RESEARCH ARTICLE Temporal and Geographical Variability of Prevalence and Incidence of Autism Spectrum Disorder Diagnoses in Children in Catalonia, Spain Laura Pérez-Crespo, Albert Prats-Uribe, Aurelio Tobias, Enric Duran-Tauleria, Ricard Coronado, Amaia Hervás, and Mònica Guxens This study aims to estimate the prevalence of autism spectrum disorders (ASD) in 2017 and the ASD diagnosis incidence between 2009 and 2017 in children living in Catalonia region in Spain, and their temporal and geographical variability. We used administrative data for all children aged 2–17 years who were insured in the public Catalan Health System between 2009 and 2017. We identified all ASD cases diagnosed between 2009 and 2017 (ICD-9 codes 299.0, 299.1, 299.8, and 299.9). We esti- mated the ASD prevalence in 2017 and the overall annual incidence between 2009 and 2017, then stratified by sex, age group, and healthcare area. We used Poisson regression models to assess temporal trends in the incidence and mixed-effects Poisson regression models to assess geographical variability. We observed an ASD prevalence of 1.23% (95% confidence interval [CI] 1.21–1.25) in 2017, with 1.95% (95% CI 1.92–1.99) for boys and 0.46% (95% CI 0.44–0.48) for girls, the highest prevalence being in 11- to 17-year-olds (1.80%, 95% CI 1.76–1.83). The ASD diagnosis incidence increased from 0.07% (95% CI 0.06–0.09) in 2009 to 0.23% (95% CI 0.21–0.24) in 2017, with a higher increase in girls, and in children aged 2–5 years at the time of diag- nosis. We only observed geographical differences in prevalence in the 2017 data. We also detected a threefold increase in the diagnosis incidence overall, which was even more pronounced in girls and at early ages. In conclusion, the ASD prevalence observed in this study was 1.23% in 2017, with a sex ratio of 4.5 in favor of boys, which is consistent with previous studies. Autism Res 2019, 12: 1693–1705. © 2019 The Authors. Autism Research published by International Society for Autism Research published by Wiley Periodicals, Inc. Lay Summary: Autism spectrum disorders (ASD) are currently well known in our society as one of the most common neu- rodevelopmental disorders during childhood. The results of our study showed that, in 2017 in Catalonia, slightly more than one in a 100 children had an ASD diagnosis, it was more common in boys than in girls, and also in older children. In addi- tion, between 2009 and 2017, we observed an increase in the number of new cases diagnosed each year. The data presented in this study will assist in planning and evaluating the needs of health services in this geographical region. Keywords: autistic disorder; child development disorder; neurodevelopmental disorders; epidemiology; Europe Introduction while others have a more severe form that interferes with their quality of life [Farley et al., 2009; Hervas, 2016]. Addi- Autism spectrum disorders (ASD) are a group of neu- tionally, individuals with ASD have a very high prevalence of rodevelopmental conditions caused by atypical brain devel- comorbidities, which could explain the complex, and still opment that begins during prenatal or early postnatal life not fully understood, etiology of the disorder [Murray et al., [Zachor & Curatolo, 2014]. ASDs are commonly character- 2014]. Although considerable effort has been devoted to the ized by impaired social interaction, deficits in social commu- etiological factors of ASD, no single cause has been identified. nication, and restricted, repetitive, and stereotyped behavior The available data suggest that this condition results from patterns [American Psychiatric Association, 2013]. The symp- different sets of causal factors, including genetic and environ- toms are clinically heterogeneous, and some patients are able mental risk factors [Huguet, Ey, & Bourgeron, 2013; to have an independent and fulfilling life with little support, Modabbernia, Velthorst, & Reichenberg, 2017]. From the ISGlobal, Barcelona, Spain (L.Pérez-Crespo., M.G.); Pompeu Fabra University, Barcelona, Spain (L.Pérez-Crespo., M.G.); Spanish Consortium for Research on Epidemiology and Public Health (CIBERESP), Instituto de Salud Carlos III, Madrid, Spain (L.Pérez-Crespo., M.G.); Preventive Medicine and Public Health Training Unit, Parc de Salut Mar—Pompeu, Fabra University—Public Health Agency of Barcelona, Barcelona, Spain (A.P.-U.); Centre for Sta- tistics in Medicine, Botnar Research Centre, NDORMS, University of Oxford, Oxford, UK (A.P.-U.); Institute of Environmental Assessment and Water Research (IDAEA), Spanish Council for Scientific Research (CSIC), Barcelona, Spain (A.T.); Institut Global d’Atenció Integral del Neurodesenvolupament (IGAIN), Barcelona, Spain (E.D.-T., A.H.); Hospital General de Granollers, Granollers, Spain (R.C.); Department of Pediatrics, Obstetrics and Gynecol- ogy and Preventive Medicine and Public Health, Autonomous University of Barcelona, Barcelona, Spain (R.C.); Child and Adolescent Mental Health Unit, Hospital Universitari Mútua de Terrassa, Barcelona, Spain (A.H.); Department of Child and Adolescent Psychiatry/Psychology, Erasmus University Medi- cal Centre–Sophia Children’s Hospital, Rotterdam, The Netherlands (M.G.) Received January 16, 2019; accepted for publication June 27, 2019 Address for correspondence and reprints: Mònica Guxens, Barcelona Institute for Global Health (ISGlobal), Campus Mar, Carrer Dr. Aiguader 88, 08003 Barcelona, Spain. E-mail: monica.guxens@isglobal.org This is an open access article under the terms of the Creative Commons Attribution License, which permits use, distribution and reproduction in any medium, provided the original work is properly cited. Published online 17 July 2019 in Wiley Online Library (wileyonlinelibrary.com) DOI: 10.1002/aur.2172 © 2019 The Authors. Autism Research published by International Society for Autism Research published by Wiley Periodicals, Inc. INSAR Autism Research 12: 1693–1705, 2019 1693 ASD are a major public health problem due to the world- although few studies have assessed ASD incidence, they wide increase in their prevalence in recent decades reaching have all shown an increased incidence in the last years [Bar- values of between 0.01% in Oman [Al-Farsi et al., 2011] and baresi, Katusic, Colligan, Weaver, & Jacobsen, 2005; Chien, 2.64% in South Korea [Kim et al., 2011]. The observed differ- Lin, Chou, & Chou, 2011; Davidovitch et al., 2013; ences between studies might be due to methodological, cul- Hinkka-Yli-Salomäki et al., 2014; Honda, Shimizu, Misumi, tural, and environmental factors [Hansen, Schendel, & Niimi, & Ohashi, 1996; Jensen, Steinhausen, & Lauritsen, Parner, 2015; Lyall et al., 2017]. Therefore, it is difficult to 2014; Lauritsen, Pedersen, & Mortensen, 2004; Nassar compare estimates of ASD prevalence and incidence from et al., 2009; Pelly et al., 2015; Plubrukarn, Piyasil, different studies because they may not have been conducted Moungnoi, Tanprasert, & Chutchawalitsakul, 2005; Powell homogeneously. ASD epidemiological studies have mainly et al., 2000; Raz, Weisskopf, Davidovitch, Pinto, & Levine, been based on registers or on multistage screening 2015; Taylor, Jick, & MacLaughlin, 2013; Williams et al., approaches. Most studies have been performed in America 2005; Wong & Hui, 2008]. As for prevalence, the incidence and Europe, whereas less data is available for the Western varies notably between countries and regions; the inci- Pacific, South East Asia, and Eastern Mediterranean regions dence in Europe and North America is approximately [Chaaya, Saab, Maalouf, & Boustany, 2016; Chinawa et al., 0.15% [Davidovitch et al., 2013; Diallo et al., 2018; Jensen 2017; Davidovitch, Hemo, Manning-Courtney, & et al., 2014; Pelly et al., 2015; Raz et al., 2015; Taylor et al., Fombonne, 2013; Dekkers, Groot, Díaz Mosquera, Andrade 2013], while that in South Asia and Western Pacific regions Zúñiga, & Delfos, 2015; Elsabbagh et al., 2012; Fombonne, is ~0.05% [Chien et al., 2011; Plubrukarn et al., 2005; Wil- 2009b, 2009a; Fortea Sevilla, Escandell Bermúdez, & Castro liams et al., 2005; Wong & Hui, 2008]. Sánchez, 2013; Hewitt et al., 2016; Hill, Zuckerman, & Overall, very few studies have been carried out in South- Fombonne, 2014; Huang et al., 2014; Idring et al., 2015; ern European countries, particularly in our region, where lit- Isaksen, Diseth, Schjølberg, & Skjeldal, 2013; May, Sciberras, tle is known about the age of diagnosis and the distribution Brignell, & Williams, 2017; Pelly, Vardy, Fernandez, of prevalence and incidence of diagnosis in each sex. These Newhook, & Chafe, 2015; Poovathinal et al., 2016; Raina data are crucial to understanding how ASD diagnosis is et al., 2017; Rudra et al., 2017; Saemundsen, Magnússon, evolving, and to evaluating and planning the needs of the Georgsdóttir, Egilsson, & Rafnsson, 2013; Skonieczna- healthcare system. Therefore, the aim of this epidemiological Żydecka, Gorzkowska, Pierzak-Sominka, & Adler, 2017; Sun study was to assess the prevalence of ASD in 2017, and its et al., 2015; van Bakel et al., 2015; Zahorodny et al., incidence of diagnosis between 2009 and 2017 among the 2014]. Studies carried out during the last two decades indi- child population of the Catalonia region in Spain, as well as cate that North America and Europe have the highest preva- their temporal and geographical variability. lences of ASD, with median rates of 0.90% and 0.61%, respectively. In the Western Pacific, ASD prevalence esti- mates are slightly lower, with a median rate of 0.50%, while Methods Population and Study Design in South America, Eastern Mediterranean, and South East Asia, ASD prevalence estimates were much lower, with We used administrative data for children insured under the median rates not exceeding 0.39%. In Europe, several studies Catalan Public Health Service between 2009 and 2017. In have been conducted, mostly in Northern countries, and this cohort, we identified children diagnosed with ASD at have used existing registers. However, in Spain, the most public healthcare centers between 2009 and 2017, and we recent studies of ASD were based on a two-phase screening defined population-based denominators based on the total approach. The first study that used this approach was con- population of children in 2017, and the total population of ducted in the Canary Islands in 2012, and reported an esti- children at risk for each year between 2009 and 2017 mated ASD prevalence of 0.61% in toddlers [Fortea Sevilla (described below). Spain’s National Health System provides et al., 2013]. The other study was conducted in one area of universal access to health services for all native and foreign- Catalonia and showed an estimated ASD prevalence of born children [García-Armesto, Begoña Abadía-Taira, Durán, 1.55% in preschool children, and 1.00% in primary school Hernández-Quevedo, & Bernal-Delgado, 2010], regardless of children [Morales-Hidalgo, Roigé-Castellví, Hernández- socioeconomic/financial status or racial/ethnic identification Martínez, Voltas, & Canals, 2018]. Regarding sex differences [García-Altés, Ruiz-Muñoz, Colls, Mias, & Martín Bassols, in ASD diagnosis, most previous studies have shown a 2018; Rajmil, 2000]. This study was reviewed and approved higher ASD prevalence in boys than in girls, with a median by the Clinical Research Ethics Committee of the Parc de sex ratio of 4 (boy:girl) [Elsabbagh et al., 2012]. Although Salut Mar. ASD symptoms tend to manifest themselves during early childhood, [American Psychiatric Association, 2013] some ASD Cases studies have reported higher ASD prevalence in children aged 4–9 years [Al-Farsi et al., 2011; Skonieczna-Żydecka In Catalonia, all children diagnosed with ASD in a public et al., 2017; Yeargin-Allsopp et al., 2003]. Furthermore, healthcare center are registered by the Catalan Health 1694 Pérez-Crespo et al./Autism spectrum disorders in children INSAR Service in the Minimum Basic Data Set register. This popu- We calculated the annual ASD diagnosis incidence between lation register collects information on the patients’ activity 2009 and 2017 by dividing the number of newly diagnosed and health morbidity, and is populated using information cases of ASD among 2- to 17-year-olds (numerator) by the provided by four different types of public healthcare cen- total population at-risk 2- to 17-year-olds for each year ters: mental health centers, mental health outpatient (denominator). Annual ASD diagnosis incidence between clinics, primary care centers, and hospital discharges. 2009 and 2017 was stratified by sex, age group (at diagnosis), In this study, we included 2- to 17-year-old children and healthcare area. As observations were independent with ASD diagnosed between 2009 and 2017 based on between years, we used Poisson regression models to assess the following International Classification of Diseases 9th the annual trend of the overall ASD diagnosis incidence edition (ICD-9) codes: 299.0 (autistic disorder), 299.1 between 2009 and 2017, and that for each sex and age group (childhood disintegrative disorder), 299.8 (other specified (at diagnosis). We performed mixed-effects Poisson regression pervasive developmental disorders), and 299.9 (unspecified analysis to assess differences between healthcare areas pervasive developmental disorder). In the register, ASD according to sex and age group (at diagnosis). The Aran cases might be duplicated when they are registered in dif- healthcare area was excluded because the mixed-effects ferent public healthcare centers or by different health pro- Poisson regression model did not converge for this area, most fessionals. For our study, we selected the first entry in the likely due to the low number of ASD cases diagnosed and the register as the first ASD diagnosis as children have a small population size. Finally, we also created maps of the dis- unique identifier in all centers. We excluded ASD cases tribution of ASD diagnosis incidence for each year and for registered before the age of 2 because ASD diagnosis each healthcare area. We provide the 95% confidence interval before this age might not be reliable [American Psychiatric (95% CI) for all the variables analyzed. Association, 2013; Charman & Baird, 2002]. Statistical analyses were performed using STATA (ver- The Catalan territory is divided into 29 healthcare areas sion 14.2; Stata Corporation, College Station, TX) and based on geographic, socio-economic, and demographic Open Epi (version 3.01; Dean AG, Sullivan KM, Soe MM), factors. For each patient, we collected information about and the maps were drawn using QGIS (version 3.0.2 sex, healthcare area where ASD was diagnosed, and age Girona; QGIS Development Team). (at diagnosis and in 2017). Age was stratified in three cat- egories: 2–5 years, 6–10 years, and 11–17 years. Results Characteristics of the Study Population Population-Based Denominators In 2017, 1,326,666 children aged 2–17 years were insured We estimated the total population of children in 2017 and under the Catalan Health Service. Boys (51.5%) and girls the total population of children at risk for each year between (48.5%) were almost equally represented, and with regards 2009 and 2017, as registered by the Catalan Health Service to age groups 2–5 years old were 22.7%, 6–10 years old through the Central Register of Insured Persons. The total 31.6%, and 11–17 years old 45.7%. The ratios were similar population of children was defined as the number of 2- to for the overall studied population and by sex and age 17-year-olds in 2017, and the total population of children at groups for the other years of study. Out of the total studied risk as the number of 2- to 17-year-olds who were ASD population in 2017, 15,466 had been diagnosed with ASD. diagnosed-free but still at risk of being diagnosed of ASD for Most ASD cases were in the 6–10 years age group (48.2%) each year between 2009 and 2017. For both population- followed by the 2–5 years (30.3%), and the 11–17 years based denominators, we used the population on 1st of groups (21.5%). January of each year. Both population-based denominators were stratified by sex, age group, and healthcare areas, using ASD Prevalence in 2- to 17-Year-Olds in 2017 the same category groups as those for the ASD cases. We observed an overall ASD prevalence of 1.23% (95% CI Statistical Analyses 1.21–1.25) in 2017 (Table 1), significantly higher in boys (1.95%, 95% CI 1.92–1.99) than in girls (0.46%, 95% CI We calculated the ASD prevalence in 2017 by dividing the 0.44–0.48, P < 0.001) with 4.5-fold higher prevalence in number of 2- to 17-year-olds with an ASD diagnosis in boys (12,647 boys vs. 2,819 girls). We also found signifi- 2017 (numerator) by the total number of 2- to 17-year-olds cant differences in ASD prevalence between age groups in the population in 2017(denominator). The ASD preva- according to the patient’s age in 2017 (P < 0.001), with lence in 2017 was stratified by sex, age group (age in 2017), the highest prevalence in the 11–17 years age group and healthcare area. The differences in ASD prevalence (1.80%, 95% CI 1.76–1.83) followed by the 6–10 years between sexes, age groups, and healthcare areas were esti- (1.18%, 95% CI 1.14–1.21) and the 2–5 years groups mated using the chi-squared test. We created a map of the (0.29%, 95% CI 0.26–0.32). Regarding geographical vari- distribution of ASD prevalence by healthcare area. ability, the ASD prevalence varied significantly between INSAR Pérez-Crespo et al./Autism spectrum disorders in children 1695 1696 Table 1. Autism Spectrum Disorders Prevalence Rate (%) in 2017 and Diagnosis Incidence Rates (%) Between 2009 and 2017 by Sex and Age Group Sex Age groupb Total Boys Girls Sex-ratioa 2–5 years old 6–10 years old 11–17 years old Prevalence 2017 N 15,466 12,647 2,819 4.5 869 4,937 9,660 Prevalence rate (95% CI) 1.23 (1.21; 1.25) 1.95 (1.92; 1.99) 0.46 (0.44; 0.48) 0.29 (0.26; 0.32) 1.18 (1.14; 1.21) 1.80 (1.76; 1.83) Incidence 2009 N 919 768 151 5.1 168 390 361 Incidence rate (95% CI) 0.07 (0.06; 0.09) 0.12 (0.10; 0.14) 0.02 (0.00; 0.05) 0.05 (0.02; 0.08) 0.10 (0.07; 0.13) 0.07 (0.05; 0.10) 2010 N 1198 1014 184 5.5 284 499 415 Incidence rate (95% CI) 0.10 (0.08; 0.11) 0.16 (0.14; 0.18) 0.03 (0.01; 0.06) 0.08 (0.05; 0.11) 0.13 (0.10; 0.16) 0.08 (0.06; 0.11) 2011 N 1225 1004 221 4.5 282 548 395 Incidence rate (95% CI) 0.10 (0.08; 0.11) 0.15 (0.13; 0.18) 0.04 (0.01; 0.06) 0.08 (0.05; 0.11) 0.14(0.11; 0.16) 0.08 (0.05; 0.10) 2012 N 1485 1238 247 5.0 398 635 452 Incidence rate (95% CI) 0.12 (0.10; 0.13) 0.19 (0.17; 0.21) 0.04 (0.02; 0.07) 0.11 (0.08; 0.14) 0.16 (0.13; 0.18) 0.09 (0.06; 0.12) 2013 N 2029 1674 355 4.7 570 871 588 Incidence rate (95% CI) 0.16 (0.14; 0.18) 0.26 (0.24; 0.28) 0.06 (0.03; 0.08) 0.16 (0.13; 0.19) 0.21 (0.18; 0.24) 0.12 (0.09; 0.14) 2014 N 2227 1791 436 4.1 541 961 725 Incidence rate (95% CI) 0.18 (0.16; 0.19) 0.27 (0.25; 0.29) 0.07 (0.05; 0.10) 0.16 (0.13; 0.19) 0.23 (0.20; 0.26) 0.13 (0.10; 0.15) 2015 N 2393 1933 460 4.2 587 962 844 Incidence rate (95% CI) 0.19 (0.17; 0.21) 0.30 (0.28; 0.32) 0.08 (0.05; 0.10) 0.18 (0.15; 0.21) 0.23 (0.20; 0.26) 0.14 (0.12; 0.17) 2016 N 2668 2144 524 4.1 695 1055 918 Incidence rate (95% CI) 0.21 (0.20; 0.23) 0.33 (0.31; 0.35) 0.09 (0.06; 0.11) 0.22 (0.19; 0.25) 0.25 (0.22; 0.28) 0.18 (0.15; 0.20) Pérez-Crespo et al./Autism spectrum disorders in children 2017 N 2833 2272 561 4.1 756 1144 933 Incidence rate (95% CI) 0.23 (0.21; 0.24) 0.35 (0.33; 0.37) 0.09 (0.07; 0.12) 0.25 (0.22; 0.28) 0.27 (0.25; 0.30) 0.17 (0.15; 0.20) a Boys versus girls. b Age groups based on child’s age in 2017 for ASD prevalence and age groups at diagnosis for ASD incidence. Abbreviations: CI, confidence interval; N, number of children with autism spectrum disorders. INSAR Figure 1. Autism spectrum disorders prevalence rate (%) among children from 2 to 17 years old in 2017 stratified by healthcare areas of Catalonia region, Spain. 1: Aran; 2: Alt Pirineu; 3: Lleida; 4: Terres de l’Ebre; 5: Baix Camp–Priorat; 6: Alt Camp–Conca de Barberà; 7: Tarragonés–Baix Penedès; 8: Anoia; 9: Solsonès–Bages–Berguedà; 10: Osona; 11: Girona Nord; 12: Girona Sud; 13: Vallès Oriental; 14: Vallès Occidental Est; 15: Vallès Occidental Oest; 16: Baix Llobregat Nord; 17: Alt Penedès–Garraf; 18: Baix Llobregat Centre-Litoral i L’Hospitalet de Llobregat; 19: Barcelonès Nord i Maresme; 20: Barcelona Ciutat Vella; 21: Barcelona Sant Martí; 22: Barcelona les Corts; 23: Barcelona Sants–Montjuïc; 24: Barcelona Sarrià–Sant Gervasi; 25: Barcelona Eixample; 26: Barcelona Gràcia; 27: Barcelona Horta– Guinardó; 28: Barcelona Nou Barris; 29: Barcelona Sant Andreu. Abbreviation: ASD, autism spectrum disorders. healthcare areas in 2017 (P < 0.001), ranging from 0.55% from 0.07% (95% CI 0.06–0.09) in 2009 to 0.23% (95% CI (95% CI 0.46–0.64) in the Barcelona Sarrià-Sant Gervasi 0.21–0.24) in 2017 (Table 1 and Fig. 2A). ASD diagnosis healthcare area to 1.84% (95% CI 1.71–1.98) in Solsonès- incidence increased linearly by 15% (95% CI 1.14–1.15) Bages-Berguedà healthcare area (Fig. 1). each year. Remarkably, we observed a greater increase in diagnosis incidence between 2012 (0.12%) and 2013 ASD Diagnosis Incidence in 2- to 17-Year-Olds From 2009 (0.16%) than in other years of the study period. When to 2017 stratified by sex, we observed a greater increase in ASD diagnosis incidence in girls (0.02% in 2009 and 0.09% in During the period between 2009 and 2017, there was a 2017) than in boys. However, throughout the entire study significant increase in the overall ASD diagnosis incidence period, the ASD diagnosis incidence remained higher in INSAR Pérez-Crespo et al./Autism spectrum disorders in children 1697 (A) 0.25 0.23 Autism spectrum disorders incidence (%) 0.21 0.20 0.19 0.18 0.16 0.15 0.12 0.10 0.10 0.10 0.07 0.05 0.00 2009 2010 2011 2012 2013 2014 2015 2016 2017 Years (B) 0.40 Autism spectrum disorders incidence (%) 0.35 0.35 0.33 0.30 0.30 0.27 0.26 0.25 0.19 0.20 0.16 Boys 0.15 0.15 0.12 Girls 0.08 0.09 0.09 0.10 0.07 0.06 0.04 0.04 0.05 0.02 0.03 0.00 2009 2010 2011 2012 2013 2014 2015 2016 2017 Years (C) 0.30 0.27 Autism spectrum disorders incidence (%) 0.25 0.25 0.25 0.23 0.23 0.21 0.22 0.20 0.18 0.16 0.16 0.16 0.15 0.14 0.18 0.17 0.13 2 to 5 years old 0.11 0.10 0.14 6 to 10 years old 0.10 0.08 0.08 0.12 0.13 0.07 11 to 17 years old 0.08 0.08 0.09 0.05 0.05 0.00 2009 2010 2011 2012 2013 2014 2015 2016 2017 Years Figure 2. Autism spectrum disorders diagnosis incidence rate (%) among children from 2 to 17 years old between 2009 and 2017 in Catalonia region, Spain. (A) Overall autism spectrum disorders diagnosis incidence rate (%). (B) Autism spectrum disorders diag- nosis incidence rate (%) stratified by sex. (C) Autism spectrum disorders diagnosis incidence rate (%) stratified by age group at diagnosis. 1698 Pérez-Crespo et al./Autism spectrum disorders in children INSAR INSAR Pérez-Crespo et al./Autism spectrum disorders in children Figure 3. Autism spectrum disorders diagnosis incidence rate (%) among children from 2 to 17 years old between 2009 and 2017 stratified by healthcare areas of Catalonia region, Spain. (A) Healthcare areas of Catalonia (excluding Barcelona city). (B) Healthcare areas of Barcelona city. 1: Aran; 2: Alt Pirineu; 3: Lleida; 4: Terres de l’Ebre; 5: Baix Camp– Priorat; 6: Alt Camp–Conca de Barberà; 7: Tarragonés–Baix Penedès; 8: Anoia; 9: Solsonès–Bages–Berguedà; 10: Osona; 11: Girona Nord; 12: Girona Sud; 13: Vallès Oriental; 14: Vallès Occidental Est; 15: Vallès Occidental Oest; 16: Baix Llobregat Nord; 17: Alt Penedès–Garraf; 18: Baix Llobregat Centre-Litoral i L’Hospitalet de Llobregat; 19: Barcelonès Nord i Maresme; 20: Barcelona Ciutat Vella; 21: Barcelona Sant Martí; 22: Barcelona les Corts; 23: Barcelona Sants–Montjuïc; 24: Barcelona Sarrià–Sant Gervasi; 25: Barcelona Eixample; 26: Barcelona Gràcia; 27: Barcelona Horta–Guinardó; 28: Barcelona Nou Barris; 29: Barcelona Sant Andreu. Abbreviation: ASD, autism spectrum disorders. 1699 boys than in girls. In addition, there was a significant et al., 2012] and a median ASD prevalence of 0.61% in increase in diagnosis incidence between 2012 and 2013 in Europe [Elsabbagh et al., 2012; Fortea Sevilla et al., 2013; boys that was not detected in girls (Table 1 and Fig. 2B). Idring et al., 2015; Saemundsen et al., 2013; Skonieczna- Regarding the age groups at diagnosis, we observed a Żydecka et al., 2017; van Bakel et al., 2015]. Among the greater increase in ASD diagnosis incidence in 2- to 5-year- most recent European studies, the highest ASD preva- olds (0.05% in 2009 and 0.25% in 2017) than in the other lence was found in Spain with an estimate of 1.55% age groups (Table 1 and Fig. 2C). However, throughout among preschool children [Morales-Hidalgo et al., the entire period between 2009 and 2017, more children 2018]. In contrast, the lowest ASD prevalence was found were diagnosed with ASD in the 6–10 years age group than in Poland with an overall estimate of 0.35% among chil- in the two other age groups at the time of diagnosis. The dren up to 18 years old in 2014 [Skonieczna-Żydecka age group with the fewest children diagnosed with ASD et al., 2017]. The methodologies applied in the different was the 11–17 years old group. The mean age of ASD diag- studies differ in case definition, identification, and diag- nosis was estimated to be 8.8 years, and the median was nosis, making it difficult to compare studies. Most of the 8 years. When studying ASD diagnosis incidence by studies of ASD prevalence include population-based healthcare area, we noted an increase between 2009 and screening studies, which are based on samples that are 2017 in all areas, although there was no clear pattern not always representative of the reference population and across healthcare areas (Fig. 3 and Supporting Information would therefore provide biased estimations. Other stud- Table S1). During the study period, when taking the ASD ies, particularly those including large populations, have incidence within each healthcare area separately, the inci- used different strategies or datasets to define ASD cases, dence varied significantly in most of the healthcare areas including parental reports, healthcare provider datasets, of Catalonia, except in Alt Camp–Conca de Barberà, special educational datasets, or national registers. There- Osona, Aran, Barcelona Ciutat Vella, Barcelona les Corts, fore, when interpreting and comparing studies on ASD Barcelona Sants, Barcelona Eixample, Barcelona Gràcia, prevalence it is crucial to keep in mind the limitations Barcelona Horta, and Barcelona NouBarris. However, inherent to each study design in order to draw meaning- when assessing differences in the annual diagnosis inci- ful conclusions. dence between healthcare areas, no difference across them In our study, the observed sex ratio of 4.5 (boy:girl) is was found. Similarly, there were no significant differences in agreement with the median ratio of 4.0 reported in by sex and age group between healthcare areas from 2009 most epidemiological studies of ASD prevalence [Ameri- to 2017 (Supporting Information Fig. S1). can Psychiatric Association, 2013]. Notwithstanding that, this ratio varies considerably between studies, ranging from 1.0 [Chaaya et al., 2016] to 15.7 [Baird et al., Discussion 2006]. The causes of the observed male bias in ASD preva- lence are not completely understood, but may include The present study has assessed the ASD prevalence in genetic factors or differences in detection and diagnosis. 2017 and the ASD diagnosis incidence between 2009 and Some studies have demonstrated that girls could be 2017, as well as their temporal and geographical variabil- protected from inherited and de novo genetic risk vari- ity, among children aged 2–17 years in Catalonia region ants for ASD [Constantino, Zhang, Frazier, Abbacchi, & in Spain. The ASD prevalence in 2017 was estimated at Law, 2010; Sebat et al., 2007]. Others suggest that genes 1.23%, with a sex ratio of 4.5 (boy:girl), and with the on the sex chromosomes and sex hormones, particularly highest prevalence observed in 11- to 17-year-olds. The testosterone, may modulate the effects of genetic varia- incidence of new ASD diagnosed cases increased signifi- tion on the appearance of an autistic phenotype cantly between 2009 and 2017, from 0.07% to 0.23%, [Auyeung et al., 2009; Baron-Cohen et al., 2011; with a greater increase in girls and, notably, in those aged Werling & Geschwind, 2013]. Differences in detection 2–5 years at the time of diagnosis. We also observed dif- and diagnosis of ASD might be due to the fact that girls ferences in the ASD prevalence between geographical with ASD show higher social motivation and a greater regions in 2017, but not in the annual diagnosis inci- capacity for traditional friendships than boys with ASD dence across the period studied. When stratified by sex [Bargiela, Steward, & Mandy, 2016; Sedgewick, Hill, and age group, there were no significant differences Yates, Pickering, & Pellicano, 2016]. Moreover, girls also between geographical regions in the annual diagnosis score lower on measures of repetitive and stereotyped incidence over the study period. behavior, which might lead to an underdiagnosis of ASD Prior to 1990, the first studies on ASD prevalence in [Rubenstein, Wiggins, & Lee, 2015]. children reported a prevalence of 0.05% [Lotter, 1966], Early diagnosis of ASD is critical, but it is often delayed which has since increased dramatically. Studies carried until school age. Evidence suggests that interventions to out between 2000 and 2016 showed a median worldwide improve the functioning in ASD children may be more ASD prevalence of 0.70% [Baxter et al., 2015; Elsabbagh effective in younger patients, and early treatment 1700 Pérez-Crespo et al./Autism spectrum disorders in children INSAR improves long-term prognosis [Mandell, 2005]. ASD still increases each year, which means that the diagnostic symptoms are typically recognized during the second stability of ASD has not yet been achieved. Therefore, there year of life (12–24 months), but the disorder can be diag- is a need to invest more resources to improve ASD diagno- nosed either before 12 months, if developmental symp- sis, especially at early ages. toms are severe, or after 24 months, if symptoms are The main strengths of our study are the design, which is more subtle [American Psychiatric Association, 2013]. In based on administrative data, and the large sample size our study, the highest ASD prevalence was observed in included. It is one of the few studies carried out in South- 11- to 17-year-olds, with a mean age at diagnosis of ern Europe based on register data including a large annual 8.8 years, and a median of 8 years. Our findings are con- number of births (80,000 per year approximately) [Idescat, sistent with another study conducted in children up to 2018]. The last studies conducted in Spain based on multi- 17 years old in Sweden, which found that the ASD preva- stage screening approach have the advantage of detecting lence increased with age, with a higher ASD prevalence and evaluating the cases of ASD with better precision but among teenagers and a median age at diagnosis of 8 years they have the limitation that only a point prevalence rate [Idring et al., 2015]. However, the results of our study dif- can be estimated. In our study, the use of administrative fer from those of other studies that compared ASD preva- data has allowed us to estimate the prevalence, temporal lence between age groups [Al-Farsi et al., 2011; trends of the diagnosis incidence as well as the geographi- Davidovitch et al., 2013; Latif & Williams, 2007; May cal variability for both prevalence and incidence. Our study et al., 2017; Sheldrick, Maye, & Carter, 2017; Skonieczna- also included data from a relatively long time period, from Żydecka et al., 2017; Yeargin-Allsopp et al., 2003]. Glob- 2009 to 2017, allowing a thorough analysis of temporal ally, these studies found that ASD prevalence was higher trends. Finally, we studied the differences in prevalence among 5- to 9-year-olds, and that the mean age of diag- and incidence of ASD diagnosis for each sex, age group, nosis was ~5.5 years. In two additional studies, the ASD and across geographical healthcare areas. However, we prevalence did not differ between age groups [Harrison, acknowledge that the small numbers in the incidence of 2005; Kogan et al., 2009]. The lower prevalence rates in ASD diagnosis across geographical healthcare areas might younger children in our study could be due to clinicians’ have limited our capability of detecting differences. reluctance to label children as having ASD at very young Our study has also some limitations. The main limita- ages, or a lack of training among health professionals in tion of our study is its retrospective nature: studies based diagnosing ASD at early ages. Nevertheless, our temporal on registers only capture individuals who accessed the trend data suggest that ASD diagnosis at early ages has healthcare service, rather than sampling from the popula- improved over the study period. tion at large. As a result, participants with the studied dis- We found a significant threefold increase in overall order who are not in contact with these services are yet ASD diagnosis incidence between 2009 and 2017, consis- unidentified and not included as cases, which may lead tent with trends in other countries [Davidovitch et al., to underestimation of ASD diagnosis prevalence and inci- 2013; Jensen et al., 2014; Nassar et al., 2009; Pelly et al., dence rates. While we recognize these limitations, other 2015; Raz et al., 2015]. This increase could be due to etio- study designs such as the methods of case confirmation logic factors, but is most likely related to non-etiologic have also important limitations including a low participa- factors; changes in the definition of ASD during the tion of families in the study that would lead to bias esti- period could explain the observed overall increase in mations. On the contrary, the use of data registers is less the incidence of ASD diagnosis. In the last edition of the costly and more time-efficient and may provide a more Diagnostic and Statistical Manual of mental disorders complete coverage of a large population. Furthermore, (DSM) [American Psychiatric Association, 2013], ASD Spain’s National Health System provides universal access included several disorders that were not included in the to health services for all native and foreign-born children previous editions, which would explain that more chil- [García-Altés et al., 2018; García-Armesto et al., 2010]. dren could be diagnosed as having ASD. Another possible Some people also have health cover from private insurers, explanation could be the increasing awareness of ASD in which covers services offered by the public system, as recent years among parents, teachers, psychiatrists, and well as services beyond those offered by the public system pediatricians. In the region studied, several professional (e.g., elective surgery or dental care). This provides easier training actions in diagnostic tools for ASD (Autism Diag- individual direct access to healthcare, more personalized nostic Interview Revised (ADI-R) and Autism Diagnostic care, shorter waiting lists, and private rooms in private Observational Schedule [ADOS]) have been addressed to hospitals [Fusté, Séculi, Brugulat, Medina, & Juncà, 2005]. professionals working in public mental healthcare centers Therefore, the double coverage allows the use of public since 2011. In addition, the Catalan Health Department services and, at the same time, those that are offered by implemented a comprehensive plan in some geographical private insurers according to the user’s preferences and areas to improve ASD detection, diagnosis, and treatment. needs; thus, this private cover supplements the public However, the data showed that ASD diagnosis incidence system. In Catalonia, the rate of this double cover is INSAR Pérez-Crespo et al./Autism spectrum disorders in children 1701 much higher than in other Spanish regions, estimated to expect that most of these ASD diagnosed cases would also reach about 25% of the population [Direcció General de had at least one contact with the public healthcare cen- Planificació en Salut, 2016]. The remaining 75% of the ters in relation to their ASD diagnosis or for other reasons population is covered only by the public health system. and notify the pediatrician about the diagnosis, that is, at Therefore, we expect to have a low number of ASD diag- a regular visits within the “Healthy childhood” program, nosed cases not included in our study, which would lead and that their ASD diagnosis would be registered. to a slight underestimation of prevalence and incidence Although the age of diagnosis might not be accurately estimates found in our study. In addition, study design registered for these cases, we were able to include them in masked variation in ASD diagnostic practices. As our the overall ASD estimations. Future studies should focus study is based on register data, we cannot confirm that on harmonizing and merging the data from all existing ASD diagnosis assessment has been made in the standard registers to provide a more accurate estimate of ASD diag- way for all ASD cases. While some clinicians work within nosis prevalence and incidence. a multidisciplinary team, others work independently to In conclusion, in Catalonia region, we found an ASD make a diagnostic decision. This fact could partly explain prevalence of 1.23% in 2017. This prevalence was higher the variability in ASD diagnosis prevalence and incidence in boys than in girls (sex ratio 4.5), and was highest in the between healthcare areas. Nevertheless, in all the public 11–17 years age group. We also observed a threefold mental healthcare centers of Catalonia where the diagno- increase in ASD diagnosis incidence between 2009 and sis of ASD could be made, there is at least one profes- 2017, and that this was more pronounced among girls and sional specialized in ADI-R and ADOS tools as clinical in children aged 2–5 years at the time of diagnosis. We guidelines recommended. This fact, however, does not observed geographical differences in the ASD prevalence in ensure that the ASD diagnosis would be done using these 2017 between healthcare areas, but not for the annual ASD diagnostic tools. As an example, the time constraints for diagnosis incidence during the period 2009 and 2017. This each visit in the public mental healthcare centers can study provides new epidemiological information on the lead to the health professionals to perform shorter diag- prevalence and incidence of ASD diagnosis in children in nosis based on the DSM criteria instead of ADI-R and Catalonia region, and is one of the few studies that has ADOS. Additionally, no efforts are made by the data pro- been carried out in the South of Europe. The data collected vider to perform a case validation of the ASD cases here will add to the available information for planning and included in the register. For future studies on the epide- evaluating the needs of the healthcare services in Catalonia, miological monitoring of ASD based on this data, a case especially those related to early diagnosis and intervention. validation study should be done as well as an evaluation Finally, it is also important to point out to the need to of the completeness of the data. It is also possible that establish improved methods of registering the diagnosis of ASD diagnosis was underestimated if children were first ASD to ensure that the health burden of the condition is diagnosed with an attention deficit and hyperactivity dis- accurately measured. order, which occurs concomitantly in 30% to 50% of children with ASD [Davis & Kollins, 2012]. Finally, we Acknowledgments only retrieved information on ASD cases from the public healthcare centers that are under the Catalan Health The research described in this article was conducted Department. We were unable to access data from private under contract with the Health Effects Institute (HEI), an healthcare centers, the Child Development and Early organization funded jointly by the United States Environ- Care Centers that are under the Catalan Welfare Depart- mental Protection Agency (EPA; Assistance Award No. R- ment, and the Psycho-Pedagogical Assessment Teams 82811201) and certain motor vehicle and engine manu- that are under the Catalan Educational Department. facturers. The contents of this article do not necessarily However, in Catalonia there is a health program called reflect the views of HEI or its sponsors, nor do they neces- “Healthy Childhood” aimed at children aged 0–14 years sarily reflect the views and policies of the EPA or motor old [Direcció General de Salut Pública, 2008]. This pro- vehicle and engine manufacturers. M.G. is funded by a gram is about issues related to childhood health (physical Miguel Servet fellowship (MS13/00054, CP18/00018) and psychomotor development, routine immunizations, from the Spanish Institute of Health Carlos III. A.P. is locomotor system, genitourinary system, ophthalmology, supported by the Medical Research Council (grant num- otorhinolaryngology, oral health, etc.) and visits to the bers MR/K501256/1, MR/N013468/1). The authors would pediatrician are scheduled throughout the whole child- like to express their gratitude to Joan Albert d’Escofet, hood. Although it is estimated that a maximum of 25% from the Agency for Health Quality and Assessment of of the population has double health coverage [Direcció the Catalan Health Department, for providing essential General de Planificació en Salut, 2016], it is important to data and support for this study. The research materials point out that most of the children insured in the Cata- supporting this publication can be accessed by contacting lan Health System follow this program. Therefore, we the corresponding author on reasonable request. 1702 Pérez-Crespo et al./Autism spectrum disorders in children INSAR Conflict of Interest Constantino, J. N., Zhang, Y., Frazier, T., Abbacchi, A. M., & Law, P. (2010). Sibling recurrence and the genetic epidemiology of A.T. is the father of a child with ASD and a member of autism. The American Journal of Psychiatry, 167, 1349–1356. Davidovitch, M., Hemo, B., Manning-Courtney, P., & two patient associations (Aprenem and JuntsAutisme) Fombonne, E. (2013). Prevalence and incidence of autism that strive for the inclusion of people with ASD. spectrum disorder in an Israeli population. Journal of Autism and Developmental Disorders, 43, 785–793. Davis, N. O., & Kollins, S. H. (2012). Treatment for co-occurring attention deficit/hyperactivity disorder and autism spectrum References disorder. Neurotherapeutics, 9, 518–530. Al-Farsi, Y. M., Al-Sharbati, M. M., Al-Farsi, O. A., Al- Dekkers, L. M. S., Groot, N. A., Díaz Mosquera, E. N., Andrade Shafaee, M. S., Brooks, D. R., & Waly, M. I. (2011). Brief Zúñiga, I. P., & Delfos, M. F. (2015). Prevalence of autism report: Prevalence of autistic spectrum disorders in the sultan- spectrum disorders in Ecuador: A pilot study in quito. Journal ate of Oman. Journal of Autism and Developmental Disor- of Autism and Developmental Disorders, 45, 4165–4173. ders, 41, 821–825. Diallo, F. B., Fombonne, É., Kisely, S., Rochette, L., Vasiliadis, H.-M., American Psychiatric Association. (2013). Diagnostic and statisti- Vanasse, A., … Lesage, A. (2018). Prevalence and correlates of cal manual of mental disorders: DSM-5 (5th ed.). autism spectrum disorders in Quebec: Prévalence et corrélats des Washington, DC: American Psychiatric Association. troubles du spectre de l’autisme au Québec. The Canadian Jour- Auyeung, B., Baron-Cohen, S., Ashwin, E., Knickmeyer, R., nal of Psychiatry, 63, 231–239. Taylor, K., & Hackett, G. (2009). Fetal testosterone and autis- Direcció General de Planificació en Salut. (2016). Enquesta de tic traits. British Journal of Psychology, 100, 1–22. salut de Catalunya. Comportaments relacionats amb la salut, Baird, G., Simonoff, E., Pickles, A., Chandler, S., Loucas, T., l’estat de salut i l’ús de serveis sanitaris a Catalunya. Informe dels Meldrum, D., & Charman, T. (2006). Prevalence of disorders principals resultats 2015. Barcelona. of the autism spectrum in a population cohort of children in Direcció General de Salut Pública. (2008). Protocol d’activitats South Thames: The Special Needs and Autism Project (SNAP). preventives i de promoció de la salut a l’edat pediàtrica: The Lancet, 368, 210–215. infància amb salut. Barcelona. Barbaresi, W. J., Katusic, S. K., Colligan, R. C., Weaver, A. L., & Elsabbagh, M., Divan, G., Koh, Y.-J., Kim, Y. S., Kauchali, S., Jacobsen, S. J. (2005). The incidence of autism in Olmsted Marcín, C., … Fombonne, E. (2012). Global prevalence of County, Minnesota, 1976–1997: Results from a population- autism and other pervasive developmental disorders: Global based study. Archives of Pediatrics & Adolescent Medicine, epidemiology of autism. Autism Research, 5, 160–179. 159, 37–44. Farley, M. A., McMahon, W. M., Fombonne, E., Jenson, W. R., Bargiela, S., Steward, R., & Mandy, W. (2016). The experiences of Miller, J., Gardner, M., … Coon, H. (2009). Twenty-year out- late-diagnosed women with autism spectrum conditions: An come for individuals with autism and average or near-average investigation of the female autism phenotype. Journal of cognitive abilities. Autism Research, 2, 109–118. Autism and Developmental Disorders, 46, 3281–3294. Fombonne, E. (2009a). A wrinkle in time: From early signs to a Baron-Cohen, S., Lombardo, M. V., Auyeung, B., Ashwin, E., diagnosis of autism. Journal of the American Academy of Chakrabarti, B., & Knickmeyer, R. (2011). Why are autism Child & Adolescent Psychiatry, 48, 463–464. spectrum conditions more prevalent in males? PLoS Biology, Fombonne, E. (2009b). Epidemiology of pervasive developmen- 9, e1001081. tal disorders. Pediatric Research, 65, 591–598. Baxter, A. J., Brugha, T. S., Erskine, H. E., Scheurer, R. W., Vos, T., & Fortea Sevilla, M. S., Escandell Bermúdez, M. O., & Castro Scott, J. G. (2015). The epidemiology and global burden of autism Sánchez, J. J. (2013). Estimación de la prevalencia de los spectrum disorders. Psychological Medicine, 45, 601–613. trastornos del espectro autista en Canarias. Anales de Chaaya, M., Saab, D., Maalouf, F. T., & Boustany, R.-M. (2016). Pediatría, 79, 352–359. Prevalence of autism spectrum disorder in nurseries in Leba- Fusté, J., Séculi, E., Brugulat, P., Medina, A., & Juncà, S. (2005). non: A cross sectional study. Journal of Autism and Develop- Población con cobertura pública o doble cobertura de mental Disorders, 46, 514–522. aseguramiento sanitario. ¿Cuál es la diferencia? Gaceta Sani- Charman, T., & Baird, G. (2002). Practitioner review: Diagnosis taria, 19, 15–21. of autism spectrum disorder in 2- and 3-year-old children. García-Altés, A., Ruiz-Muñoz, D., Colls, C., Mias, M., & Martín Journal of Child Psychology and Psychiatry, and Allied Disci- Bassols, N. (2018). Socioeconomic inequalities in health and plines, 43, 289–305. the use of healthcare services in Catalonia: Analysis of the Chien, I.-C., Lin, C.-H., Chou, Y.-J., & Chou, P. (2011). Preva- individual data of 7.5 million residents. Journal of Epidemiol- lence and Incidence of autism spectrum disorders among ogy and Community Health, 72, 871–879. national health insurance enrollees in Taiwan from 1996 to García-Armesto, S., Begoña Abadía-Taira, M., Durán, A., Hernández- 2005. Journal of Child Neurology, 26, 830–834. Quevedo, C., & Bernal-Delgado, E. (2010). Spain: Health system Chinawa, J. M., Manyike, P. C., Aniwada, E. C., Chinawa, A. T., review. Health Systems in Transition, 12, 1–295. Obu, H. A., Odetunde, O. I., … Ibekwe, R. R. (2017). Preva- Hansen, S. N., Schendel, D. E., & Parner, E. T. (2015). Explaining lence and socioeconomic correlates of autism among children the increase in the prevalence of autism spectrum disorders: attending primary and secondary schools in south east Nige- The proportion attributable to changes in reporting practices. ria. African Health Sciences, 16, 936. JAMA Pediatrics, 169, 56–62. INSAR Pérez-Crespo et al./Autism spectrum disorders in children 1703 Harrison, M. J. (2005). Prevalence of autistic spectrum disorders in disorders: A Danish population-based study. Psychological Lothian, Scotland: An estimate using the “capture-recapture” Medicine, 34, 1339–1346. technique. Archives of Disease in Childhood, 91, 16–19. Lotter, V. (1966). Epidemiology of autistic conditions in young Hervas, A. (2016). One autism, several autisms. Phenotypical var- children: 1. Prevalence. Social Psychiatry, 1, 124–135. iability in autism spectrum disorders. Revista De Neurologia, Lyall, K., Croen, L., Daniels, J., Fallin, M. D., Ladd-Acosta, C., 62, S9–S14. Lee, B. K., … Newschaffer, C. (2017). The changing epidemi- Hewitt, A., Hall-Lande, J., Hamre, K., Esler, A. N., Punyko, J., ology of autism spectrum disorders. Annual Review of Public Reichle, J., & Gulaid, A. A. (2016). Autism spectrum disorder Health, 38, 81–102. (ASD) prevalence in Somali and non-Somali children. Journal Mandell, D. S. (2005). Factors associated with age of diagnosis of Autism and Developmental Disorders, 46, 2599–2608. among children with autism spectrum disorders. Pediatrics, Hill, A. P., Zuckerman, K., & Fombonne, E. (2014). Handbook of 116, 1480–1486. autism and pervasive developmental disorders. Diagnosis, May, T., Sciberras, E., Brignell, A., & Williams, K. (2017). Autism development, and brain mechanisms (4th ed.). New York, spectrum disorder: Updated prevalence and comparison of NY: Wiley. two birth cohorts in a nationally representative Australian Hinkka-Yli-Salomäki, S., Banerjee, P. N., Gissler, M., sample. BMJ Open, 7, e015549. Lampi, K. M., Vanhala, R., Brown, A. S., & Sourander, A. Modabbernia, A., Velthorst, E., & Reichenberg, A. (2017). Environ- (2014). The incidence of diagnosed autism spectrum disorders mental risk factors for autism: An evidence-based review of sys- in Finland. Nordic Journal of Psychiatry, 68, 472–480. tematic reviews and meta-analyses. Molecular Autism, 8, 13. Honda, H., Shimizu, Y., Misumi, K., Niimi, M., & Ohashi, Y. Morales-Hidalgo, P., Roigé-Castellví, J., Hernández-Martínez, C., (1996). Cumulative incidence and prevalence of childhood Voltas, N., & Canals, J. (2018). Prevalence and characteristics of autism in children in Japan. The British Journal of Psychiatry: autism spectrum disorder among Spanish school-age children. The Journal of Mental Science, 169, 228–235. Journal of Autism and Developmental Disorders, 48, 3176–3190. Huang, J. P., Cui, S. S., Han, Y., Irva, H.-P., Qi, L. H., & Zhang, X. Murray, M. L., Hsia, Y., Glaser, K., Simonoff, E., (2014). Prevalence and early signs of autism spectrum disor- Murphy, D. G. M., Asherson, P. J., … Wong, I. C. K. (2014). der (ASD) among 18–36 month-old children of Tianjin in Pharmacological treatments prescribed to people with autism China. Biomedical and Environmental Sciences: BES, 27, spectrum disorder (ASD) in primary health care. Psychophar- 453–461. macology, 231, 1011–1021. Huguet, G., Ey, E., & Bourgeron, T. (2013). The genetic land- Nassar, N., Dixon, G., Bourke, J., Bower, C., Glasson, E., de scapes of autism spectrum disorders. Annual Review of Geno- Klerk, N., & Leonard, H. (2009). Autism spectrum disorders in mics and Human Genetics, 14, 191–213. young children: Effect of changes in diagnostic practices. Idescat. (2018). Estadística de naixements. Nascuts vius segons sexe. International Journal of Epidemiology, 38, 1245–1254. Catalunya. Retrieved from https://www.idescat.cat/pub/?id= Pelly, L., Vardy, C., Fernandez, B., Newhook, L. A., & Chafe, R. naix&n=364 (2015). Incidence and cohort prevalence for autism spectrum Idring, S., Lundberg, M., Sturm, H., Dalman, C., Gumpert, C., disorders in the Avalon Peninsula, Newfoundland and Labra- Rai, D., … Magnusson, C. (2015). Changes in prevalence of dor. CMAJ Open, 3, E276–E280. autism spectrum disorders in 2001–2011: Findings from the Plubrukarn, R., Piyasil, V., Moungnoi, P., Tanprasert, S., & Stockholm Youth Cohort. Journal of Autism and Develop- Chutchawalitsakul, V. (2005). Trend study of autistic spec- mental Disorders, 45, 1766–1773. trum disorders at Queen Sirikit National Institute of Child Isaksen, J., Diseth, T. H., Schjølberg, S., & Skjeldal, O. H. (2013). Health. Journal of the Medical Association of Thailand, 88, Autism spectrum disorders—Are they really epidemic? 891–897. European Journal of Paediatric Neurology, 17, 327–333. Poovathinal, S. A., Anitha, A., Thomas, R., Kaniamattam, M., Jensen, C. M., Steinhausen, H.-C., & Lauritsen, M. B. (2014). Melempatt, N., Anilkumar, A., & Meena, M. (2016). Preva- Time trends over 16 years in incidence-rates of autism spec- lence of autism spectrum disorders in a semiurban commu- trum disorders across the lifespan based on Nationwide Dan- nity in south India. Annals of Epidemiology, 26, 663. ish Register Data. Journal of Autism and Developmental e8–665.e8. Disorders, 44, 1808–1818. Powell, J. E., Edwards, A., Edwards, M., Pandit, B. S., Sungum- Kim, Y. S., Leventhal, B. L., Koh, Y.-J., Fombonne, E., Laska, E., Paliwal, S. R., & Whitehouse, W. (2000). Changes in the inci- Lim, E.-C., … Grinker, R. R. (2011). Prevalence of autism spec- dence of childhood autism and other autistic spectrum disor- trum disorders in a total population sample. American Jour- ders in preschool children from two areas of the West nal of Psychiatry, 168, 904–912. Midlands, UK. Developmental Medicine and Child Neurol- Kogan, M. D., Blumberg, S. J., Schieve, L. A., Boyle, C. A., Perrin, ogy, 42, 624–628. J. M., Ghandour, R. M., … van Dyck, P. C. (2009). Prevalence Raina, S., Chander, V., Bhardwaj, A., Kumar, D., Sharma, S., of parent-reported diagnosis of autism spectrum disorder Kashyap, V., … Bhardwaj, A. (2017). Prevalence of autism among children in the US, 2007. Pediatrics, 124, 1395–1403. spectrum disorder among rural, urban, and tribal children Latif, A. H. A., & Williams, W. R. (2007). Diagnostic trends in (1–10 years of age). Journal of Neurosciences in Rural Practice, autistic spectrum disorders in the South Wales valleys. 8, 368–374. Autism, 11, 479–487. Rajmil, L. (2000). The quality of care and influence of double Lauritsen, M. B., Pedersen, C. B., & Mortensen, P. B. (2004). The health care coverage in Catalonia (Spain). Archives of Disease incidence and prevalence of pervasive developmental in Childhood, 83, 211–214. 1704 Pérez-Crespo et al./Autism spectrum disorders in children INSAR Raz, R., Weisskopf, M. G., Davidovitch, M., Pinto, O., & increasing prevalence of autism spectrum disorders in a Levine, H. (2015). Differences in autism spectrum disorders French area from register-based data. Journal of Autism and incidence by sub-populations in Israel 1992–2009: A total Developmental Disorders, 45, 3255–3261. population study. Journal of Autism and Developmental Dis- Werling, D. M., & Geschwind, D. H. (2013). Sex differences in autism orders, 45, 1062–1069. spectrum disorders. Current Opinion in Neurology, 26, 146–153. Rubenstein, E., Wiggins, L. D., & Lee, L.-C. (2015). A review of Williams, K., Glasson, E. J., Wray, J., Tuck, M., Helmer, M., the differences in developmental, psychiatric, and medical Bower, C. I., & Mellis, C. M. (2005). Incidence of autism spec- endophenotypes between males and females with autism trum disorders in children in two Australian states. The Medi- spectrum disorder. Journal of Developmental and Physical cal Journal of Australia, 182, 108–111. Disabilities, 27, 119–139. Wong, V. C. N., & Hui, S. L. H. (2008). Epidemiological study of Rudra, A., Belmonte, M. K., Soni, P. K., Banerjee, S., autism spectrum disorder in China. Journal of Child Neurol- Mukerji, S., & Chakrabarti, B. (2017). Prevalence of autism ogy, 23, 67–72. spectrum disorder and autistic symptoms in a school-based Yeargin-Allsopp, M., Rice, C., Karapurkar, T., Doernberg, N., cohort of children in Kolkata, India: Prevalence of ASD in Boyle, C., & Murphy, C. (2003). Prevalence of autism in a US India. Autism Research, 10, 1597–1605. metropolitan area. JAMA, 289, 49–55. Saemundsen, E., Magnússon, P., Georgsdóttir, I., Egilsson, E., & Zachor, D. A., & Curatolo, P. (2014). Recommendations for early Rafnsson, V. (2013). Prevalence of autism spectrum disorders diagnosis and intervention in autism spectrum disorders: An in an Icelandic birth cohort. BMJ Open, 3, e002748. Italian–Israeli consensus conference. European Journal of Pae- Sebat, J., Lakshmi, B., Malhotra, D., Troge, J., Lese-Martin, C., diatric Neurology, 18, 107–118. Walsh, T., … Wigler, M. (2007). Strong association of de novo Zahorodny, W., Shenouda, J., Howell, S., Rosato, N. S., copy number mutations with autism. Science, 316, 445–449. Peng, B., & Mehta, U. (2014). Increasing autism prevalence in Sedgewick, F., Hill, V., Yates, R., Pickering, L., & Pellicano, E. metropolitan New Jersey. Autism, 18, 117–126. (2016). Gender differences in the social motivation and friendship experiences of autistic and non-autistic adoles- cents. Journal of Autism and Developmental Disorders, 46, Supporting Information 1297–1306. Sheldrick, R. C., Maye, M. P., & Carter, A. S. (2017). Age at first Additional supporting information may be found online identification of autism spectrum disorder: An analysis of two in the Supporting Information section at the end of the US surveys. Journal of the American Academy of Child & article. Adolescent Psychiatry, 56, 313–320. Skonieczna-Żydecka, K., Gorzkowska, I., Pierzak-Sominka, J., & Table S1. Autism spectrum disorders prevalence rate (%) Adler, G. (2017). The prevalence of autism spectrum disorders in in 2017 and ASD diagnosis incidence rates (%) between West Pomeranian and Pomeranian regions of Poland. Journal of 2009 and 2017 among children between 2 to 17 years old Applied Research in Intellectual Disabilities, 30, 283–289. by healthcare areas of the Catalonia region, Spain. Sun, X., Allison, C., Matthews, F. E., Zhang, Z., Auyeung, B., Baron- Figure S1. Incidence risk ratio of autism spectrum disor- Cohen, S., & Brayne, C. (2015). Exploring the underdiagnosis ders diagnosis incidence over the study period from 2009 and prevalence of autism spectrum conditions in Beijing: Under- to 2017 between healthcare areas of the Catalonia region, diagnosis of autism in China. Autism Research, 8, 250–260. Spain, stratified by sex and age group at diagnosis. (A) Taylor, B., Jick, H., & MacLaughlin, D. (2013). Prevalence and Boys from 2 to 5 years old (B) Boys from 6 to 10 years old incidence rates of autism in the UK: Time trend from 2004–2010 in children aged 8 years. BMJ Open, 3, e003219. (C) Boys from 11 to 17 years old (D) Girls from 2 to van Bakel, M. M. E., Delobel-Ayoub, M., Cans, C., Assouline, B., 5 years old (E) Girls from 6 to 10 years old (F) Girls from Jouk, P.-S., Raynaud, J.-P., & Arnaud, C. (2015). Low but 11 to 17 years old. INSAR Pérez-Crespo et al./Autism spectrum disorders in children 1705