The Variation of Psychopharmacological Prescription Rates for PeopleWith Autism Spectrum Disorder (ASD) in 30 CountriesAngel Y.S. Wong, Yingfen Hsia, Esther W. Chan, Declan G.M. Murphy, Emily Simonoff, Jan K. Buitelaar,and Ian C.K. Wong

There is significant variation in prescriptions among countries in clinical practice for the treatment of comorbiditiesassociated with autism spectrum disorder (ASD). It has been suggested that many people with mental health disordersin low-/middle-income countries do not receive adequate treatment. Hence, this study investigated psychopharmaco-logical treatment patterns for ASD comorbidities in 30 countries and the association between country’s income andprescription rates.

The IMS Prescribing Insights database was used to investigate prescription patterns for ASD comorbidity treatmentfrom 2007 to 2012. Data were obtained from 30 countries in continents of Europe, Asia, Oceania, Central America, SouthAmerica, and Africa. The gross domestic product (GDP) per capita was used as a proxy for each country’s income.Spearman correlation was used to examine the association between prescription rate and GDP per capita.

The highest prescription rates were found in Western Europe (3.89–36.36/10,000) while the lowest prescription rateswere found in Asian countries, such as Turkey, Indonesia, Saudi Arabia, and Pakistan (0.04–0.82/10,000). The mostcommonly prescribed drug for ASD comorbidity treatment in most of the countries was risperidone, but antidepressantsand antiepileptic drugs were also frequently prescribed. There was a significant positive correlation between GDP percapita and prescription rate (Spearman ρ = 0.60; P = 0.0011; 95% confidence interval 0.27–0.81), that is, the higher theGDP per capita, the higher the prescription rate.

There are marked international differences in prescription rates, and this is partially accounted by economic factors.Future research should combine more data for ASD comorbidity treatment to explore the disparity of psychopharma-cological treatment between countries. Autism Res 2014, ••: ••–••. © 2014 International Society for Autism Research,Wiley Periodicals, Inc.

Keywords: epidemiology; Gross Domestic Product; Psychopharmacology; Multinational study

Introduction

Autism spectrum disorder (ASD) is a neurodevelopmentalcondition characterized by impaired social and commu-nicative behavior as well as rigid and stereotyped patternsof behavior [American Psychiatric Association, 1994;

World Health Organization, 2012]. ASD is oftencomorbid with mental health conditions includinganxiety and mood disorders, attention-deficit hyperactiv-ity disorder (ADHD), seizure disorder, sleep disorder, andlearning disabilities [Bradley & Bolton, 2006; MacNeil,Lopes, & Minnes, 2009; Matson & Neal, 2009; Simonoff

From the Centre for Safe Medication Practice and Research, Department of Pharmacology and Pharmacy, Li Ka Shing Faculty of Medicine, The Universityof Hong Kong, Hong Kong, SAR, China (A.Y.S.W., Y.H., E.W.C., I.C.K.W.); Sackler Institute for Translational Neurodevelopmental Research, Departmentof Forensic and Developmental Sciences, Institute of Psychiatry, King’s College London, London, United Kingdom (D.G.M.M.); Department of Child andAdolescent Psychiatry, Institute of Psychiatry and Biomedical Research Centre for Mental Health, King’s College London, London, United Kingdom(E.S.); Donders Institute for Brain, Cognition and Behavior, Department of Cognitive Neuroscience, Radboud University Nijmegen Medical Center,Nijmegen, The Netherlands (J.K.B.); Centre for Paediatric Pharmacy Research, Department of Practice and Policy, UCL School of Pharmacy, UniversityCollege London, London, United Kingdom (Y.H., I.C.K.W.)

Received September 21, 2013; accepted for publication April 21, 2014Address for correspondence and reprints: Ian Chi Kei Wong, Centre for Safe Medication Practice and Research, Department of Pharmacology and

Pharmacy, Li Ka Shing Faculty of Medicine, The University of Hong Kong, Pokfulam, Hong Kong, SAR, China. E-mail: wongick@hku.hkConflict of interest: E. S., Y. H., E. W. C., and A. W. have no conflict of interest. D. G. M. has received research funding and honoraria from variouspharmaceutical companies, including Shire and Janssen-Cilag, and has previously received honoraria from Janssen-Cilag and Wyeth. He is currentlyreceiving funding from the IMI for EU-AIMS to identify new treatment targets for ASD.

J. B. has been a consultant to/member of advisory board of and/or speaker for Janssen Cilag BV, Eli Lilly, Shire, Novartis, Roche, and Servier in the last3 years. He is not an employee or stock shareholder of any of these companies. He has no other financial or material support, including expert testimony,patents, and royalties. He currently receives funding from the European Union’s Seventh Framework Programme to investigate the safety of risperidonein children (PERS), the long-term safety of methylphenidate (ADDUCE) and EU-AIMS to identify new treatment targets for ASD.

I. W. has received research funding and honoraria from various pharmaceutical companies, including Shire, Janssen-Cilag, and Bristol-Myers Squibb.He currently receives funding from the European Union’s Seventh Framework Programme to investigate the safety of risperidone in children (PERS) andthe long-term safety of methylphenidate (ADDUCE). He is also a director of Healthcare Innovation Technology Service Limited, which received fundingfrom the IMI for taking part in the EU-AIMS project.Published online in Wiley Online Library (wileyonlinelibrary.com)DOI: 10.1002/aur.1391© 2014 International Society for Autism Research, Wiley Periodicals, Inc.

RESEARCH ARTICLE

INSAR 1Autism Research ••: ••–••, 2014

et al., 2008; Souders et al., 2009]. Recent epidemiologicalstudies report a prevalence of ASD in approximately 1.1%in children and adolescents in the United Kingdom andthe United States [Baird et al., 2006; Kogan et al., 2009],but prevalence varies in different countries and acrossstudies [Elsabbagh et al., 2012]. Approximately 70% ofindividuals with ASD suffer from additional mentalhealth conditions [Simonoff et al., 2008].

The clinical management of ASD aims to maximize theaffected individual’s functional independence and qualityof life, and to mitigate carers’ distress and burden[Carbone, Farley, & Do, 2010; Myers et al., 2007] by mini-mizing core symptoms, lowering the impact of associatedmental health symptoms, and reducing maladaptivebehaviors [Cadman et al., 2012]. A comprehensive treat-ment approach includes educational and behavioral pro-grams and pharmacological treatment as an adjunct in themanagement of ASD [Myers et al., 2007]. There is cur-rently no highly effective pharmacological treatment forthe core symptoms of ASD; therefore, current approachestarget the minimization of comorbid symptoms. Cur-rently, however, there are limited practical guidelines forthe pharmacological treatment of associated symptoms inchildren and adolescents with ASD [Frazier et al., 2011]. Inthe United States, only risperidone and aripiprazole areapproved by the Food and Drug Administration (FDA) forthe treatment of irritability in children with ASD [Marcuset al., 2009; Owen et al., 2009; Shea et al., 2004]. In otherwestern countries, risperidone is also approved for theindication of behavioral disturbance among children andadolescents with ASD by the European Medicines Agency(EMA), the Medicines and Healthcare Products RegulatoryAgency of the United Kingdom, and the Australian Thera-peutic Goods Administration [EMA, 2007; WHO, 2013].Despite insufficient evidence on the efficacy and long-term safety of psychotropic drugs for ASD treatment,various medications have been used for ASD comorbiditytreatment in clinical practice such as α-adrenergic agents,β-blockers, mood stabilizers, antiepileptic medication,and antipsychotics [Hollander, Phillips, & Yeh, 2003;McPheeters et al., 2011; Myers et al., 2007; Rosenberget al., 2010].

Previous studies have estimated that between 27% and60% of children and adolescents received at least onemedication for ASD treatment in the United States andCanada [Coury et al., 2012; Frazier et al., 2011; Mandellet al., 2008]. In the United Kingdom, a study using aprimary care database reported approximately 29%(1,619/5,651) of young people aged less than 25 years withASD received psychotropic drugs; and the commonly pre-scribed drug classes were sleep medications (9.7%), stimu-lants (7.9%), and antipsychotics (7.3%) [Murray et al.,2014]. Although previous studies have demonstrated theprescription patterns of ASD comorbidity treatment in theUnited States and the United Kingdom, generalization of

the findings to other countries is unknown. This is ofrelevance due to geographic and cultural differences, aswell as potential variations in clinical practice [Murrayet al., 2014; Rani, Murray, Byrne, & Wong, 2008]. A face-to-face interview survey study conducted by the WHO in17 countries showed that the proportion of patients withmental health disorders (e.g. anxiety, mood, and sub-stance disorders) receiving either pharmacotherapy (≥1month of a medication, plus ≥4 visits to a medical doctor)or psychotherapy (≥8 visits with any professional) wassmaller in low- and middle-income countries comparedwith high-income countries [Wang et al., 2007]. In addi-tion, it suggested that the majority of people with mentalhealth disorders in low-/middle-income countries do notreceive the treatment they need [Lund et al., 2012]. In ourprevious work, we demonstrated significant variation inpsychopharmacological prescription patterns between2010 and 2012 in children and adolescents with ASDacross ten countries [Hsia et al., 2014]. However, we didnot investigate the extent to which a country’s income isassociated with variation in drug treatment. There arecurrently very limited data from cross-national studies onthe treatment of ASD comorbidities. Therefore, we con-ducted a descriptive comparison of prescription rates in 30countries that previously had no published data. We inves-tigated whether a country’s income was associated withASD comorbidity treatment.

MethodData Source

This was a descriptive study using the IMS PrescribingInsights (MIDAS) database to investigate prescriptionsissued for patients with a diagnosis of ASD in 30 coun-tries. The IMS MIDAS database is an audit drawn from arepresentative sample of practitioners in each countryand contains diagnoses and prescription data from differ-ent countries [Wong & Murray, 2005]. The sampled prac-titioners were recruited and asked to record details of theofficial prescription forms for each patient visit. As nopatient identifier was available in our dataset, it wasnot possible to determine the number of patients whohad visited these practitioners. The prescription dataof sampled practitioners were adjusted according to dif-ferent stratification methods (Table 1), and a projectednational total of prescriptions data per year was calcu-lated for each country. Table 1 summarizes the detailsof data collection including the number of doctorsinvolved, stratification methods, and reporting timefor each country. Previous studies have used the MIDASdatabase to investigate the prescription trends and pat-terns of psychotropic prescriptions [Balkrishnan, Phatak,Gleim, & Karve, 2009; Livermore et al., 2006; Wong,Murray, Camilleri-Novak, & Stephens, 2004]. These

INSAR2 Wong et al./Variation of prescription rates for ASD

studies indicate that this database is suitable for investi-gating prescription patterns on a global level.

Study Population

Prescriptions for patients with a diagnosis of ASD wereidentified using International Classification of Disease(ICD), tenth revision codes under F84 (Pervasive Devel-opment Disorder; PDD) (Supporting InformationTable S1) between January 2007 and December 2012(study period) in the IMS MIDAS database. Data wereobtained from 12 European countries (Austria, Belgium,the Netherlands, Finland, Ireland, Switzerland, Greece,Czech Republic, Hungary, Poland, Portugal, and Slova-kia), six Asian countries (South Korea, Thailand, Indone-sia, Pakistan, Turkey, and Saudi Arabia), two South

American countries (Colombia and Venezuela), twocountries in Oceania (Australia and New Zealand), andtwo African countries (Egypt and South Africa). The pre-scription data of Central America were drawn fromsampled practitioners in Guatemala, Honduras, El Salva-dor, Nicaragua, Costa Rica, and Panama then adjustedaccording to stratifications to obtain the projected pre-scriptions. Therefore, “Central America” was used to rep-resent these six countries’ projected prescription datathroughout this study.

Drug Prescribing

We used several drug categories to classify ASDcomorbidities treatment: (a) stimulants (dexamfetamineand methylphenidate) and atomoxetine; (b) antidepres-

Table 1. Method of Data Collection in Each Country

Countries by continent

Number of doctorsinvolved in data

collection Stratification of sampling Reporting time

EuropeWestern Europe

The Netherlands 360 Region, speciality, community size 7 consecutive days per quarterBelgium 520 Region, speciality 7 consecutive days per quarterSwitzerland 304 Region, speciality 5 consecutive days per semesterAustria 300 Region, speciality 7 consecutive days per quarter

Southern EuropePortugal 679 Region, speciality 7 consecutive days per quarterGreece 474 Region, speciality 7 consecutive days per quarter

Northern EuropeFinland 400 Region, speciality, practice type 5 consecutive days per semesterIreland 200 Region 6 consecutive days per quarter

Eastern EuropeSlovakia 390 Speciality 7 consecutive days per semesterPoland 565 Speciality 7 consecutive days per semesterCzech Republic 470 Speciality 7 consecutive days per semesterHungary 540 Speciality 7 consecutive days per semester

AsiaSouth Korea 701 Regions, speciality, practice type 7 consecutive days per semesterThailand 440 Region, speciality 7 consecutive days per semesterIndonesia 450 Region, speciality 7 consecutive days per semesterPakistan 540 Region, speciality 7 consecutive days per semesterTurkey 705 Region, speciality 7 consecutive days per semesterSaudi Arabia 320 Region, speciality 7 consecutive days per quarter

Central AmericaGuatemala, Honduras,

El Salvador, Nicaragua,Costa Rica, and Panama

740 Country, speciality 7 consecutive days per semester

South AmericaColombia 385 Region, speciality, practice type 7 consecutive days per semesterVenezuela 500 Region, speciality 7 consecutive days per semester

AustraliaAustralia 420 Region, speciality 7 consecutive days per quarterNew Zealand 200 Region 7 consecutive days per quarter

AfricaEgypt 525 Region, speciality 7 consecutive days per quarterSouth Africa 384 Region, speciality, community size, location of general

practitioner’s practice (metropolitan or nonmetropolitan)7 consecutive days per quarter

3Wong et al./Variation of prescription rates for ASDINSAR

sants; (c) antipsychotics; (d) antiepileptics and mood sta-bilizers; (e) benzodiazepines; (f) sleep medications(including melatonin); (g) clonidine; (h) β-blockers; (i)antiparkinson medication; and (j) anxiolytics (buspironeand hydroxyzine).

Estimated Prescription Rate

The “average annual projected number of prescriptions”was calculated by the summation of the projectednumber of prescriptions between 2007 and 2012 (inclu-sive), followed by the division of the sum by 6. Theestimated annual prescription rates were calculated bydividing “average annual projected number of prescrip-tions” by the estimated annual population. The esti-mated annual population figures for each country in2010 were obtained from the United Nations, Depart-ment of Economic and Social Affairs (http://esa.un.org/wpp/excel-Data/population.htm). Using an average of6-year data might reduce random variation within thedata. As noted above, it has been reported that peoplewith mental health disorders receive less treatment inlow-/middle-income countries compared with high-income countries [MacArthur & The Academy of MedicalSciences, 2008]. In order to investigate whether low-/middle-income countries are associated with a lower pre-scription rate for ASD comorbidity treatment comparedwith high-income countries, the gross domestic product(GDP) per capita in U.S. dollars in 2010 was used as anindicator of each country’s living standard and its asso-ciation with ASD comorbidity treatment calculated. TheGDP per capita data were obtained from the UnitedNations, National Accounts Main Aggregates Data-base (http://unstats.un.org/unsd/snaama/dnllist.asp). ForCentral America, the GDP per capita was the mean of thesum of GDP per capita of the six countries.

Data Analysis

We analyzed the data by drug class and individual drugfor each country. Spearman correlation was used to inves-tigate the association between prescription rate and GDPper capita; this was selected (rather than Pearson correla-tion) because it is less sensitive to outliers, which mayoccur in relation to GDP per capita figures. Data manage-ment and analyses were performed using Microsoft Excel2010® (Microsoft US, Washington, United States), Statis-tical Analysis System (SAS) v9.3 (SAS Inc., Cary, NorthCarolina, USA), and Stata MP version 11.2 (StataCorp,College Station, TX, USA).

Results

There was a wide range of estimated prescription rates indifferent countries between 2007 and 2012 (Table 2). The

Western European countries (except Austria) were foundto have the highest prescription rate (3.89–36.36/10,000).The lowest prescription rates were observed in Egypt(0.74/10,000), Venezuela (0.83/10,000), and Asian coun-tries such as Turkey, Saudi Arabia, Indonesia, and Pakistan(0.04–0.82/10,000).

Figure 1 shows the proportion of different psychotropicclass for ASD treatment by country between 2007 and2012. Antipsychotics were the most commonly pre-scribed drug class for ASD comorbidity treatment inall countries except Austria, Finland, and New Zealand.In Finland, benzodiazepines (27.7%) and anxiolytics(22.1%) were the most commonly prescribed drug classesfor ASD comorbidity treatment. Antidepressants were thesecond most commonly prescribed drug class in Austra-lia, South Africa, and most of the European countriesexcept the Czech Republic, Austria, Belgium, Portugal,and Slovakia. Antiepileptics and mood stabilizers werethe second most commonly prescribed drug class in theCzech Republic, Hungary, Poland, Portugal, and Slovakia.

There was a significant positive relation between GDPper capita and prescription rate (Spearman’s coefficientρ = 0.60; P = 0.0011) (Fig. 2). In general, countries withlow GDP per capita have low estimated prescription rates.However, it is important to note that this was not uni-versally the case. For instance, some countries (e.g. Swit-zerland, Austria, Finland, and Australia) have high GDPper capita but low estimated prescription rates for ASDcomorbidity treatment; in contrast, Colombia and Thai-land have low GDP per capita but high estimated pre-scription rates.

Table 3 shows the most common drug prescribed forASD comorbidity treatment in each country. Risperidonewas a commonly prescribed drug in all countries exceptFinland. Buspirone (an anxiolytic) was the drug mostcommonly prescribed for ASD treatment in Finland.Stimulants (methylphenidate), atomoxetine, and anti-epileptic drugs (e.g. lamotrigine, carbamazepine, andvalproic acid) were also prescribed for treatment in mostcountries.

Discussion

There is considerable variation in prescription ratesamong countries. The variation of psychopharmacologi-cal prescription patterns may be explained by a numberof factors such as socioeconomic status, awareness orlevels of knowledge about ASD, perceptions toward ASDtreatment, lack of consistent guidelines for pharmaco-logical management of ASD, and differential access tohealth services. The difference in diagnostic guidance (i.e.ICD-10 and Diagnostic and statistical manual of mentaldiseases, 4th edition) and different health-care policiesbetween countries may also attribute to the variation of

INSAR4 Wong et al./Variation of prescription rates for ASD

prescribing patterns [Zito et al., 2008]. Variations in phy-sician specialties in different geographic locations alsoinfluence psychotropic drug use. In Western Europe,general practitioners prescribed most of the psychotropicdrugs [Zito et al., 2008]. It has been suggested that Britishphysicians are more conservative in prescribing psycho-tropic drugs than those in the United States [Murrayet al., 2014; Rani et al., 2008]. Diverse autism prevalencecan also result in wide variations of prescription ratesbetween countries. A systematic review on the globalprevalence of autism commissioned by the WHO hasshown high variation of autism prevalence across and

within countries. This has been reported to range from 13per 10,000 to 116 per 10,000 in European countries (theUnited Kingdom, Iceland, Denmark, and Sweden) frompopulation-based studies in 2007–2012. Similarly, thevariation of autism prevalence was also reported inAmerica (from 13 per 10,000 to 110 per 10,000) and Asia(2.8 per 10,000 to 181 per 10,000). The authors of theWHO review suggested that variation of ASD prevalencemay account for differential case ascertainment becauseof different diagnostic criteria, service availability, andawareness of the condition in both the lay and profes-sional public between countries [Elsabbagh et al., 2012].

Table 2. Estimated Prescription Rates for Autism Spectrum Disorder (ASD) Treatment in 2007–2012a

Country

Average projected numberof prescriptions for patients

with ASD per yearEstimated population

in 2010b

Estimated prescription rate(number of ASD prescriptions

per 10,000 people in thegeneral public)c

EuropeWestern Europe

The Netherlands 60,400 16,613,000 36.36Belgium 34,010 10,712,000 31.75Switzerland 11,764 7,664,000 15.35Austria 3,269 8,394,000 3.89

Southern EuropePortugal 10,690 10,676,000 10.01Greece 1,526 11,359,000 1.34

Northern EuropeFinland 2,150 5,365,000 4.01Ireland 2,656 4,470,000 5.94

Eastern EuropeSlovakia 3,056 5,462,000 5.59Poland 15,192 38,277,000 3.97Czech Republic 4,105 10,493,000 3.91Hungary 1,683 9,984,000 1.69

AsiaSouth Korea 69,184 48,184,000 14.36Thailand 69,010 69,122,000 9.98Indonesia 9,587 239,871,000 0.40Pakistan 651 173,593,000 0.04Turkey 5,941 72,752,000 0.82Saudi Arabia 1,389 27,448,000 0.51

Central AmericaGuatemala, Honduras, El Salvador,

Nicaragua, Costa Rica, and Panama4,924 42,146,000 1.17

South AmericaColombia 29,678 46,295,000 6.41Venezuela 2,400 28,980,000 0.83

OceaniaAustralia 6,103 22,268,000 2.74New Zealand 3,553 4,368,000 8.13

AfricaEgypt 6,039 81,121,000 0.74South Africa 6,631 50,133,000 1.32

aStudy drug includes: stimulants and atomoxetine; antidepressants; antipsychotics; antiepileptics and mood stabilizers; benzodiazepines; sleepmedication; clonidine; β-blockers; antiparkinsonians; and other anxiolytics.

bThe country classification by continents and population was obtained from United Nations, Department of Economic and Social Affairs, PopulationDivision, World Population Prospects. Available from http://esa.un.org/unpd/wpp/Excel-Data/population.htm.

cNumbers presented in two decimal places.

5Wong et al./Variation of prescription rates for ASDINSAR

Figure 1. Percentage of different psychotropic drug classes for the treatment of autism spectrum disorder. WE, Western Europe; SE,Southern Europe; NE, Northern Europe; EE, Eastern Europe; CA, Central America; SA, South America.

Figure 2. Scatter plot of estimated prescription rates for autism spectrum disorder (ASD) treatment vs. gross domestic products (GDPs)per capita in each country. GDP per capita at current prices in U.S. dollars in 2010 was obtained from United Nations, National AccountsMain Aggregates Database. Available from http://unstats.un.org/unsd/snaama/dnllist.asp.

INSAR6 Wong et al./Variation of prescription rates for ASD

Table 3. Common Drugs Prescribed for ASD Treatment in 2007–2012

Countries by region Drug

Projected numberof prescriptions

per year %

EuropeWestern Europe

The Netherlands Risperidone 22,696 37.6Aripiprazole 6,275 10.4Methylphenidate 6,144 10.2Total projected prescriptions 60,400 100.0

Belgium Risperidone 11,816 34.7Aripiprazole 4,036 11.9Pipamperone 2,671 7.9Total projected prescriptions 34,010 100.0

Switzerland Risperidone 1,902 16.2Aripiprazole 1,512 12.8Quetiapine 905 7.7Total projected prescriptions 11,764 100.0

Austria Lamotrigine 910 27.8Risperidone 654 20.0Atomoxetine 544 16.6Total projected prescriptions 3,269 100.0

Southern EuropePortugal Risperidone 3,709 34.7

Methylphenidate 659 6.2Valproic acid 631 5.9Total projected prescriptions 10,690 100.0

Greece Olanzapine 509 33.3Risperidone 241 15.8Clomipramine 207 13.5Venlafaxine 207 13.5Total projected prescriptions 1,526 100.0

Northern EuropeFinland Buspirone 451 21.0

Escitalopram 322 15.0Oxazepam 303 14.1Total projected prescriptions 2,150 100.0

Ireland Zuclopenthixol 497 18.7Risperidone 431 16.2Methylphenidate 354 13.3Total projected prescriptions 2,656 100.0

Eastern EuropeSlovakia Risperidone 747 24.4

Clonazepam 347 11.4Quetiapine 294 9.6Total projected prescriptions 3,056 100.0

Poland Risperidone 3,657 24.1Valproic acid 1,869 12.3Lamotrigine 1,614 10.6Total projected prescriptions 15,192 100.0

Czech Republic Risperidone 2,021 49.2Chlorprothixene 403 9.8Atomoxetine 308 7.5Total projected prescriptions 4,105 100.0

Hungary Carbamazepine 300 17.8Risperidone 296 17.6Zuclopenthixol 189 11.2Total projected prescriptions 1,683 100.0

AsiaSouth Korea Risperidone 32,792 47.4

Methylphenidate 4,285 6.2Valproic acid 3,266 4.7Total projected prescriptions 69,184 100.0

7Wong et al./Variation of prescription rates for ASDINSAR

Another systematic review also reported that diagnosticcriteria, age of the children screened, and study locationwere significantly associated with estimated prevalenceof ASD [Williams, Higgins, & Brayne, 2006]. As dataobtained in this study are limited, further investigation is

needed to determine the factors associated with the varia-tion of prescription rates for ASD comorbidity treatmentbetween countries.

In general, antipsychotic drugs were the most fre-quently prescribed drug class for patients with ASD, and

Table 3. Continued

Countries by region Drug

Projected numberof prescriptions

per year %

Thailand Risperidone 38,116 55.2Valproic acid 13,496 19.6Methylphenidate 5,113 7.4Total projected prescriptions 69,010 100.0

Indonesia Risperidone 3,472 36.2Aripiprazole 2,573 26.8Piracetam 1,224 12.8Total projected prescriptions 9,587 100.0

Pakistan Risperidone 651 100.0Total projected prescriptions 651 100.0

Turkey Risperidone 4,427 74.5Carbamazepine 372 6.3Valproic acid 239 4.0Lamotrigine 236 4.0Total projected prescriptions 5,941 100.0

Saudi Arabia Risperidone 898 64.7Carbamazepine 266 19.2Piracetam 75 5.4Total projected prescriptions 1,389 100.0

Central AmericaGuatemala, Honduras,

El Salvador, Nicaragua,Costa Rica, and Panama

Risperidone 2,608 53.0Haloperidol 432 8.8Piracetam 312 6.3Total projected prescriptions 4,924 100.0

South AmericaColombia Risperidone 15,257 51.4

Fluoxetine 3,835 12.9Aripiprazole 2,008 6.8Total projected prescriptions 29,678 100.0

Venezuela Risperidone 789 32.9Clonazepam 526 21.9Atomoxetine 289 12.1Total projected prescriptions 2,400 100.0

OceaniaAustralia Risperidone 2,859 46.8

Methylphenidate 552 9.0Sertraline 545 8.9Total projected prescriptions 6,103 100.0

New Zealand Methylphenidate 958 27.0Risperidone 556 15.6Melatonin 338 9.5Total projected prescriptions 3,553 100.0

AfricaEgypt Risperidone 3,792 62.8

Piracetam 632 10.5Carbamazepine 548 9.1Total projected prescriptions 6,039 100.0

South Africa Risperidone 3,158 47.6Citalopram 1,688 25.5Methylphenidate 924 13.9Total projected prescriptions 6,631 100.0

INSAR8 Wong et al./Variation of prescription rates for ASD

risperidone was commonly prescribed in all countries.There are two FDA-approved medications (risperidoneand aripiprazole) licensed for the treatment of irritabilityor aggression in patients with ASD. Risperidone is alsolabeled for the treatment of behavioral problems inpeople with ASD in some countries, including Europeancountries. In addition, risperidone has been on themarket longer than aripiprazole and is off-patent; there-fore, the generic products are more affordable thanaripiprazole [Almandil & Wong, 2011]. Research support-ing the use of risperidone has also been available longerthan aripiprazole. This may explain the popularity of thedrug for patients with ASD. However, risperidone is morelikely to cause adverse metabolic effects than aripiprazole[Almandil et al., 2013]. Some severe adverse drug reac-tions were also reported to be associated withantipsychotics [Rani, Byrne, Cranswick, Murray, & Wong,2011; Rani, Byrne, Murray, Carter, & Wong, 2009; Staret al., 2012]. Therefore, careful monitoring on prescribingantipsychotics is needed.

Stimulants and atomoxetine, antiepileptics, mood sta-bilizers, and antidepressants were also frequently pre-scribed for ASD comorbidity treatment in our study. Wedo not have information on individual comorbidities.Therefore, the actual indication for each of these medi-cations is uncertain. For example, carbamazepine can beused to treat epilepsy or mood swings. With limited dataon the efficacy and long-term adverse reactions of thesedrugs in patients with ASD, further research is warranted.

There was significant association between psychotropicdrug use for ASD comorbidity treatment and GDP percapita. Western Europe (except Austria) and Portugal hadhigher prescription rates compared with other Europeancountries. In Asia, Korea had exceptionally high prescrip-tion rates among the Asian countries. Of note, thesecountries had a higher GDP per capita compared with theothers. Our results indicate that, in general, countrieswith low GDP tend to have lower prescription rates. Lowprescription rates found in low-/middle-income countriescould be due to smaller health budgets for mental healthpolicy and services. We also conducted a separate analysison ten countries, which was reported in our previouswork [Hsia et al., 2014]. These include France, Germany,Italy, Spain, Mexico, Brazil, Japan, Canada, the UnitedKingdom and the United States. Due to differences in theduration of the previous work and this study, we areunable to combine the two datasets for analysis. Thecorrelation between the prescription rate and GDP percapita among the ten countries was also found to besignificantly positive (Spearman ρ = 0.68; 95% confi-dence interval 0.097–0.92; P = 0.027; Supporting Infor-mation Fig. S1). Our current study findings are supportedby similar correlation coefficients.

A study involving 191 countries reported that a smallerproportion of the health budget was allocated to mental

health in lower-income countries compared with higherincome countries. In addition, patients in many low-income countries rely on medication purchased withtheir own out-of-pocket expenditure [Saxena, Sharan, &Saraceno, 2003]. Compared with tax and insurance-basedhealth-care financing, out-of-pocket expenditure couldcreate financial burdens for patients and their families,and thus may lower the rate of access to health care inlow-/middle-income countries. Another survey reported asimilar association between the overall expenditure onhealth care and the percentage of patients receiving ser-vices [Wang et al., 2007]. In 2006, the Academy ofMedical Sciences showed that access to mental healthtreatment among patients living in lower-/middle-income countries was limited compared with higher-income countries [MacArthur & The Academy of MedicalSciences, 2008]. It also reported that low availability ofmental health professionals in low-/middle-incomecountries might have an impact on access to treatment.For instance, there is up to a 200-fold difference betweensome countries in the number of available psychiatrists,nurses, psychologists, and social workers. This means thaton average, there are five psychiatrists per million peoplein low-/middle-income countries compared with 40 psy-chiatrists per million people in the United Kingdom. Inaddition to limited access to health services, patients whowere treated for mental health disorders also received lessfollow-up care in low-/middle-income countries [Wanget al., 2007]. Reduced follow-up care could also contrib-ute to lower prescription rates in low-/middle-incomecountries. Further research is needed to investigate theassociation between ASD comorbidity treatment andhealth-care expenditure on mental health in differentcountries.

To our knowledge, this is the largest multinationalcomparison of psychopharmacological use in ASDcomorbidity treatment. There are, however, some notablelimitations. First, as IMS MIDAS is not a patient-basedmedical record database, we were unable to report thenumber of patients who visited practitioners for ASDcomorbidity treatment in this study. Consequently, wecannot estimate the treatment rates in patients with ASD.Second, our stratified sampling was intended to provide arepresentative sample for each country. Some patients,however, may be under the care of psychiatrists for themanagement of ASD-related symptoms, and our access tothe sample of practitioners represents a small percentageof psychiatrists. Therefore, the audit may have underes-timated the number of prescriptions. Third, individualswith ASD have very high comorbidity prevalence, such asADHD, anxiety, learning disabilities, and epilepsy[MacNeil et al., 2009; Simonoff et al., 2008]. We wereunable to obtain data on these comorbidities and treat-ment indications, and it is uncertain whether psychotro-pic drugs were prescribed for the treatment of ASD core

9Wong et al./Variation of prescription rates for ASDINSAR

symptoms or other psychiatric comorbidities. This limi-tation can be addressed in future studies using differentdatabases that contain patient medical records such asThe Health Improvement Network and Clinical PracticeResearch Datalink. Finally, despite the significant positivecorrelation between GDP per capita and prescriptionrate for ASD comorbidity treatment, potential confound-ers could not be adjusted due to limited informationin the dataset (e.g. age of patients and the use ofcomplementary/ alternative treatments).

Conclusion

There is wide variation in the psychopharmacologicaltreatment for patients with ASD among countries. Ingeneral, European countries have the highest prescriptionrates and Asian regions have the lowest. An associationalso exists between psychotropic drug use and GDP percapita in most (but not all) countries. Future researchshould incorporate more data including treatment rates,comorbidities, age of patients, and the use of comple-mentary or alternative treatments for ASD to obtain anin-depth understanding of the disparity of psychophar-macological treatment between countries.

Acknowledgments

This research was supported by the Innovative MedicinesInitiative (IMI) Joint Undertaking under grant agreementno. 115300: European Autism Interventions (EU-AIMS),which are composed of financial contributions from theEuropean Union’s Seventh Framework Programme (FP7/2007-2013), and the European Federation of Pharmaceu-tical Industries and Associations (EFPIA). D.G.M.M.,I.C.K.W and E. S. also received funding from the NationalInstitute of Health Research (UK) for a program grant onthis topic and for a neurodevelopmental theme in theBiomedical Research Centre at the Institute of Psychiatry,London, United Kingdom.

References

Almandil, N.B., Liu, Y., Murray, M.L., Besag, F.M., Aitchison, K.J.,& Wong, I.C. (2013). Weight gain and other metabolicadverse effects associated with atypical antipsychotic treat-ment of children and adolescents: A systematic review andmeta-analysis. Paediatric Drugs, 15, 139–150.

Almandil, N.B., & Wong, I.C. (2011). Review on the current useof antipsychotic drugs in children and adolescents. Archivesof Disease in Childhood. Education and Practice Edition, 96,192–196.

American Psychiatric Association. (1994). Diagnostic and statis-tical manual of mental diseases (DSM-IV) (4th ed.). Washing-ton, DC: American Psychiatric Publishing.

Baird, G., Simonoff, E., Pickles, A., Chandler, S., Loucas, T.,et al. (2006). Prevalence of disorders of the autism spectrumin a population cohort of children in South Thames:The Special Needs and Autism Project (SNAP). Lancet, 368,210–215.

Balkrishnan, R., Phatak, H., Gleim, G., & Karve, S. (2009). Assess-ment of the use of angiotensin receptor blockers in majorEuropean markets among paediatric population for treatingessential hypertension. Journal of Human Hypertension, 23,420–425.

Bradley, E., & Bolton, P. (2006). Episodic psychiatric disorders inteenagers with learning disabilities with and without autism.The British Journal of Psychiatry: The Journal of MentalScience, 189, 361–366.

Cadman, T., Eklund, H., Howley, D., Hayward, H., Clarke, H.,et al. (2012). Caregiver burden as people with autism spec-trum disorder and attention-deficit/hyperactivity disordertransition into adolescence and adulthood in the UnitedKingdom. Journal of the American Academy of Child andAdolescent Psychiatry, 51, 879–888.

Carbone, P.S., Farley, M., & Do, T.D. (2010). Primary care forchildren with autism. American Family Physician, 81, 453–460.

Coury, D.L., Anagnostou, E., Manning-Courtney, P., Reynolds,A., Cole, L., et al. (2012). Use of psychotropic medication inchildren and adolescents with autism spectrum disorders.Pediatrics, 130(Suppl. 2), S69–S76.

Elsabbagh, M., Divan, G., Koh, Y.J., Kim, Y.S., Kauchali, S.,et al. (2012). Global prevalence of autism and other perva-sive developmental disorders. Autism Research: OfficialJournal of the International Society for Autism Research, 5,160–179.

European Medicines Agency (EMA). (2007). Assessment ofthe paediatric needs psychiatry. Retrieved September 2,2013 from http://www.ema.europa.eu/docs/en_GB/document_library/Other/2009/10/WC500004046.pdf

Frazier, T.W., Shattuck, P.T., Narendorf, S.C., Cooper, B.P.,Wagner, M., & Spitznagel, E.L. (2011). Prevalence andcorrelates of psychotropic medication use in adolescentswith an autism spectrum disorder with and withoutcaregiver-reported attention-deficit/hyperactivity disorder.Journal of Child and Adolescent Psychopharmacology, 21,571–579.

Hollander, E., Phillips, A.T., & Yeh, C.C. (2003). Targeted treat-ments for symptom domains in child and adolescent autism.Lancet, 362, 732–734.

Hsia, Y., Wong, A.Y., Murphy, D.G.M., Simonoff, E., Buitelaar,J.K., & Wong, I.C. (2014). Psychopharmacological pre-scriptions for people with autistic spectrum disorder(ASD): A multinational study. Psychopharmacology, 231,999–1009.

Kogan, M.D., Blumberg, S.J., Schieve, L.A., Boyle, C.A., Perrin,J.M., et al. (2009). Prevalence of parent-reported diagnosis ofautism spectrum disorder among children in the US, 2007.Pediatrics, 124, 1395–1403.

Livermore, D.M., Reynolds, R., Stephens, P., Duckworth,G., Felmingham, D., et al. (2006). Trends in penicillin andmacrolide resistance among pneumococci in the UK andthe Republic of Ireland in relation to antibiotic sales to

INSAR10 Wong et al./Variation of prescription rates for ASD

pharmacies and dispensing doctors. International Journal ofAntimicrobial Agents, 28, 273–279.

Lund, C., Tomlinson, M., De Silva, M., Fekadu, A., Shidhaye, R.,et al. (2012). PRIME: A programme to reduce the treatmentgap for mental disorders in five low- and middle-incomecountries. PLoS Medicine, 9, e1001359.

MacArthur, G., & The Academy of Medical Sciences. (2008).Challenges and priorities for global mental health research inlow- and middle-income countries: Symposium report.Retrieved July 4, 2013 from http://www.eldis.org/go/topics/resource-guides/health-systems/governance-and-health&id=43964&type=Document#.UdUtFPk9Fio

MacNeil, B.M., Lopes, V.A., & Minnes, P.M. (2009). Anxiety inchildren and adolescents with autism spectrum disorders.Research in Autism Spectrum Disorders, 3, 1–21.

Mandell, D.S., Morales, K.H., Marcus, S.C., Stahmer, A.C., Doshi,J., & Polsky, D.E. (2008). Psychotropic medication use amongmedicaid-enrolled children with autism spectrum disorders.Pediatrics, 121, e441–e448.

Marcus, R.N., Owen, R., Kamen, L., Manos, G., McQuade, R.D.,et al. (2009). A placebo-controlled, fixed-dose study ofaripiprazole in children and adolescents with irritabilityassociated with autistic disorder. Journal of the AmericanAcademy of Child and Adolescent Psychiatry, 48, 1110–1119.

Matson, J.L., & Neal, D. (2009). Seizures and epilepsy and theirrelationship to autism spectrum disorders. Research inAutism Spectrum Disorders, 3, 999–1005.

McPheeters, M.L., Warren, Z., Sathe, N., Bruzek, J.L.,Krishnaswami, S., et al. (2011). A systematic review ofmedical treatments for children with autism spectrum disor-ders. Pediatrics, 127, e1312–e1321.

Murray, M.L., Hsia, Y., Glaser, K., Simonoff, E., Murphy,D.G., et al. (2014). Pharmacological treatments prescribedto people with autism spectrum disorder (ASD) inprimary health care. Psychopharmacology, 231, 1011–1021.

Myers, S.M., Johnson, C.P., & American Academy of PediatricsCouncil on Children with Disabilities. (2007). Managementof children with autism spectrum disorders. Pediatrics, 120,1162–1182.

Owen, R., Sikich, L., Marcus, R.N., Corey-isle, P., Manos, G., et al.(2009). Aripiprazole in the treatment of irritability in childrenand adolescents with autistic disorder. Pediatrics, 124, 1533–1540.

Rani, F., Murray, M.L., Byrne, P.J., & Wong, I.C. (2008). Epide-miologic features of antipsychotic prescribing to children andadolescents in primary care in the United Kingdom. Pediat-rics, 121, 1002–1009.

Rani, F.A., Byrne, P., Cranswick, N., Murray, M.L., & Wong, I.C.(2011). Mortality in children and adolescents prescribed anti-psychotic medication: A retrospective cohort study using theUK general practice research database. Drug Safety: An Inter-national Journal of Medical Toxicology and Drug Experience,34, 773–781.

Rani, F.A., Byrne, P.J., Murray, M.L., Carter, P., & Wong, I.C.(2009). Paediatric atypical antipsychotic monitoring safety(PAMS) study: Pilot study in children and adolescentsin secondary- and tertiary-care settings. Drug Safety: An

International Journal of Medical Toxicology and Drug Expe-rience, 32, 325–333.

Rosenberg, R.E., Mandell, D.S., Farmer, J.E., Law, J.K., Marvin,A.R., & Law, P.A. (2010). Psychotropic medication use amongchildren with autism spectrum disorders enrolled in anational registry, 2007–2008. Journal of Autism and Devel-opmental Disorders, 40, 342–351.

Saxena, S., Sharan, P., & Saraceno, B. (2003). Budget and financ-ing of mental health services: Baseline information on 89countries from WHO’s project atlas. The Journal of MentalHealth Policy and Economics, 6, 135–143.

Shea, S., Turgay, A., Carroll, A., Schulz, M., Orlik, H.,et al. (2004). Risperidone in the treatment of disruptivebehavioral symptoms in children with autistic and otherpervasive developmental disorders. Pediatrics, 114, e634–e641.

Simonoff, E., Pickles, A., Charman, T., Chandler, S., Loucas, T., &Baird, G. (2008). Psychiatric disorders in children with autismspectrum disorders: Prevalence, comorbidity, and associatedfactors in a population-derived sample. Journal of the Ameri-can Academy of Child and Adolescent Psychiatry, 47, 921–929.

Souders, M.C., Mason, T.B., Valladares, O., Bucan, M., Levy, S.E.,et al. (2009). Sleep behaviors and sleep quality in childrenwith autism spectrum disorders. Sleep, 32, 1566–1578.

Star, K., Iessa, N., Almandil, N.B., Wilton, L., Curran, S., et al.(2012). Rhabdomyolysis reported for children and adoles-cents treated with antipsychotic medicines: A case seriesanalysis. Journal of Child and Adolescent Psychopharmacol-ogy, 22, 440–451.

Wang, P.S., Aguilar-Gaxiola, S., Alonso, J., Angermeyer,M.C., Borges, G., et al. (2007). Use of mental health servicesfor anxiety, mood, and substance disorders in 17 countriesin the WHO world mental health surveys. Lancet, 370,841–850.

Williams, J.G., Higgins, J.P., & Brayne, C.E. (2006). Systematicreview of prevalence studies of autism spectrum disorders.Archives of Disease in Childhood, 91, 8–15.

Wong, I.C., & Murray, M.L. (2005). The potential of UK clinicaldatabases in enhancing paediatric medication research.British Journal of Clinical Pharmacology, 59, 750–755.

Wong, I.C., Murray, M.L., Camilleri-Novak, D., & Stephens, P.(2004). Increased prescribing trends of paediatric psychotro-pic medications. Archives of Disease in Childhood, 89, 1131–1132.

World Health Organization (WHO). (2012). The ICD-10 Classi-fication of Mental and Behavioural Disorders: Clinicaldescriptions and diagnostic guidelines. Retrieved May 24,2013 from http://www.who.int/classifications/icd/en/bluebook.pdf

World Health Organization (WHO). (2013). Application forinclusion to the 19th expert committee on the selectionand use of essential medicines: Risperidone. Retrieved August25, 2013 from http://www.who.int/selection_medicines/committees/expert/19/applications/Risperidone_24_A_Ad_Final.pdf

Zito, J.M., Safer, D.J., de Jong-van den Berg, L.T., Janhsen, K.,Fegert, J.M., et al. (2008). A three-country comparison of psy-

11Wong et al./Variation of prescription rates for ASDINSAR

chotropic medication prevalence in youth. Child and Ado-lescent Psychiatry and Mental Health, 2, 1–8.

Supporting Information

Additional Supporting Information may be found in theonline version of this article at the publisher’s web-site:

Figure S1. Scatter plot of estimated prescription ratesfor autism spectrum disorder (ASD) treatment vs. grossdomestic products (GDPs) per capita in ten countriesa

[Hsia et al., 2014].Table S1. The International Classification of Diseases(ICD), tenth version codes to identify autism spectrumdisorder (ASD) patients.

INSAR12 Wong et al./Variation of prescription rates for ASD