REVIEW

Counting Fetal Alcohol Spectrum Disorder in Australia: Theevidence and the challenges

LUCY BURNS1, COURTNEY BREEN1, CAROL BOWER3, COLLEEN O’ LEARY4 &ELIZABETH JANE ELLIOTT2

1National Drug and Alcohol Research Centre, University of New SouthWales, Sydney, Australia, 2Paediatrics and ChildHealth, University of Sydney and The Children’s Hospital atWestmead, Sydney, Australia, 3Telethon Institute for ChildHealth Research, Centre for Child Health Research, University ofWestern Australia, Perth, Australia, and 4Curtin HealthInnovation Research Institute—Population Health Research, Curtin University, Perth, Australia

AbstractIssues. Alcohol exposure in utero is associated with a range of adverse outcomes in pregnancy and can cause long-termdisability. Fetal alcohol spectrum disorder (FASD) is an umbrella term to describe a range of effects from prenatal alcoholexposure including fetal alcohol syndrome (FAS). Determining the prevalence of FASD is challenging. Approach. Thisnarrative review collates information on the prevalence of FASD in Australia and documents the various methods used forattaining estimates and the limitations of the available data. Key Findings. Birth prevalence of FASD is most commonlymeasured through clinic-based studies, passive surveillance systems and active case ascertainment.Alcohol use in pregnancy andFAS in Australia is predominantly monitored through passive surveillance systems and under-ascertainment of cases is likely.State- and territory-based studies have reported birth prevalence rates of FAS of between 0.01 and 0.68 per 1000 live births.Prevalence rates of FASD have not been estimated in Australia.As reflected in the international data, Australian studies havefound higher rates of FAS among some Indigenous communities. This likely reflects patterns of alcohol use and othersocioeconomic risk factors. Implications. Under-recognition of FASD reflects incomplete and inconsistent data collectionsrecording alcohol use in pregnancy, lack of awareness among health professionals and a lack of diagnostic and support services.Conclusion. Accurate measurement of FASD prevalence is crucial to inform policy, resource and service development in theareas of health, education, justice and community.There is a need for consensus on the collection and best use of data. [BurnsL, Breen C, Bower C, O’ Leary C, Elliott EJ. Counting fetal alcohol spectrum disorders in Australia: the evidence andthe challenges. Drug Alcohol Rev 2013;32:461–467]

Key words: FASD, alcohol, pregnancy.

Introduction

Fetal alcohol spectrum disorder (FASD) is anumbrella term to describe a range of disordersresulting from prenatal alcohol exposure including:partial FAS (pFAS), alcohol-related birth defects(ARBD) and alcohol-related neurodevelopmental

disorder (ARND) and fetal alcohol syndrome(FAS) at the most visible end of the spectrum. Anumber of challenges exist in establishing population-based prevalence of FASD including limitationsin prenatal alcohol screening, variations in datacollection methods and incomplete data collections[1].

Lucy Burns PhD, MPH, Grad Cert Health Policy, Senior Lecturer, Chief Investigator, Courtney Breen PhD, MPH, Research Fellow, Carol BowerMBBS, MSc, PhD, FAFPHM, Winthrop Research Professor, Colleen O’ Leary PhD, NHMRC Postdoctoral Research Fellow, Elizabeth JaneElliott MD MPhil FRACP FRCPCH FRCP, Professor, Consultant Paediatrician. Correspondence to Dr Lucinda Burns, National Drug andAlcohol Research Centre, UNSW, Sydney, NSW 2052, Australia. Tel: +1 02 9385 0258; Fax: +1 02 9385 0222; E-mail: l.burns@unsw.edu.au

Received 4 February 2013; accepted for publication 2 April 2013.

bs_bs_banner

R E V I E W

Drug and Alcohol Review (September 2013), 32, 461–467DOI: 10.1111/dar.12047

© 2013 Australasian Professional Society on Alcohol and other Drugs

International prevalence data

There is a wide range of FAS and FASD prevalanceestimates from international studies [1]. The popula-tion prevalence estimates for FAS range from 0.1 per1000 among Caucasians in the USA to 120 per 1000 inan isolated community in British Columbia [2]. Thedifferences between study estimates may reflect truedisparity in prevalence between locations or differencesin study methodologies (clinic based, active or passiveascertainment) including the diagnostic criteria used toassess FAS/FASD.

International studies have reported higher prevalenceamong some Indigenous populations [3], and higherrates of FAS have been shown to be associated withdisadvantage [4]. Recent active ascertainment studiesof school children have reported FASD rates rangingfrom 65.2 to 74.2 per 1000 children in the WesternCape of South Africa [5], 23.1–62.6 per 1000 livebirths in Italy and 8.1–14.8 per 1000 live births in theUSA [1]. It is argued that these higher prevalence ratesamong school children reflect higher general popula-tion rates than previously estimated. The authorssuggest FAS estimates of 2–7 per 1000 in typical, mixedracial and socioeconomic populations in the USA [1].

This review collates information on the prevalence ofFASD in Australia and documents, the variousmethods used for attaining estimates and the limita-tions of the available data.

Diagnosing fetal alchol spectrum disorders

To obtain accurate incidence and prevalence data for acondition, accurate data on the condition need to becollected and the condition diagnosed. The lack of anobjective diagnostic test may lead to subjectivity in thediagnostic process, thus increasing the risk of diagnosticmisclassification [6]. A diagnosis of FAS relies on atriad of features: characteristic facial dysmorphology;impaired growth prenatally and/or postnatally; andstructural and/or functional abnormalities of the centralnervous system (CNS). A diagnosis of FAS can bemade without a history of maternal alcohol use, if dif-ferential diagnoses have been excluded and the triad ofabnormalities consistent with FAS is identified [7].

In 1996, the Institute of Medicine published the firstdiagnostic criteria for FAS, pFAS, ARBD and ARND.Complete agreement of the diagnostic categories ofFASD has not been reached, and a number of diagnos-tic guidelines have been published in the past decade[8–12]. The guidelines have only subtle differences inthe diagnostic criteria for FAS and pFAS but use dif-ferent criteria and terminolgy for other disorders in theFASD spectrum [13].

Diagnosis is particularly problematic for disorders inthe FASD spectrum in which characteristic facialabnormalities are not present because the other prob-lems [such as growth restriction, CNS dysfunction andbirth defects] are not unique to FASD and may beassociated with factors such as low socioeconomicstatus and poor maternal nutrition [14]. Even the diag-nosis of FAS may be overlooked if facial abnormalitiesare not recognised at birth and, although they are mostdistinct in early childhood, they become less obvious inadolescents and adults [15]. The 4-Digit DiagnosticCode is one method that can be utilised to describechildren who do not have sufficient features to fulfil thediagnosis of FAS and enables objective description ofthe severity and mix of abnormalities in the otherFASD [8]. It includes an assesment of growth defi-ciency, FAS facial features, CNS damage and prenatalalcohol exposure and allocates a score of 1–4 to each.

Determining the prevalence of FASD

There are a number of ways in which the incidence andprevalence of FASD can be determined. The threemain methods are: clinic-based studies, passive systemsand active case ascertainment [1,16].

Clinic-based studies

Clinic-based studies examining alcohol use are gener-ally conducted in prenatal clinics of large hospitalswhere alcohol consumption data is collected from preg-nant women. This may involve collection of biologicalspecimens during pregnancy and the use of standardscreening instruments (e.g. the T-ACE [17], theTWEAK [18] or the AUDIT-C [19]). Some studiescollect data at multiple time points.

Antenatal clinic-based studies are advantageousbecause detailed maternal history data are collected(sometimes prospectively), they can include largenumbers of pregnancies with various levels of alcoholexposure and babies at risk can be referred for assess-ment for FASD. However, women at highest risk areless likely to attend prenatal clinics regularly, and manydo not attend at all, making access to the highest riskcases less likely. Data from paediatric and FASD clinicsare also subject to selection bias and may overestimateFASD rates. Many clinic-based studies have been con-ducted in publicly funded hospitals and clinics wheredisadvantaged populations predominate. These studiesmay overestimate the population prevalence of alcoholuse in pregnancy and FASD. Since FASD are mostoften diagnosed between the ages of three and 12 years,studies in neonates will underestimate the prevalence ofFASD unless there is long-term follow-up [16]. Thereare no clinic-based studies from Australia, although two

462 L. Burns et al.

© 2013 Australasian Professional Society on Alcohol and other Drugs

NHMRC-funded birth cohorts examining outcomesfollowing alcohol use in pregnancy are in progress[20,21].

Passive surveillance and record review systems

Passive surveillance systems use existing record or datacollections in a particular geographical area for a par-ticular timeframe. Records reviewed generally include:birth records, special registries for children with devel-opmental disabilities or birth defects, or the medicalcharts of hospitals and physicians. Passive surveillancestudies often use multiple types of records to identify asmany cases as possible, as cases of FASD are oftendocumented in more than one place over time. Themajor advantages of this method are that it uses existingrecords, is relatively inexpensive and is easier to under-take than more time-intensive methods. The disadvan-tages are that data on maternal alcohol use duringpregnancy are not routinely collected [22,23] and adiagnosis of FASD involves multiple assessments forindicators of physical, developmental and behaviouralabnormalities and application of diagnostic criteria.Many features are not obvious, or are difficult to iden-tify at particular ages. Consequently, the informationrequired to make a diagnosis may not be recorded onthese data systems, or may not be identified as relatedto prenatal alcohol exposure [16]. Passive surveillancesystems rely on reporting of cases with predefinedinclusion criteria by particular groups, for example,clinicians or midwives, to particular databases, forexample, birth defects registers. Data may be collectedprospectively in this way, but case ascertainment islower than with active methods of ascertainment, evenif reporting is mandatory. There have been studies ofFAS involving passive survelliance in Australia [24–27]but none include the full FASD spectrum. Recent Aus-tralian research has demonstrated the potential forlinkage of routinely collected administrative datasetsto be used to monitor FASD and has providedpopulation-based estimates the propotion of intellec-tual disability [28] and cerebral palsy [29] that areattributable to maternal alcohol-use disorders. Routinecollection of data on maternal alcohol use during preg-nancy would enhance the use of this method for sur-veillance of FASD.

Active case ascertainment methods

One method of active case ascertainment involves iden-tifying and recruiting children with possible diagnosesof FASD from populations. All children are examinedto determine the final diagnosis and the overall preva-lence and characteristics of FASD in the population.This method has at least three advantages. Firstly, the

primary focus is on finding children with FASD at anage at which an accurate diagnosis can be made byclinical specialists. Secondly, the active method ofrecruitment means a higher possibility of identifyingchildren with FASD.Thirdly, by studying total popula-tions, bias is reduced. Active case ascertainment there-fore produces the most complete assessment of theprevalence and characteristics of FASD in a particularpopulation. This method is very labour intensive, timeconsuming and costly and relies on skilled health pro-fessionals [7,16]. It has been used in remote Aboriginalcommunities in Western Australia in the Lililwanproject, in an entire population of children born in2002 and 2003 [30]. Of these children, ~50% wereexposed to high levels of alcohol during pregnancy, andFASD prevalence data are currently being analysed.Another approach to active case identification is totarget specialist groups, such as paediatricians andgeneticists, who are likely to make the diagnosis ofFASD and to use a reminder mechanism to promptcentral notification of new cases on a regular basis.Thismethod was used by the Australian Paediatric Surveil-lance Unit to identify incident cases of FAS seen bypaediatricians nationally. This type of system providestimely, detailed, clinical data from which minimumincidence and birth prevalence rates can be estimatedin the population of children seen by paediatricians.

Monitoring FAS in Australia

State prevalence data

In Australia, there are Birth Defects Registers in fourjurisdictions (Western Australa, Victoria, New SouthWales and South Australia). These registers are popu-lation based, passive surveillance systems established tomonitor defects detected during pregnancy or at birthor diagnosed in childhood. It is likely that there isunder-ascertainment of FAS through this method. Inthe NSW, register FAS are not included, and informa-tion on maternal alcohol consumption during preg-nancy is not routinely collected.

The first estimate of the birth prevalence of FAS inAustralia demonstrated the importance of using multi-ple data sources to increase ascertainment [25]. Whendata from the Birth Defects Registry and the RuralPaediatric Service database in Western Australia werelinked the birth prevalence of FAS increased by 38%from that estimated from the Birth Defects Registryalone, giving a rate of 0.02 per 1000 live births fornon-Indigenous children and 2.76 per 1000 for Indig-enous children [25] (Table 1). The Rural PaediatricService database collates information on children seenby the rural visiting paediatrician and was initiallydesigned as a patient recall system with informal nota-

Counting FASD 463

© 2013 Australasian Professional Society on Alcohol and other Drugs

tion of diagnosis.The diagnostic notation was reviewedfor FAS cases and was used to determine additionalcases not already registered in the Birth DefectsRegistry.

Two subsequent studies estimating birth prevalenceof FAS in the Northern Territory and Victoria reportedsimilar rates. The medical charts of all children seen atthe Royal Darwin Hospital in the Northern Territory

Table 1. Prevalence estimates of FASD in Australia

Region Study type Outcome

Rate per 1000

Totalpopulation

Non-Indigenous Indigenous

AustraliaWestern Australia Birth Defect

Register plus Rural PaediatricService 1980–1997 [25]

Passive surveillance: multiplesources of reports includinghospitals and privatepractitioners, Department ofHealth databases (midwives,mortality and hospitalmorbidity systems) andinvestigative and treatmentcentres

FAS 0.18 0.02 2.76

Western Australia Birth DefectRegister plus Rural PaediatricService 2000–2004 [24]

Passive surveillance: multiplesources of reports as above

FAS 0.4 — —

Northern Territory [26] Passive surveillance: retrospectivemedical case notes review

FAS 0.68 — 1.87

Northern Territory [26] Passive surveillance: retrospectivemedical case notes review

FAS pluspartial FAS

1.7 — 4.70

Victorian Perinatal DataCollection and Birth DefectsRegister [23]

Passive surveillance: multiplesources including 50% frommidwives and the other 50%from hospital inpatient andoutpatient listings, maternaland child health nurses,cytogenic laboratories, deathcertificates and autopsyreports, and privatepaediatricians.

FAS 0.01–0.03 0.01–0.03 —

South Australian Birth DefectsRegister 1986–2005 [27]

Passive surveillance: multiplesources including from doctorsand other health professionalsinvolved with; the care ofchildren with birth defects inhospital, special paediatricassessment, treatment andrehabilitation centres andprivate practices. ThePregnancy Outcome Unit ofSouth Australia Health, TheState Perinatal AutopsyService, Diagnostic servicesincluding laboratoriesdiagnosing cytogenetic orbiochemical abnormalities,and organ imagingdepartments.

FAS 0.02 — —

Australian PaediatricSurveillance Unit [31]

Active, prospective surveillance:reporting by child healthspecialists

FAS 0.06 0.004 0.146

FASD, fetal alcohol spectrum disorder.

464 L. Burns et al.

© 2013 Australasian Professional Society on Alcohol and other Drugs

over a 10-year period were reviewed to identify childrenwith FAS and partial FAS [26]. All children identifiedwere Indigenous. The estimated birth prevalence ofFAS was 1.9 per 1000 Indigenous live births and ofFAS plus partial FAS was 4.7 per 1000 Indigenous livebirths. The birth prevalence for the population overallwas 0.68 per 1000 live births for FAS and 1.7 per 1000live births for FAS plus partial FAS.

In Victoria, estimates of birth prevalence for FASwere calculated using linked data from two passive sur-veillance systems, the Victorian Perinatal Data Collec-tion and the Victorian Birth Defects Register [23].Birth prevalence for FAS was estimated at 0.01 to 0.03per 1000 live births in the general population. NoIndigenous cases of FAS were identified. Similar rateswere reported from South Australia using data from theSouth Australia Birth Defects Register, which recorded13 cases of FAS between 1986 and 2005 representing arate of 0.03 per 1000 births [27].

The Western Australian Birth Defects Registryrecorded 74 children born with FAS between 1980 and2002 and notified by 2003. Following a Royal Australa-sian College of Physicians workshop on FASD held in2004, an additional nine cases were notified for chil-dren born 1980–2003, representing a 15% increase inbirth prevalence. With increasing awareness of FASresulting from research studies, educational seminarsand a particular focus inWA on alcohol and pregnancy,the reported birth prevalence has increased over timefrom 0.1 per 1000 in 1980–1984 to 0.4 per 1000 for the5-year period 2000–2004 [24].

National incidence and prevalence data

The national incidence and birth prevalence figures forFAS were estimated using the Australian PaediatricSurveillance Unit, an active surveillance system withmonthly reporting of incident cases by child healthspecialists [31]. Between 2001 and 2004, a total of 92children with FAS was reported. The overall incidencewas 0.58 per 105 children aged less than 15 years (0.18in non-Indigenous and 8.11 in Indigenous children)and 1.14 per 105 children under five years (0.37 non-Indigenous and 14.60 in Indigenous children) perannum.

The overall birth prevalence (based on children bornduring the study period) was 0.06 per 1000 live births(0.004 per 1000 non-Indigenous children and 0.146per 1000 Indigenous children). Although the rates ofFAS are likely to have been underestimated, these arethe only prospective national data available on FASthroughout the world.

The reported birth prevalence and incidence of FASin Indigenous Australian children is much higher thanfor non-Indigenous children, a finding reflected in a

number of other Indigenous populations [3,7,25,26,32,33]. However, the birth prevalence of FAS for non-Indigenous Australians is one-tenth that reported forother countries [3,6]. This may reflect under-ascertainment of cases or the distribution of maternalrisk factors and/or lower alcohol intake or under-recognition of maternal alcohol use during pregnancy inthe non-Aboriginal population [22,23,25,34]. Theremay also be reluctance to diagnose conditions related toalcohol use in pregnancy because of fear of stigmatisingchildren and families and a lack of services and profes-sionals available to treat diagnosed individuals.

Challenges for monitoring FASD in Australia

Studies in Australia have predominantly been based onpassive surveillance systems, which have a number oflimitations, particularly under-ascertainment [16].Thislikely reflects under-diagnosis and under-reporting ofcases and lack of screening of children at risk fromalcohol exposure.

Screening

In the Victorian study, where estimates were based onpassive reporting, inadequate documentation of mater-nal alcohol use was a limitation [23]. In particular,there was no record of maternal alcohol use in eitherthe antenatal or infant records in 28% of the auditedpopulation and 39% of the cases with microcephaly[23]. Only 22% of the mothers of children born in1985–2006 who had a diagnosis of FAS recorded onthe Western Australian Register of DevelopmentalAnomalies had an alcohol-related diagnosis recordedduring pregnancy [22]. Australian research has shownthat fewer than half (45%) of health professionals sur-veyed routinely ask pregnant women about alcohol useduring pregnancy and only 23% of paediatricians do sowhen taking a pregnancy history [35,36]. A third ofWestern Australian health professionals surveyed saidthat lack of referral resources affected their practice ofassessing alcohol intake in pregnant women [36].

In each Australian state and territory the MidwivesData Collection provides perinatal data on obstetricconditions, procedures and outcomes, neonatal mor-bidity and birth defects for every birth of at least 20weeks gestation, or if gestation is unknown at least400 g birthweight. Alcohol consumption data duringpregnancy are not routinely collected, self-reportedalcohol use in pregnancy being included in onlyTasma-nia, Northern Territory and Australian Capital Terri-tory collections. Only data from the Northern Territoryare published [37]. There is difficulty linking prenatalalcohol exposure with diagnoses of FASD in the clinicalsetting and in establishing a causal relationship.

Counting FASD 465

© 2013 Australasian Professional Society on Alcohol and other Drugs

Diagnosis and reporting

There are documented barriers to health professionalsmaking the diagnosis of FASD, including lack of knowl-edge of the diagnostic criteria. The Australian govern-ment has recently funded development of a nationaldiagnostic instrument for FASD [38,39], which nowrequires testing for utility in the Australian setting.Reporting of FASD is required in only three state birthdefects registers (Western Australia, South Australiaand Victoria) and is not included as a reportable con-dition in the largest state (New South Wales). FASDneeds to be incuded in all registers. In addition to thepractical issues of reporting, there is a reluctance tomake the diagnosis of FASD for fear of stigmatising themother, child and/or family [35,36]. Health profession-als also express lack of confidence to manage a childwith FAS and cite lack of knowledge about where torefer children for confirmation of the diagnosis andmanagement and lack of proof of effective interventionsas barriers to making a FAS diagnosis [35,36].

Conclusion and recommendations

Accurate FASD prevalence data are required as knowl-edge of the size and nature of the problem is essential toinform and monitor prevention initiatives, service needsand service development. It is difficult to obtain preva-lence data in Australia as few data required for thediagnosis are routinely collected and clinicians lackknowledge and are reluctant to diagnose. There is norequirement to count or report FASD, birth defectregisters do not exist in many states, data on alcohol usein pregnancy is not routinely collected and recordedand there are no accepted national diagnostic criteriafor FASD.

There is a need for final consensus on what datashould be collected and an implementation plan devel-oped for the collection of this data.This would includethe selection of data items to be included in nationaladministrative datasets.This should be accompanied byefforts to improve the capacity and confidence of healthprofessionals through education and training to addressthis public health problem. In addition, further studiesusing active case ascertainment need to be undertakenin Australia.

Acknowledgement

The National Drug and Alcohol Research Centre at theUniversity of NSW is supported by funding from theAustralian Government under the Substance MisusePrevention and Service Improvements Grants Fund.

References

[1] May PA, Gossage JP, Kalberg WO, et al. Prevalence andepidemiologic characteristics of FASD from various

research methods with an emphasis on recent in-schoolstudies. Dev Disabil Res Rev 2009;15:176–92.

[2] Robinson G, Conry J, Conry R. Clinical profile and preva-lence of fetal alchol syndrome in an isolated community inBritish Columbia. CMAJ 1987;137:203–7.

[3] Egeland G, Perham-Hester K, Gessner B, Ingle D, Berner J,Middaugh J. Fetal alcohol syndrome in Alaska, 1977through 1992: an administrative prevalence derived frommultiple data sources. Am J Public Health 1998;88:781–6.

[4] Abel EL, Hannigan JH. Maternal risk factors in fetalalcohol syndrome: provocative and permissive influences.Neurotoxicol Teratol 1995;17:445–62.

[5] Viljoen D, Gossage J, Brooke L, et al. Fetal alcohol syn-drome epidemiology in a South African community: asecond study of a very high prevalence area. J Stud Alcohol2005;66:593–604.

[6] Abel E. An update on incidence of FAS: FAS is not an equalopportunity birth defect. Neurotoxicol Teratol 1995;17:437–43.

[7] Stratton K, Howe C, Bataglia FF, etal. Alcohol Syndrome:Diagnosis, epidemiology, prevention, and treatment. Wash-ington, DC: National Academy Press, 1996.

[8] Astley S. Diagnostic guide for fetal alcohol spectrum disor-ders: the 4-digit diagnostic code, 3rd edn. Seattle: Univer-sity of Washington, 2004.

[9] Astley S. Comparison of the 4-digit diagnostic code and theHoyme diagnostic guidelines for fetal alcohol spectrum dis-orders. [Erratum appears in Pediatrics. 2007 Jan;119(1):227]. Pediatrics 2006;118:1532–45.

[10] Bertrand J, Floyd RL, Weber M, et al. Fetal alcohol syn-drome: guidelines for referral and diagnosis. Atlanta:Centers for Disease Control and Prevention, 2004.

[11] Chudley A, Conry J, Cook J, Loock C, Rosales T, LeBlancN. Fetal alcohol spectrum disorder: Canadian guidelines fordiagnosis. CMAJ 2005;172 (5 Suppl):S1–S21.

[12] Hoyme H, May P, Kalberg W, et al. A practical clinicalapproach to diagnosis of fetal alcohol spectrum disorders:clarification of the 1996 Institute of Medicine criteria. Pedi-atrics 2005;115:39–47.

[13] Western Australian Department of Health. Fetal alcoholspectrum disorder model of care. Perth, WA: WA Depart-ment of Health, 2010.

[14] O’Leary C. Fetal alcohol syndrome: diagnosis, epidemiol-ogy, and developmental outcomes. J Paediatr Child Health2004;40:2–7.

[15] Streissguth A, Aase J, Clarren S, Randels S, LaDue R,Smith D. Fetal alcohol syndrome in adolescents and adults.JAMA 1991;265:1961–7.

[16] May PA, Gossage JP. Estimating the prevalence of fetalalcohol syndrome: a summary. Alcohol Res Health. 2001;25:159–68.

[17] Sokol RJ, Martier SS, Ager JW. The T-ACE questions:practical prenatal detection of risk-drinking. Am J ObstetGynecol 1989;160:863–8.

[18] Russell M. New assessment tools for drinking in pregnancy:T-ACE, TWEAK, and others. Alcohol Health Res World1994;18:55–61.

[19] Burns E, Gray R, Smith L. Brief screening questionnaires toidentify problem drinking during pregnancy: a systematicreview. Addiction 2010;105:601–14.

[20] Bucello C, Fielder H, Hutchinson D, et al. The Triple BStudy: pilot data on the association between drugand alcohol use in pregnancy and child development at oneand three years of age. Drug Alcohol Rev 2012;31 (S1):29.

466 L. Burns et al.

© 2013 Australasian Professional Society on Alcohol and other Drugs

[21] Halliday J. Asking Questions about Alcohol in Pregnancy(AQUA): pregnancy alcohol exposure measured in a longi-tudinal cohort study in Australia 5th International Confer-ence on Fetal Alcohol Spectrum Disorder 2013;VancouverCanada. 2013.

[22] O’Leary CM, Halliday J, Bartu A, D’Antoine H, Bower C.Alcohol-use disorders during and within one year of preg-nancy: a population-based cohort study 1985-2006. BJOG2013; DOI: 10.1111/1471-0528.12167.

[23] Allen K, Riley M, Goldfeld S, Halliday J. Estimating theprevalence of fetal alcohol syndrome in Victoria using rou-tinely collected administrative data. Aust N Z J PublicHealth 2007;31:62–6.

[24] Bower C, Rudy E, Callaghan A, Cosgrove P, Quick J.Report of the birth defects registry of Western Australia1980–2006. Report Number 14. Perth: King EdwardMemorial Hospital, 2007.

[25] Bower C, Silva D, Henderson T, Ryan A, Rudy E. Ascer-tainment of birth defects: the effect on completeness ofadding a new source of data. J Paediatr Child Health2000;36:574–6.

[26] Harris K, Bucens I. Prevalence of fetal alcohol syndrome inthe top end of the Northern Territory. J Paediatr ChildHealth 2003;39:528–33.

[27] Van Essen P, Gibson C, Scott H, Willoughby C, Chan A,Haan EA. 2005 annual report of the South Australian BirthDefects Register, incorporating the 2005 annual report ofprenatal diagnosis in South Australia. Adelaide: South Aus-tralian Birth Defects Register, Children, Youth andWomen’s Health Service, 2008.

[28] O’Leary CM, Leonard H, Bourke J, D’Antoine H, Bartu A,Bower C. Intellectual disability: population-based estimatesof the proportion attributable to maternal alcohol-use dis-order during pregnancy. Dev Med Child Neurol 2013;55:271–7.

[29] O’Leary CM, Watson L, D’Antoine H, Stanley F, Bower C.Heavy maternal alcohol consumption and cerebral palsy inthe offspring. Dev Med Child Neurol 2012;54:224–30.

[30] Elliott E, Latimer J, Fitzpatrick J, Oscar J, Carter M.There’s hope in the valley. J Paediatr Child Health2012;48:190–2.

[31] Elliott EJ, Payne J, Morris A, Haan E, Bower C. Fetalalcohol syndrome: a prospective national surveillance study.Arch Dis Child 2008;93:732–7.

[32] Sampson PD, Streissguth AP, Bookstein FL, et al. Incidenceof fetal alcohol syndrome and prevalence of alcohol-relatedneurodevelopmental disorder. Teratology 1997;56:317–26.

[33] May P. Fetal alcohol effects among North AmericanIndians. Alcohol Health Res World 1991;15:239–48.

[34] Peadon E, O’Leary C, Bower C, Elliott E. Impacts ofalcohol use in pregnancy: the role of the general practi-tioner. Aust Fam Physician 2007;36:935–9.

[35] Elliott E, Payne J, Haan E, Bower C. Diagnosis of foetalalcohol syndrome and alcohol use in pregnancy: a survey ofpaediatricians’ knowledge, attitudes and practice. J PaediatrChild Health 2006;42:698–703.

[36] Payne J, Elliott E, D’Antoine H, et al. Health professionals’knowledge, practice and opinions about fetal alcohol syn-drome and alcohol consumption in pregnancy. Aust N Z JPublic Health 2005;29:558–64.

[37] Tew K, Zhang X. Northern Territory Midwives’ collection:mothers and babies 2006. Darwin: Department of Healthand Families, 2010.

[38] Watkins R, Elliott E, Mutch R, et al. Health professionalsperceptions about the adoption of existing guidelines for thediagnosis of fetal alcohol spectrum disorders in Australia.BMC Pediatrics. 2012;12:69.

[39] Watkins R, Elliott E, Mutch R, et al. Consensus diagnosticcriteria for fetal alcohol spectrum disorders in Australia: amodified Delphi study. BMJ Open 2012;25:5.

Counting FASD 467

© 2013 Australasian Professional Society on Alcohol and other Drugs