively manage and mitigate the future ASD disease burden.
Limitations
While our study provides important insights, several limitations need to be addressed. First, the heterogeneity introduced by varying screening procedures and diagnostic criteria across different GBD studies provided by different countries warrant consideration.26 Second, the ARIMA model usually assumes that the data is stationary or can be transformed into a stationary sequence through methods such as differencing. However, the factors influencing ASD prevalence and burden are multifaceted and can result in complex, non-stationary data that may not be adequately captured through simple differencing.27 Third, the ARIMA model projected the ASD trends based on historical data in this study. The external factors have a very large impact on diseases such as advances in medical technology, changes in public health policies, changes in population structure, etc. Thus, the ARIMA model incorporates the exogenous variables, which may lead to the conformity from the actual situation in long-term predictions.28 Additionally, it is important to acknowledge the inherent limitations of the GBD database. While the GBD study integrates data from various sources to provide robust estimates, there are potential biases and uncertainties associated with data collection and reporting. For example, differences in data availability and quality across countries and regions may introduce variability in the reported prevalence and DALYs. These limitations should be considered when interpreting the results of this study.
Conclusion
Our research provides a comprehensive overview of the current and projected disease burden of ASD using extensive data from the GBD database, revealing a significant increase in the burden in China from 1990 to 2030 and key sex and age disparities in its impact. These findings underscore the growing public health challenge of ASD both in China and globally, emphasising the urgent need for sustained efforts to enhance diagnostic tools, expand research for improved data collection and implement targeted policies.
Supplementary material
Acknowledgements
We highly appreciate the works by the Global Burden of Diseases, Injuries, and Risk Factors Study 2021 collaborators. We also acknowledge the valuable insights and support from our colleagues, whose feedback significantly contributed to the refinement of this study.
Footnotes
Funding: This work was supported by the National Foreign Experts Project (number G2022011021L to LW) and the National Natural Science Foundation of China(U24A20770).
Provenance and peer review: Part of a topic collection; not commissioned; externally peer-reviewed.
Patient consent for publication: Not applicable.
Ethics approval: Not applicable.
Map disclaimer: The depiction of boundaries on this map does not imply the expression of any opinion whatsoever on the part of BMJ (or any member of its group) concerning the legal status of any country, territory, jurisdiction or area or of its authorities. This map is provided without any warranty of any kind, either express or implied.
Patient and public involvement: Patients and/or the public were not involved in the design, or conduct, or reporting, or dissemination plans of this research.
Data availability statement
All data relevant to the study are included in the article or uploaded as supplementary information.
References
- 1.Diagnostic and statistical manual of mental disorders, 5th edn. Arlington, VA: American Psychiatric Association; 2013. [Google Scholar]
- 2.Zeidan J, Fombonne E, Scorah J, et al. Global prevalence of autism: A systematic review update. Autism Res . 2022;15:778–90. doi: 10.1002/aur.2696. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 3.Salari N, Rasoulpoor S, Rasoulpoor S, et al. The global prevalence of autism spectrum disorder: a comprehensive systematic review and meta-analysis. Ital J Pediatr . 2022;48:112. doi: 10.1186/s13052-022-01310-w. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 4.Zhou H, Xu X, Yan W, et al. Prevalence of Autism Spectrum Disorder in China: A Nationwide Multi-center Population-based Study Among Children Aged 6 to 12 Years. Neurosci Bull . 2020;36:961–71. doi: 10.1007/s12264-020-00530-6. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 5.Cakir J, Frye RE, Walker SJ. The lifetime social cost of autism: 1990–2029. Res Autism Spectr Disord. 2020;72:101502. doi: 10.1016/j.rasd.2019.101502. [DOI] [Google Scholar]
- 6.Murray CJ, Acharya AK. Understanding DALYs (disability-adjusted life years) J Health Econ . 1997;16:703–30. doi: 10.1016/s0167-6296(97)00004-0. [DOI] [PubMed] [Google Scholar]
- 7.Vos T, Lim SS, Abbafati C, et al. Global burden of 369 diseases and injuries in 204 countries and territories, 1990–2019: a systematic analysis for the Global Burden of Disease Study 2019. Lancet. 2020;396:1204–22. doi: 10.1016/S0140-6736(20)30925-9. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 8.Murray CJL. The Global Burden of Disease Study at 30 years. Nat Med . 2022;28:2019–26. doi: 10.1038/s41591-022-01990-1. [DOI] [PubMed] [Google Scholar]
- 9.Liu BM, Kim HJ, Feuer EJ, et al. Joinpoint Regression Methods of Aggregate Outcomes for Complex Survey Data. J Surv Stat Methodol. 2023;11:967–89. doi: 10.1093/jssam/smac014. [DOI] [Google Scholar]
- 10.Yoneoka D, Nomura S, Tanaka S, et al. Prediction of disability-adjusted life years for diseases due to low fruit intake in 2017-2040 in Japan. Public Health Nutr . 2021;24:3156–66. doi: 10.1017/S1368980020004541. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 11.Gillis D, Edwards BPM. The utility of joinpoint regression for estimating population parameters given changes in population structure. Heliyon . 2019;5:e02515. doi: 10.1016/j.heliyon.2019.e02515. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 12.Sun X, Allison C, Wei L, et al. Autism prevalence in China is comparable to Western prevalence. Mol Autism. 2019;10:7. doi: 10.1186/s13229-018-0246-0. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 13.Milner V, McIntosh H, Colvert E, et al. A Qualitative Exploration of the Female Experience of Autism Spectrum Disorder (ASD) J Autism Dev Disord . 2019;49:2389–402. doi: 10.1007/s10803-019-03906-4. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 14.Baron-Cohen S, Lombardo MV, Auyeung B, et al. Why are autism spectrum conditions more prevalent in males? PLoS Biol . 2011;9:e1001081. doi: 10.1371/journal.pbio.1001081. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 15.Robinson EB, Lichtenstein P, Anckarsäter H, et al. Examining and interpreting the female protective effect against autistic behavior. Proc Natl Acad Sci U S A . 2013;110:5258–62. doi: 10.1073/pnas.1211070110. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 16.Jiang X, Chen X, Su J, et al. Prevalence of autism spectrum disorder in mainland china over the past 6 years: a systematic review and meta-analysis. BMC Psychiatry. 2024;24:404. doi: 10.1186/s12888-024-05729-9. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 17.Sun X, Allison C, Matthews FE, et al. Prevalence of autism in mainland China, Hong Kong and Taiwan: a systematic review and meta-analysis. Mol Autism . 2013;4:7. doi: 10.1186/2040-2392-4-7. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 18.Keehn B, Monahan P, Enneking B, et al. Eye-Tracking Biomarkers and Autism Diagnosis in Primary Care. JAMA Netw Open . 2024;7:e2411190. doi: 10.1001/jamanetworkopen.2024.11190. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 19.Oberman LM, Kaufmann WE. Autism Spectrum Disorder Versus Autism Spectrum Disorders: Terminology, Concepts, and Clinical Practice. Front Psychiatry. 2020;11:484. doi: 10.3389/fpsyt.2020.00484. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 20.Duque-Cartagena T, Dalla MDB, Mundstock E, et al. Environmental pollutants as risk factors for autism spectrum disorders: a systematic review and meta-analysis of cohort studies. BMC Public Health . 2024;24:2388. doi: 10.1186/s12889-024-19742-w. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 21.Grosvenor LP, Croen LA, Lynch FL, et al. Autism Diagnosis Among US Children and Adults, 2011-2022. JAMA Netw Open . 2024;7:e2442218. doi: 10.1001/jamanetworkopen.2024.42218. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 22.Sun J. Forecasting COVID-19 pandemic in Alberta, Canada using modified ARIMA models. Comput Methods Programs Biomed Update. 2021;1:100029. doi: 10.1016/j.cmpbup.2021.100029. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 23.Qi CY, Wang Y. Why Is Rehabilitation Assistance Policy for Children With Disabilities Deviated in Supply-Demand? A Case Study in Mainland China. Front Public Health. 2021;9:666333. doi: 10.3389/fpubh.2021.666333. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 24.WHO . Mental health atlas 2020. J Autism Dev Disord; 2020. [Google Scholar]
- 25.Lobanov-Rostovsky S, He Q, Chen Y, et al. Growing old in China in socioeconomic and epidemiological context: systematic review of social care policy for older people. BMC Public Health . 2023;23:1272. doi: 10.1186/s12889-023-15583-1. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 26.Global, regional, and national burden of 12 mental disorders in 204 countries and territories, 1990–2019: a systematic analysis for the Global Burden of Disease Study 2019. Lancet Psychiatry. 2022;9:137–50. doi: 10.1016/S2215-0366(21)00395-3. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 27.Schaffer AL, Dobbins TA, Pearson SA. Interrupted time series analysis using autoregressive integrated moving average (ARIMA) models: a guide for evaluating large-scale health interventions. BMC Med Res Methodol . 2021;21:58. doi: 10.1186/s12874-021-01235-8. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 28.Wang H, Yao R, Hou L, et al. A methodology for calculating the contribution of exogenous variables to arimax predictions. Proceedings of the Canadian Conference on Artificial Intelligence; 2021. [Google Scholar]