Key PointsQuestion
Are persons with autism spectrum disorder (ASD) or attention-deficit/hyperactivity disorder (ADHD) at a higher risk of dying compared with the general population?
Findings
In this systematic review and meta-analysis of 27 studies, persons with ASD or ADHD had higher mortality rates than the general population. When causes of death were examined, ASD and ADHD were associated with higher mortality due to unnatural causes (eg, injury, poisoning, and other), and only persons with ASD had an increased risk of mortality from natural causes of death (eg, neurologic, respiratory system, and cancer).
Meaning
Having ASD or ADHD may be associated with higher mortality risks.
Importance
Autism spectrum disorder (ASD) and attention-deficit/hyperactivity disorder (ADHD) are childhood-onset disorders that may persist into adulthood. Several studies have suggested that they may be associated with an increased risk of mortality; however, the results are inconsistent.
Objective
To assess the risk of mortality among persons with ASD or ADHD and their first-degree relatives.
Data Sources
A search of MEDLINE, Embase, Scopus, Web of Science, and PsycINFO (published from inception to April 1, 2021) was supplemented by searching reference lists of the retrieved articles.
Study Selection
Cohort and case-control studies that reported mortality rate ratios (RRs) in persons with ASD or ADHD and/or their first-degree relatives compared with the general population or those without ASD/ADHD were included.
Data Extraction and Synthesis
Screening, data extraction, and quality assessment were performed by at least 2 researchers independently. A random-effects model was used to meta-analyze individual studies and assessed heterogeneity (I2).
Main Outcomes and Measures
All-cause mortality in association with ASD or ADHD. Secondary outcome was cause-specific mortality.
Results
Twenty-seven studies were included, with a total of 642 260 individuals. All-cause mortality was found to be higher for persons with ASD (154 238 participants; 12 studies; RR, 2.37; 95% CI, 1.97-2.85; I2, 89%; moderate confidence) and persons with ADHD (396 488 participants; 8 studies; RR, 2.13; 95% CI, 1.13-4.02; I2, 98%; low confidence) than for the general population. Among persons with ASD, deaths from natural causes (4 studies; RR, 3.80; 95% CI, 2.06-7.01; I2, 96%; low confidence) and deaths from unnatural causes were increased (6 studies; RR, 2.50; 95% CI, 1.49-4.18; I2, 95%; low confidence). Among persons with ADHD, deaths from natural causes were not significantly increased (4 studies; RR, 1.62; 95% CI, 0.89-2.96; I2, 88%; low confidence), but deaths from unnatural causes were higher than expected (10 studies; RR, 2.81; 95% CI, 1.73-4.55; I2, 92%; low confidence).
Conclusions and Relevance
This systematic review and meta-analysis found that ASD and ADHD are associated with a significantly increased risk of mortality. Understanding the mechanisms of these associations may lead to targeted strategies to prevent avoidable deaths in high-risk groups. The substantial heterogeneity between studies should be explored further.
Autism spectrum disorder (ASD) and attention-deficit/hyperactivity disorder (ADHD) are common childhood onset neurodevelopmental disorders,1,2 with estimated worldwide prevalence of 58.6 million cases among children and young people 20 years and younger.2 Although the cause of ASD and ADHD remains largely unknown, a complex interaction of multiple factors is thought to contribute to the development of both conditions, generally persisting into adulthood.3-5 Both disorders have been found to be associated with psychosocial functional impairments and a range of adverse outcomes in patients and their families.4-9
Deaths in people with mental disorders have been examined in multiple epidemiologic studies10-12 suggesting that people with these disorders may experience a significant reduction in life expectancy, with increased mortality rates over the general population. Similarly, several studies13-19 have suggested that persons with ASD or ADHD may be associated with an increased risk of mortality; however, the results are inconsistent. For example, Pickett et al13 reported a 2-fold higher mortality rate ratio (RR) of 2.4 (95%, 2.2-2.8) in persons with ASD compared with those without it in the California Development Disability System. Bilder et al14 showed a mortality RR of 11.59 (95% CI, 6.24-21.53) in the Utah/University of California, Los Angeles Autism Epidemiologic Study. However, Smith et al15 suggested a nonstatistically significant increase in mortality (RR, 1.10; 95% CI, 0.50-2.50) in a large cohort study of school pupils with and without ASD. Conversely, Dalsgaard et al16 and Sun et al17 suggested increased mortality risk in persons with ADHD in large population-based cohorts from Denmark (fully adjusted RR, 2.07; 95% CI, 1.70-2.50) and Sweden (RR, 3.94; 95% CI, 3.51-4.43), respectively. London and Landes18 suggested similar results, with an adjusted RR of 1.78 (95% CI, 1.01-3.12). More recently, a nationwide population-based cohort study in Taiwan19 showed that having ADHD was associated with having a marginally increased risk of mortality after full adjustment (fully adjusted hazard ratio, 1.07; 95% CI, 1.00-1.17).
To our knowledge, no systematic review and meta-analysis of epidemiologic studies has specifically evaluated the mortality risk and main causes of death in persons with ASD or ADHD. A systematic review20 of studies published until 2010 examined ASD and 2 outcomes (epilepsy and mortality). Only 5 studies investigating mortality associations were located,20 so lack of data precluded exploring potential variations in all-cause mortality and quantifying cause-specific mortality risks. Therefore, in our study, we conducted a comprehensive knowledge synthesis to identify and evaluate all available observational studies of mortality associations in people with ASD or ADHD, which provide more complete information, and the most updated epidemiologic evidence than previous studies. Our systematic review and meta-analysis of epidemiologic studies for ASD and ADHD has only become possible in the past few years because several large population-based observational studies16,17,19,21 have been published that have reported similar mortality-related outcome measures. In this study, we aimed to evaluate the risk of both all-cause and cause-specific mortality among persons with ASD or ADHD and their first-degree relatives.
Our systematic review was reported in accordance with the 2020 Preferred Reporting Items for Systematic Reviews and Meta-analyses (PRISMA) reporting guideline.22 We also reported this study in accordance with the guidance provided in the Meta-analysis of Observational Studies in Epidemiology (MOOSE) reporting guideline23 (eTables 1 and 2 in the Supplement). We developed and registered a review protocol (PROSPERO registration number: CRD42017059955). Our methods are briefly described here (and explained in more detail in the protocol24 and in eTables 1-12 in the Supplement).
Information Sources and Search Strategy
An information specialist (A.A.-A.) searched MEDLINE (via PubMed), Scopus, Embase, Web of Science Core Collection, and PsycINFO (via ProQuest) to identify all relevant observational studies in humans that examined the risk of mortality among persons with ASD or ADHD, published from inception to April 1, 2021, without language restrictions. The search strategies for all electronic databases are presented in the eTable 5 in the Supplement. References of all relevant primary studies and review articles were also screened to identify additional data sources.
To be included, primary studies had to be observational (cohort, case-control) studies of persons with ASD or ADHD according to standard operationalized diagnostic criteria (eg, International Classification of Diseases, Ninth Revision [ICD-9] or Tenth revision [ICD-10] or third, fourth, or fifth editions of the Diagnostic and Statistical Manual of Mental Disorders criteria; ASD, ICD-9 codes 299.0 and 299.8 and ICD-10 code F84; ADHD, ICD-9 codes 314.00 and 314.01 and ICD-10 code F90) and report the mortality RR for the risk of mortality in people with ASD/ADHD compared with the general population or those without ASD/ADHD, or have enough data (eg, number of cases and sample size; observed and expected cases) to compute these estimates. We excluded studies in which ASD or ADHD were not the exposure of interest and mortality was not reported as the outcome. Studies not presenting study-specific data or sufficient data for an outcome measure to be calculated were also excluded. We excluded case reports, case series, in vitro studies, and animal studies.
The primary outcome was all-cause mortality (ICD-10 codes A00-Y98 and ICD-9 codes 001-E999). Cause-specific mortality rates were evaluated as a secondary outcome of interest (eTable 3 in the Supplement).
Screening and Selection Process
Two investigators (F.C.-L., D.M.S.-G., and/or M.R.) independently screened all titles and abstracts and retrieved the full text of any article considered definitely or possibly eligible. The same investigators reviewed the full-text articles against the eligibility criteria. Any disagreement between these investigators was resolved by discussion.
Data Extraction and Quality Assessment
Two investigators (F.C.-L., D.M.S.-G., and/or M.R.) independently extracted relevant information from the included studies: first author, publication year and period of recruitment, country, study design, setting, coverage, mean or median age (or age range), proportion of women participants, ethnicity, comorbidities, social status, diagnostic criteria and ascertainment of outcome, consideration of confounding factors, the number of participants with fatal outcome, and maximally adjusted RR estimates with 95% CIs. When relevant fatal outcome data were not available, we directly contacted the corresponding author of the study to request the information (eTable 7 in the Supplement). Two pairs of investigators (F.C.-L., M.J.P., D.M.S.-G., and M.R.) independently undertook methodologic quality assessment of included studies using the Newcastle-Ottawa scale25 and allocated stars for adherence to the prespecified criteria. This scale ranges from 0 (lowest quality) to 9 (highest quality) stars and judges each study regarding selection of study groups, comparability, and ascertainment of the outcome. We considered studies with 0 to 3, 4 to 6, and 7 to 9 stars to represent high, moderate, and low risk of bias, respectively. Discrepant scores were resolved by discussion among investigators.
To measure the association between the mortality in people with ADHD or ASD and the mortality in the reference control group, we did meta-analysis using the inverse variance random effects model26 (with the DerSimonian-Laird between-study variance estimator) to pool weighted all-cause and specific-cause mortality RR estimates. We explored the contributions from natural and unnatural causes of death. Unnatural deaths were defined based on ICD codes (ICD-10 codes V01-Y98 or ICD-9 codes E800-999); the remaining deaths were classified as natural. All meta-analyses were conducted for ADHD and ASD separately. We assessed heterogeneity between studies using the P value of Cochran Q test27 and the I2 statistic (with 95% CIs),28 which could reflect either genuine heterogeneity or bias. The I2 statistic ranges between 0% and 100% (with values of 0%-25% and 75%-100% taken to indicate low and considerable heterogeneity, respectively). We also calculated 95% prediction intervals, which further account for heterogeneity between studies and indicate the uncertainty for the effect that would be expected in a new study examining that same association.29 To assess the robustness of pooled results and explore possible reasons for heterogeneity, prespecified subgroup analyses24 were performed according to sex (male or female), number of comorbidities (including 0, 1, 2, or any), age group at first diagnosis of ASD or ADHD (prior to age 18 years or at 18 years or older), number of participants with ASD or ADHD (<500 vs 500-1000 vs >1000), setting (community, inpatient care, or outpatient care), follow-up (less than 1 year, 1-5 years, >5 years), study quality (low vs moderate vs high risk of bias), and statistical adjustment for potential confounders (yes vs no). Small study effects (the tendency for exposure effects estimated in smaller studies to differ from those estimated in larger studies, which can result from reporting biases, methodologic heterogeneity, epidemiologic heterogeneity, or other factors) was estimated visually by funnel plots or by Begg test and the weighted regression test of Egger. P values were determined using the Woolf method. P values were 2-tailed, and statistical significance was set at less than .05. We performed all data analyses using Stata version 17 (StataCorp).
We applied a set of criteria to classify the credibility (or certainty) of the evidence from meta-analysis based on the Global Burden of Disease30 and GRADE system31 (eTable 10 in the Supplement).
With the publication of this article, the full data set will be freely available online in the Open Science Framework,32 a secure online repository for research data. Amendments to the original protocol are listed in the eTable 4 in the Supplement.
Extent of Relevant Literature Identified
We screened 2541 titles and abstracts, followed by 68 full-text articles (Figure 1). Twenty-seven epidemiologic studies published in 32 articles13-19,21,33-56 met our inclusion criteria.
Description of Studies and Participants
Table 1 and eTable 8 in the Supplement summarize the characteristics of the included studies. All studies were published in English from 1988 through 2021. Twenty-three studies13-18,33-42,44-54,56 were from North America and Europe (11 in the US,13,14,18,33-37,41,42,46,50,51,53 5 in Denmark,16,38,39,44,49,52,56 3 in Sweden,17,40,47 2 in Finland,45,54 and 2 in the UK15,48). Seventeen studies were retrospective cohorts,13-16,18,21,33-39,43,44,48,50,52-56 8 were prospective cohorts,17,19,40-42,45,46,49 and 2 were retrospective case-control studies.47,51 Fourteen studies13-15,21,34,35,38-40,47-50,53-56 included 206 162 participants with ASD, and 12 studies16-19,33,36,37,41,42,44-46,51,52 included 433 761 participants with ADHD. The number of participants with ASD ranged from 120 to 35 929, and the number of participants with ADHD ranged from 110 to 275 980. The proportion of female participants ranged from 14% to 100%. Median (range) follow-up was 16 (3-33) years. Twelve studies15,21,33,36,37,40,42,46,48,50,52,55 followed up participants to young adult age (mean age, <40 years). Only 3 studies14,36,37,43 included first-degree relatives of people with ADHD or ASD diagnosis.
There were 49 individual study estimates for all-cause mortality in persons with ASD or ADHD and/or their first-degree relatives (eTable 9 in the Supplement). Seventeen studies14,16-19,33,38-40,47-49,51,56 provided adjusted RR estimates to control for several covariates (eg, age, sex, calendar year, or other). Twenty-one studies13,14,16-19,21,33-40,42-47,49-52,54,56 reported information on 12 cause-specific mortality outcomes. The most studied ICD-10 categories were external causes of mortality (such as injuries, unintentional incidents, suicides, and poisoning), accounting for 1215 cases (61.2%) in 21 studies13,14,16,17,19,21,33-40,42-47,49-52,54,56; followed by diseases of the circulatory system, accounting for 190 cases (9.6%) in 8 studies14,17,21,38-40,43,45,47 and neoplasms, accounting for 168 cases (8.5%) in 7 studies13,14,17,21,34,35,40,43,47 (eFigure 23 in the Supplement). Infectious diseases accounted 8 cases (0.4%) in 4 studies14,38-40,47 and complications of pregnancy, childbirth, and puerperium only accounted 1 case (0.1%) in 1 study.43 Regarding the methodologic quality (eTable 10 in the Supplement), 16 studies13,14,16,17,19,21,33-35,38-40,42,44,47-49,52,55,56 (59%) had low risk of bias and 11 studies15,18,36,37,41,43,45,46,49,51,53,54 had moderate risk of bias (Newcastle-Ottawa Scale values, 4-9).
Main Findings From Meta-analyses
Our all-cause meta-analyses focused the risk of dying in persons with ASD or ADHD compared with the general population and included 21 studies13-19,21,34-42,44-49,53-55 that contributed 46 data points (Table 2; eFigure 23 and eTable 9 in the Supplement). Thirty-four of 46 all-cause mortality RR estimates (74%) suggested an increased risk of mortality in individuals with ASD or ADHD. Among participants with ASD, all-cause mortality was increased in both sexes combined (12 studies13-15,21,34,35,38-40,47-49,53-55; RR, 2.37; 95% CI, 1.97-2.85; I2, 89%; moderate confidence) and separately for male individuals (8 studies13,14,34,35,38-40,47,49,53,54; RR, 2.09; 95% CI, 1.50-2.92; I2, 94%; low confidence) and female individuals (8 studies13,14,34,35,38-40,47,49,53,54; RR, 4.87; 95% CI, 3.07-7.73; I2, 91%; low confidence). Among persons with ADHD, all-cause mortality was also increased in both sexes combined (8 studies16-19,36,37,42,44-46; RR, 2.13; 95% CI, 1.13-4.02; I2, 98%; low confidence), and separately for male individuals (5 studies16,17,19,41,44,45; RR, 2.43; 95% CI, 1.01-5.83; I2, 99%; low confidence) and female individuals (4 studies16,17,19,44,45; RR, 2.84; 95% CI, 1.02-7.89; I2, 97%; low confidence) (Table 2 and Figure 2 and Figure 3). Meta-analysis for all-cause mortality among first-degree relatives was not possible, with only 3 studies reporting outcome data (1 study14 in siblings of children with ASD, 1 study43 in mothers of children with ASD, and 1 study36,37 in parents of children with ADHD).
Secondary outcome meta-analyses focused on the cause-specific mortality RR in persons with ASD or ADHD compared with the general population and included 17 studies13,16,17,19,21,33-39,42,44-47,49,51,52,54,56 that contributed 89 data points (Table 2 and eTable 9 and eFigures 1-7 in the Supplement). A total of 67 of 87 cause-specific mortality RR estimates (75%) were increased. Among persons with ASD, deaths from natural causes (4 studies13,34,35,47,49,54; RR, 3.80; 95% CI, 2.06-7.01; I2, 96%; low confidence) and deaths from unnatural causes were increased (6 studies13,34,35,47,49,50,54,56; RR, 2.50; 95% CI, 1.49-4.18; I2, 95%; low confidence). Among persons with ADHD, deaths from natural causes were not significantly increased (4 studies16,17,19,44,45; RR, 1.62; 95% CI, 0.89-2.96; I2, 88%; low confidence), but deaths from unnatural causes were higher than expected (10 studies16,17,19,33,36,37,42,44-46,51,52; RR, 2.81; 95% CI, 1.73-4.55; I2, 92%; low confidence).
The full details of the additional analyses are reported in eFigures 1 to 23 in the Supplement. Statistical heterogeneity was not substantially reduced when analyses were stratified by prespecified (exploratory) subgroup analyses such as number of comorbidities, age group at first diagnosis, number of participants (sample size), setting, follow-up period, study quality, and adjustment for potential confounders. Overall, the I2 statistic indicated that data were heterogeneous in many of the pooled analyses, and therefore, these summary measures must be interpreted with appropriate caution. Insufficient data were available to do preplanned subgroup analyses24 by diagnostic criteria, ethnicity, geographic region, or year of publication as there were not enough data. We undertook a funnel plot, and Begg and Egger tests of the 12 studies examining all-cause mortality in people with ASD, with results showing no clear evidence of small study effects (eFigure 22 in the Supplement).
This systematic review and meta-analysis comprehensively assess for the first time, to our knowledge, the available evidence regarding the risk of mortality in persons with ASD or ADHD. We included 27 epidemiologic studies in our quantitative evaluation, 16 of which were judged to be at low risk of bias. We found that ASD and ADHD are associated with a significantly increased risk of all-cause mortality. However, the results should be interpreted with caution because there was evidence of heterogeneity between study estimates of the mortality risks. When we examined causes of death, ASD and ADHD were associated with higher mortality risk due to unnatural (external) causes, and only persons with ASD had an increased risk of mortality from natural causes of death, but the evidence was judged as only low confidence. Fewer studies exist examining the risk of mortality among first-degree relatives of persons with ASD or ADHD, to our knowledge.
Several mechanisms, including health determinants and biological pathways, have been suggested as potential factors that might explain the excess premature mortality among children and young persons with ASD or ADHD. However, establishing a causal relationship is difficult because the associations between mortality and childhood-onset developmental disorders are complex. Severe mental and behavioral disorders appear to be associated with reduced life expectancy, both in terms of mortality from external causes and mortality from other medical conditions or diseases.10,11 Findings from previous studies and reviews have suggested that children and adults with ASD/ADHD are associated with coexisting mental and neurologic conditions (such as oppositional and conduct disorders, tic disorders, epilepsy, depression, anxiety, and substance use disorders).3-5,57 As children and young persons with ASD/ADHD age, they often experience emotional and social difficulties.3-5 Some people also exhibit impulsive forms of behavior with negative impacts on their quality of life. Behaviors such as impulsivity and/or inattention can be contributing factors for injuries and unintentional incidents in children with ASD/ADHD.58,59
Previous studies16,17,49 have tested potential modifying effects of comorbidity on mortality risk for persons with ASD/ADHD. For example, Dalsgaard et al,16 Sun et al,17 and Schendel et al49 suggested increased mortality RRs in persons with ADHD or ASD with comorbid neurologic or mental conditions (eg, oppositional defiant disorder, conduct disorder, or substance use disorder). However, it should be noted that presence of ASD and/or ADHD with any distinct additional medical condition (the so-called comorbidity or multimorbidity) may be confounded by previous exposures (such as socioeconomic factors, environmental factors, and childhood abuse).60 Prevention efforts to reduce mortality in persons with ASD or ADHD may need to address the conditions that appear to mediate causes of death. While potentially preventable, reducing excess premature mortality (eg, due to external causes) can be challenging. For example, systematic screening would be advisable in health services and social care, and preventive education can be feasible in almost all circumstances. However, some persons with ADHD/ASD are often from socioeconomically disadvantaged groups/areas and are more likely to be exposed to environmental risk factors (eg, substance misuse, violence) than other populations.61-63 Similarly, studies64,65 have shown that persons with ADHD or ASD may be less likely to receive timely diagnosis and prompt care. All these factors are possible complications of ASD/ADHD, so the implication for practice is that health care professionals should recognize their importance.
Strengths and Limitations
This study had several strengths and limitations. We used established methods to leverage a large number of records identified from each database searched and contacted study authors to identify relevant outcome data. However, our search strategy could have omitted abstracts that did not state all-cause or cause-specific mortality as an included outcome. Observational studies in this area are increasingly conducted within large routinely collected electronic health databases, sometimes with hundreds of thousands of health care users. These studies generate precise estimates of exposure effects, which sometimes may be inaccurate because of potential residual confounding. For example, electronic health records data may have limitations when used for research, including limited information on lifestyle and socioeconomic determinants of health. In this systematic review, there was heterogeneity in the included studies in terms of the design, settings, populations, and outcomes. Some of the included studies have a restricted age of follow-up (eg, individuals younger than 40 years). We used study-level data instead of individual participant data, so the small number of studies (and events) for most cause-specific mortality estimates limited the subgroup analyses that could be conducted to account for potential sources of heterogeneity. Furthermore, our results might have limited generalizability because the studies were mostly conducted in Western countries. Although we adopted reproducible definitions, our meta-analyses were challenged by the reporting of events in studies. None of the primary studies examined and reported all the fatal outcomes of interest. Also, incomplete reporting is common in the literature and limited our possibilities to optimally synthesize the data for some specific causes of death. Future studies should adopt rigorous protocols when studying the effects of ASD/ADHD on health, which should clearly define all outcomes and exposures, how the study population was selected, how data were collected, and that all relevant potential confounders should be accounted for. Reporting guidelines, such as the Strengthening the Reporting of Observational Studies in Epidemiology (STROBE) reporting guideline,66 and the Reporting of Studies Conducted Using Observational Routinely-Collected health Data (RECORD) statement,67 should be rigorously implemented for individual study reports. Finally, small study effects were not quantitatively assessed for most of the outcomes because there were inadequate numbers of included studies to properly assess a funnel plot or use more sophisticated methods.
We found suggestive evidence that ASD and ADHD are associated with a significantly increased risk of mortality. Understanding the mechanisms of these associations may lead to targeted strategies to prevent avoidable deaths in high-risk groups of children and young people as an approach to improve public health. For example, clinicians and health care professionals can be encouraged to routinely collect information on behavioral, medical conditions, and health outcomes related to ASD/ADHD, emphasizing the need to recognize and address modifiable vulnerability factors and prevent delays in health care provision. Additionally, we hope that these estimates can shed some light for future studies related to examining mortality-related health estimates in persons with ASD or ADHD.
Corresponding Author: Ferrán Catalá-López, PhD, Department of Health Planning and Economics, National School of Public Health, Institute of Health Carlos III, Monforte de Lemos 5, 28029 Madrid, Spain (ferran_catala@outlook.com).
Accepted for Publication: October 27, 2021.
Published Online: February 14, 2022. doi:10.1001/jamapediatrics.2021.6401
Open Access: This is an open access article distributed under the terms of the CC-BY License. © 2022 Catalá-López F et al. JAMA Pediatrics.
Author Contributions: Dr Catalá-López had full access to all of the data in the study and takes responsibility for the integrity of the data and the accuracy of the data analysis.
Concept and design: Catalá-López, Page, Driver, Valencia, Macías Saint-Gerons, Tabares-Seisdedos.
Acquisition, analysis, or interpretation of data: Catalá-López, Hutton, Ridao, Alonso-Arroyo, Macías Saint-Gerons, Tabares-Seisdedos.
Drafting of the manuscript: Catalá-López, Macías Saint-Gerons, Tabares-Seisdedos.
Critical revision of the manuscript for important intellectual content: Hutton, Page, Driver, Ridao, Alonso-Arroyo, Valencia, Tabares-Seisdedos.
Statistical analysis: Catalá-López.
Obtained funding: Catalá-López.
Administrative, technical, or material support: Hutton, Ridao, Alonso-Arroyo, Macías Saint-Gerons.
Supervision: Catalá-López, Driver, Valencia, Tabares-Seisdedos.
Conflict of Interest Disclosures: Dr Valencia reported grants from the Spanish Ministry of Science and Technology during the conduct of the study and grants from Spanish Ministry of Science, Spanish Ministry of Economy, European Commission Framework Program, IBM Research, Generalitat de Catalunya, and EuroHPC outside the submitted work. No other disclosures were reported.
Funding/Support: This study was supported by the Institute of Health Carlos III and Generalitat Valenciana. Drs Catalá-López and Tabarés-Seisdedos received funding from the Centro de Investigación Biomédica en Red de Salud Mental, Institute of Health Carlos III, and Generalitat Valencia. Dr Page received support from an Australian Research Council Discovery Early Career Researcher Award. Dr Hutton received support from a new investigator award from the Canadian Institutes of Health Research and the Drug Safety and Effectiveness Network. Dr Ridao received support from the Spanish Health Services Research on Chronic Patients Network and Institute of Health Carlos III.
Role of the Funder/Sponsor: Institute of Health Carlos III had no role in the design and conduct of the study; collection, management, analysis, and interpretation of the data; preparation, review, or approval of the manuscript; and decision to submit the manuscript for publication.
Disclaimer: The views expressed in this article are the views of the authors and may not be understood or quoted as being made on behalf of, or reflecting the position of, the funder(s) or any institution.
Additional Contributions: We would like to thank Jane A. Pickett, PhD (Autism BrainNet); Deborah Bilder, MD (Department of Psychiatry, University of Utah); Andrew S. London, PhD (Department of Sociology, Syracuse University); Søren Dalsgaard, PhD (National Centre for Register-based Research (NCRR), Department of Economics and Business Economics, Aarhus University); Shihua Sun, PhD (Department of Medical Epidemiology and Biostatistics, Karolinska Institutet), and the study investigators for their clarifications and/or providing their unpublished data for use in this work. We thank Eduard Vieta, PhD (Department of Psychiatry and Psychology, Institute of Neuroscience, Hospital Clínic, University of Barcelona), and Jaume Forés, PhD (Department of Medicine, University of Valencia/INCLIVA Health Research Institute and Centro de Investigación en Red de Salud Mental), for providing their feedback on our study protocol. Compensation was not received.
1.Kassebaum
N, Kyu
HH, Zoeckler
L,
et al; Global Burden of Disease Child and Adolescent Health Collaboration. Child and adolescent health from 1990 to 2015: findings from the Global Burden of Diseases, Injuries, and Risk Factors 2015 Study.
JAMA Pediatr. 2017;171(6):573-592. doi:
10.1001/jamapediatrics.2017.0250PubMedGoogle Scholar 2.GBD 2019 Diseases and Injuries Collaborators. 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(10258):1204-1222. doi:
10.1016/S0140-6736(20)30925-9PubMedGoogle ScholarCrossref 7.Erskine
HE, Norman
RE, Ferrari
AJ,
et al. Long-term outcomes of attention-deficit/hyperactivity disorder and conduct disorder: a systematic review and meta-analysis.
J Am Acad Child Adolesc Psychiatry. 2016;55(10):841-850. doi:
10.1016/j.jaac.2016.06.016PubMedGoogle ScholarCrossref 8.Nourredine
M, Gering
A, Fourneret
P,
et al. Association of attention-deficit/hyperactivity disorder in childhood and adolescence with the risk of subsequent psychotic disorder: a systematic review and meta-analysis.
JAMA Psychiatry. 2021;78(5):519-529. doi:
10.1001/jamapsychiatry.2020.4799PubMedGoogle ScholarCrossref 9.Drmic
IE, Szatmari
P, Volkmar
F. Life course health development in autism spectrum disorders. In: Halfon
N, Forrest
CB, Lerner
RM, Faustman
EM, eds.
Handbook of Life Course Health Development. Springer; 2018. doi:
10.1007/978-3-319-47143-3_11 23.Stroup
DF, Berlin
JA, Morton
SC,
et al. Meta-analysis of observational studies in epidemiology: a proposal for reporting: Meta-analysis Of Observational Studies in Epidemiology (MOOSE) group.
JAMA. 2000;283(15):2008-2012. doi:
10.1001/jama.283.15.2008PubMedGoogle ScholarCrossref 24.Catalá-López
F, Hutton
B, Page
MJ,
et al. Risk of mortality among children, adolescents, and adults with autism spectrum disorder or attention deficit hyperactivity disorder and their first-degree relatives: a protocol for a systematic review and meta-analysis of observational studies.
Syst Rev. 2017;6(1):189. doi:
10.1186/s13643-017-0581-9PubMedGoogle ScholarCrossref 30.GBD 2016 Risk Factors Collaborators. Global, regional, and national comparative risk assessment of 84 behavioural, environmental and occupational, and metabolic risks or clusters of risks, 1990-2016: a systematic analysis for the Global Burden of Disease Study 2016.
Lancet. 2017;390(10100):1345-1422. doi:
10.1016/S0140-6736(17)32366-8PubMedGoogle ScholarCrossref 32.Catalá-López F, Hutton B, Page MJ, et al. Risk of mortality among persons with autism spectrum disorder or attention deficit hyperactivity disorder and their first-degree relatives: a systematic review and meta-analysis of observational studies. OSF. Updated October 17, 2021. Accessed January 10, 2022. doi:
10.17605/OSF.IO/2NJUD 35.Shavelle
RM, Strauss
D. Comparative mortality of persons with autism in California, 1980-1996.
J Insur Med. 1998;30(4):220-225.
PubMedGoogle Scholar 36.Barkley
RA, Murphy
KR, Fischer
M. ADHD in Adults: What the Science Says. The Guilford Press; 2008.
42.Barbaresi
WJ, Colligan
RC, Weaver
AL, Voigt
RG, Killian
JM, Katusic
SK. Mortality, ADHD, and psychosocial adversity in adults with childhood ADHD: a prospective study.
Pediatrics. 2013;131(4):637-644. doi:
10.1542/peds.2012-2354PubMedGoogle ScholarCrossref 43.Fairthorne
J, Hammond
G, Bourke
J, Jacoby
P, Leonard
H. Early mortality and primary causes of death in mothers of children with intellectual disability or autism spectrum disorder: a retrospective cohort study.
PLoS One. 2014;9(12):e113430. doi:
10.1371/journal.pone.0113430PubMedGoogle Scholar 46.Hechtman
L, Swanson
JM, Sibley
MH,
et al; MTA Cooperative Group. Functional adult outcomes 16 years after childhood diagnosis of attention-deficit/hyperactivity disorder: MTA results.
J Am Acad Child Adolesc Psychiatry. 2016;55(11):945-952.e2. doi:
10.1016/j.jaac.2016.07.774PubMedGoogle ScholarCrossref 53.Akobirshoev
I, Mitra
M, Dembo
R, Lauer
E. In-hospital mortality among adults with autism spectrum disorder in the United States: a retrospective analysis of US hospital discharge data.
Autism. 2020;24(1):177-189. doi:
10.1177/1362361319855795PubMedGoogle ScholarCrossref 57.Hossain
MM, Khan
N, Sultana
A,
et al. Prevalence of comorbid psychiatric disorders among people with autism spectrum disorder: an umbrella review of systematic reviews and meta-analyses.
Psychiatry Res. 2020;287:112922. doi:
10.1016/j.psychres.2020.112922PubMedGoogle Scholar 61.Roman-Urrestarazu
A, van Kessel
R, Allison
C, Matthews
FE, Brayne
C, Baron-Cohen
S. Association of race/ethnicity and social disadvantage with autism prevalence in 7 million school children in England.
JAMA Pediatr. 2021;175(6):e210054. doi:
10.1001/jamapediatrics.2021.0054PubMedGoogle Scholar 62.Russell
AE, Ford
T, Williams
R, Russell
G. The association between socioeconomic disadvantage and attention deficit/hyperactivity disorder (ADHD): a systematic review.
Child Psychiatry Hum Dev. 2016;47(3):440-458. doi:
10.1007/s10578-015-0578-3PubMedGoogle ScholarCrossref 66.von Elm
E, Altman
DG, Egger
M, Pocock
SJ, Gøtzsche
PC, Vandenbroucke
JP; STROBE Initiative. The Strengthening the Reporting of Observational Studies in Epidemiology (STROBE) statement: guidelines for reporting observational studies.
Lancet. 2007;370(9596):1453-1457. doi:
10.1016/S0140-6736(07)61602-XPubMedGoogle ScholarCrossref 67.Benchimol
EI, Smeeth
L, Guttmann
A,
et al; RECORD Working Committee. The Reporting of Studies Conducted Using Observational Routinely-Collected Health Data (RECORD) statement.
PLoS Med. 2015;12(10):e1001885. doi:
10.1371/journal.pmed.1001885PubMedGoogle Scholar