Burden of fetal alcohol syndrome in a rural West Coast area of South Africa
BA Hons (Psych);
Trop Med, MPh, PhD;
1 Foundation for Alcohol-Related Research (FARR), Cape Town, South Africa
2 Division of Molecular Biology and Human Genetics, Faculty of Health Sciences, Tygerberg Hospital and Stellenbosch University, Tygerberg, South Africa
3 Centre for Health Policy, School of Public Health, Faculty of Health Sciences, University of the Witwatersrand, Johannesburg, South Africa
International Centre for
Reproductive Health, Department of Obstetrics and Gynaecology,
University of Ghent, Belgium
Background. Fetal alcohol syndrome (FAS) is common in parts of South Africa; rural residence is a frequently cited risk factor. We conducted a FAS school prevalence survey of an isolated rural community in a West Coast village of Western Cape Province, so obtaining the first directly measured rate, focusing specifically on a South African rural area, of FAS and partial FAS (PFAS).
Methods. The study area (Aurora village), a community of about 2 500 people in a grain-producing region, has one primary school. All learners were eligible for study inclusion. Initial anthropometry screening was followed by a diagnostic stage entailing examination by a dysmorphologist for features of FAS, neurodevelopmental assessment, and an interview assessing maternal alcohol consumption.
Results. Of 160 learners screened, 78 (49%) were screen-positive, of whom 63 (81%) were clinically assessed for FAS. The overall FAS/PFAS rate among the screened learners was 17.5% (95% confidence interval 12.0 - 24.2%), with 16 (10.0%) children having FAS and 12 (7.5%) PFAS. High rates of stunting, underweight and microcephaly were noted in all learners, especially those with FAS or PFAS. Five (18%) mothers of affected children were deceased by the time of assessment.
Conclusion. We describe very high rates of
FAS/PFAS in an isolated rural part of the Western Cape that is
not located in a viticultural region. Our study suggests that
the prevalence of FAS may be very high in isolated
communities, or in particular hot-spots. It adds to the
growing evidence that FAS/PFAS is a significant, and
underestimated, health problem in South Africa. Expanded
screening and surveillance programmes, and preventive
interventions, are urgently needed.
S Afr Med J 2013;103(6):402-405.
Fetal alcohol syndrome (FAS) and the less severe fetal consequences of maternal drinking, collectively described as fetal alcohol spectrum disorders (FASD) than FAS/partial FAS (PFAS), are endemic in impoverished communities of the Western and the Northern Cape provinces of South Africa (SA).1-4 Over the past decade, levels of FAS/PFAS ranging between 40.5 and 119.4 per 1 000 have been documented in several parts of these provinces.1 , 4 A recent study examined the prevalence of FASD (which is a broader and less well defined group) in an area of the Western Cape, and estimated the prevalence at between 135.1 and 207.5 per 1 000.5
Heavy alcohol intake during pregnancy, usually in a binge-drinking pattern, occurs in over 20% of women in the Western Cape.6 Drinking during pregnancy remains a problem despite considerable media attention, efforts by the national government to limit alcohol harm,7 and isolated initiatives to increase universal and other prevention activities in these provinces.8 Heavy drinking has been socially entrenched and ‘normalised’ through a history including, but not restricted to, the dop system. For many women, the underlying social determinants of heavy alcohol consumption remain unchanged, and include poor socio-economic conditions, single-parent families, low levels of maternal education, concomitant use of tobacco and other substances, low religiosity and lack of alternative recreational opportunities.6 , 9
Most surveys have been conducted in small towns, and have
consistently shown that FAS/PFAS is more common where mothers
were resident in a rural area at the time of pregnancy.1-3
To date, no studies have directly documented FAS/PFAS rates in
entirely rural or remote parts of the country. In 2008,
community leaders of a rural area in the West Coast region of
the Western Cape invited our research team to assess levels of
FAS/PFAS, as well as substance use, in pregnancy. The study
therefore aimed to document the prevalence and risk factors
for FAS/PFAS in this isolated rural community, and to provide
information to assist the community to advocate for increased
resources to counter substance use and related problems in the
The study site is a rural village (Aurora) of approximately 2
500 inhabitants. They work mostly as seasonal labourers at
surrounding potato and grain farms. There is one school in the
area (grades 0 - 7), at which the parents/guardians of all
learners were invited to enrol in the study. Workshops were held
with approximately 220 people to generate awareness of the
study, with involvement of community and local government
leaders, including social workers, police officers, health
workers and educators from the school. The school principal and
educators also attended a course on how to manage children with
FASD in the classroom.
FAS diagnostic process
Prevalence of FAS and PFAS was determined by active case ascertainment using a validated two-tier screening method.1 This entails an initial screening stage, followed by a comprehensive diagnostic stage (clinical evaluation, neurocognitive assessment and maternal interview). In the first stage, pupils were screened by a professional nurse who measured height, weight and head circumference (HC), using standard methods. Children ≤10th percentile of the National Center for Health Statistics charts for height and weight, or ≤10th percentile for HC, were considered screen-positive and invited to the second stage.
The diagnostic stage entailed assessment by an experienced dysmorphologist (DV) for clinical features of FAS.10 The clinical diagnosis of FAS/PFAS was substantiated by a battery of neurodevelopmental tests. Positive evidence of neurological abnormality required a HC ≤10th percentile, the presence of ‘hard’ signs on neurological examination, or significant abnormalities on neurocognitive assessment (below average in more than 4 of 10 scales) using primarily the Griffiths Mental Development Scale.11
Trained interviewers completed a semi-structured questionnaire with the mother/guardian of screen-positive children. Interviews were in the home language of participants and focused on drinking habits during the index pregnancy, which were elicited using a timeline follow-back method.12
The final diagnosis, based
on criteria published by Hoyme et
al. (2005),10 encompassed all data from the
clinical evaluation, the neurodevelopmental profile and
maternal history. Final diagnostic categories were FAS and
PFAS. The clinical diagnosis of FAS, and with less certainty
PFAS, is considered distinctive even in the absence of a
history of maternal alcohol use.
Study ethics and data management
Permission was obtained from the Department of Education (provincial and district) to evaluate learners and informed consent was given by the parents or legal guardians of each child. If consent was given but the child was not present for clinical or neurodevelopmental assessment, 2 further attempts were made to assess the child. Similarly, 3 attempts were made to secure maternal interviews.
A trained counsellor communicated the FASD diagnosis to the parents or guardians, as well as its implications and a proposed intervention tailored to each child. Children identified as having FAS/PFAS were referred to local government services, including speech and hearing therapists, occupational therapists, physiotherapists and social workers. With parental permission, school personnel were informed of each child’s FASD status to facilitate educational support and further remedial assessment. Children with other medical conditions were referred to local and regional medical services, as required.
Intercooled Stata 12.1
(Stata Corporation, College Station, USA) was used for
statistical analysis. For analysis of categorical variables,
the chi-square test was used, while for continuous variables
we used an unpaired Student’s t-test or Mann-Whitney U-test for
normally and non-normally distributed data, respectively.
Consent for participation in the study was sought for all 171
learners in the school, of whom 160 consenting learners were
available for anthropometric screening. Of the 160 screened
learners, 78 (49%) were screen-positive on anthropometry, with
60 (38%) falling below the 10th percentile for both height and
weight. Status regarding FAS/PFAS diagnosis was determined for
63 children (81% of screen-positives) (Table 1). Fifteen
learners could not be clinically assessed because they were
absent from school on all follow-up visits made to assess them.
The screened group, who were all of coloured ethnicity, were aged 4.8 - 16.4 years. About 9% (14/160) of learners were older than 14 years. Exactly half the pupils were male, though females predominated in grades 0 and 1 (62%, 32/52; p=0.04).
In the group of 160 screened learners, 16 FAS and 6 PFAS cases (13.8%; 95% CI 8.8 - 20.0) were ascertained based on full clinical assessment Mean dysmorphology score for the 16 confirmed FAS children was 17.1 (SD)±2.2, and 9.1±4.5 for children with PFAS. A further 6 were suspected of having PFAS, based on facial features of FAS, HC <10th percentile, and a history of maternal alcohol use, but did not have a neurodevelopmental assessment. Despite the lack of this assessment, they had sufficient features to meet the Hoyme criteria for PFAS or FAS.9 They were assigned as PFAS cases, and included as such for further analysis, bringing the total to 28 affected children (17.5%; 95% CI 12.0 - 24.2).
No associations were detected between having FAS/PFAS and gender or grade. Children with FAS/PFAS were a median 8.5 years old, while other children were a median 10.0 years (p=0.09). The median body mass index of children without FAS/PFAS was 15.8 kg/m2 (IQR 14.6 - 17.5), compared with a median 14.7 kg/m2 in children with FAS/PFAS (IQR 13.7 - 15.6; p=0.001). The entire cohort of children, and especially those with FAS/PFAS, showed very high levels of stunting and were underweight for age (Table 1).
Five (18%) of the mothers of the
28 subjects with FAS/PFAS had died, and in 2 instances the
father was also known to have died. The study team referred 20
children to the school psychologist.
Researchers have previously noted that levels of FAS/PFAS in parts of the Western and Northern Cape provinces are several times higher than elsewhere in the world.3 Our data represent the first directly measured FAS/PFAS prevalence data from an entirely rural community in SA, and demonstrate levels that are high even when compared with similar local studies.
Table 2 shows FAS rates across
published school prevalence studies in SA. Although direct
comparisons between studies can be misleading, all these school
surveys used similar methodologies, except that ours is the only
study to include learners beyond first grade (there was,
however, no significant difference in the prevalence of FAS
across school grades).
Though the FAS rate of 10% in Aurora is very high, the true scale of overall FASD is certainly considerably higher. A further 7.5% of children were assessed as having FAS/PFAS, although this assessment was complicated by some participants not completing all study procedures, owing mostly to truancy from school (teachers suspected a high FASD rate among truant children). In addition to study-specific issues, it is likely that FASD prevalence studies inherently underestimate its prevalence because of the difficulties in diagnosing subjects without typical facial features, and in older children and adolescents as children age.13 In this regard, the high rate of non-syndromic microcephaly found in ‘unaffected’ children may relate, in part, to sub-clinical cases of FASD.
The high prevalence of FAS/PFAS in the village of Aurora is consistent with: (i) the rural-urban gradient in FAS prevalence shown in Table 2, (ii) studies that show a higher FAS/PFAS rate in children born in rural than urban areas,1-4 and (iii) evidence that binge drinking is more prevalent in rural (20%) than urban informal (16%) and urban formal (15%) areas.15 FASD also appears more prevalent in very isolated communities – the communities with the highest published prevalence of FAS/PFAS (Aurora and De Aar) are very isolated geographically. Conversely, high rates do not appear to relate to the presence of viticulture in an area.
Despite mounting evidence from school surveys and other sources that FASD is a significant public health problem, it remains under-recognised. For example, the Western Cape Burden of Disease study considered FASD only inasmuch as it is a cause of low birthweight, rather than as a significant cause of morbidity, mortality and disability in its own right.16 In part because it is under-recognised, there is a dearth of local measures aimed at preventing FASD, despite international acceptance of the utility of a range of interventions to reduce alcohol-related harm, ranging from policy and community level approaches to setting-specific approaches such as ‘brief interventions’ in prenatal care.17 As a local example of a successful intervention, we have reported a 30% reduction in FAS/PFAS rates associated with a community-level intervention comprising intensive ‘universal prevention’ measures.8
Interventions within the health sector that should be implemented, upgraded or urgently investigated include: (i) strengthened family planning services; (ii) improved antenatal education regarding the risks of drinking; (iii) identification of risky drinking in pregnancy and implementation of brief interventions to reduce it; (iv) strategies aimed at early identification of FASD to allow early intervention in the child and prevention of recurrence in future children; and (v) national-level surveillance of risky drinking and FASD rates to improve data on the problem of FASD.
Table 2 indicates that, although there is a high level of FAS in many parts of SA, the prevalence varies. A systematic approach is needed for determining overall FASD levels in the country that is both wide in scope but also provides local actionable detail in higher-risk areas. We recommend regular surveillance of a systematic sample of schools and/or antenatal clinics across the country, which would allow provincial and national baseline rates to be established and monitored over time, and would inform more intensive and cost-effective targeting of communities at highest risk.
We found that nearly 1 in 5 of mothers to children with FAS/PFAS were deceased. This phenomenon has not been studied in detail to date, although it is noteworthy that the 3 Wellington studies1-3 reported significant maternal mortality in mothers of FAS/PFAS children. Collectively, they report 13 (7%) maternal deaths among 185 mothers of children with FASD by the time the index cases were assessed at about 7 years of age. Causes of death, where known, appeared to be potentially associated with alcohol use, including accidents such as house fires; homicide and other violent death; pulmonary tuberculosis; and liver disorders. These data are consistent with evidence that binge drinking, in addition to being associated with FASD, is also strongly associated with unintentional injury, interpersonal violence, unsafe sex and other negative health consequences for adults.19 , 20 A case of FASD should therefore be considered as a marker for adverse maternal health consequences.
This study aims to benefit the local population, NGOs and community-based organisations, as well as inform government departments, specifically Health, Education, Social Services and Agriculture. The study heightened the community’s awareness of substance abuse problems and FASD in particular. Although this district has not featured on the government’s priority list for high-risk substance abuse areas and/or other health needs to date, the community specifically requested that the study findings be publicised and used to rectify this omission. Findings will also inform development of a comprehensive NGO-led intervention to address these local problems.
Acknowledgements. We acknowledge the dedication and invaluable contributions from the children and their parents, community members and leaders in Aurora, the school principal and educators, the Women’s Agricultural Society and the Women’s Forum. Special acknowledgment goes to FARR staff (Elizabeth Pegram, Yumna Martin, Debbie Lombard, Lian Drotsky, Karina Coetzer, Safia Abrahams and Lebo Khusu) who were responsible for the screening and neurodevelopmental assessments, maternal interviews, administration, data capturing and training conducted during this project. The Department of Education is thanked for the permission granted and willingness to entrust the study to FARR. The project was funded by the Department of Education and TopSpar. Pam Tilley initiated the invitation to FARR to assess the problem in Aurora. If not for her, this study would not have taken place.
Conflict of interest statement. The FARR receives funding from the Industry Association for Responsible Alcohol Use (ARA) and donations from the Whiskey Live Festival.
Authorship. L Olivier and M Urban are equal
1. May PA, Brooke LE, Gossage JP, et al. Epidemiology of fetal alcohol syndrome in a South African community in the Western Cape Province. Am J Public Health 2000;90(12):1905-1912. [http://dx.doi.org/10.2105%2FAJPH.90.12.1905]
2. Viljoen DL, Gossage JP, Brooke L, et al. Fetal alcohol syndrome epidemiology in a South African community: a second study of a very high prevalence area. J Stud Alcohol 2005;66(5):593-604.
3. May PA, Gossage JP, Marais AS, et al. The epidemiology of fetal alcohol syndrome and partial FAS in a South African community. Drug Alcohol Depend 2007;88(2-3):259-271. [http://dx.doi.org/10.1016%2Fj.drugalcdep.2006.11.007]
4. Urban M, Chersich MF, Fourie LA, et al. Fetal alcohol syndrome among grade 1 schoolchildren in Northern Cape Province: prevalence and risk factors. S Afr Med J 2008;98(11):877-882.
5. May PA, Blankenship J, Marais A-S, et al. Approaching the prevalence of the full spectrum of fetal alcohol spectrum disorders in a South African population-based study. Alcohol Clin Exp Res 2012 Dec 14. (Epub ahead of print) [http://dx.doi.org/ 10.1111/acer.12033] (accessed 19 March 2013).
6. Croxford J, Viljoen D. Alcohol consumption by pregnant women in the Western Cape. S Afr Med J 1999;89(9):962-965.
7. South African Revenue Services. Excise duties and levies 2012. http://www.sars.gov.za/home.asp?pid=483 (accessed 1 May 2012).
8. Chersich MF, Urban M, Olivier L, et al. Universal prevention is associated with lower prevalence of fetal alcohol spectrum disorders in Northern Cape, South Africa: a multicentre before-after study. Alcohol Alcohol 2012; 7(1):67-74. [http://dx.doi.org/10.1093%2Falcalc%2Fagr145]
9. Viljoen D, Croxford J, Gossage JP, et al. Characteristics of mothers of children with fetal alcohol syndrome in the Western Cape Province of South Africa: a case control study. J Stud Alcohol 2002;63(1):6-17.
10. Hoyme HE, May PA, Kalberg WO, et al. A practical clinical approach to diagnosis of fetal alcohol spectrum disorders: clarification of the 1996 Institute of Medicine Criteria. Pediatrics 2005;115:39-47. [http://dx.doi.org/10.1542%2Fpeds.2005-0702]
11. Griffiths R. The Ability of Young Children: A Comprehensive System of Mental Measurement for the First Eight Years of Life. Oxford, UK: The Test Agency, 1970.
12. Sobell LC, Agrawal S, Annis H, et al. Cross-cultural evaluation of two drinking assessment instruments: alcohol timeline followback and inventory of drinking situations. Subst Use Misuse 2001;36(3):313-331. [http://dx.doi.org/10.1081%2FJA-100102628]
13. Alberta Partnership on Fetal Alcohol Syndrome. Diagnosis of fetal alcohol syndrome. Can Child Adol Psych Rev 2003;12(3):81-86.
14. Viljoen D, Craig P, Hymbaugh K, et al. Fetal alcohol syndrome - South Africa 2001. MMWR 2003;52(28):660-662.
15. Harker N, Kader R, Myers B, et al. Substance abuse trends in the Western Cape, a review of studies conducted since 2000. http://www.sahealthinfo.org/admodule/substance.pdf (accessed 22 April 2012).
16. Sanders D, Reynolds L, Eley B, et al. Western Cape Burden of Disease Reduction Project Volume 7: Decreasing the Burden of Childhood Disease, 2007. http://www.westerncape.gov.za/eng/pubs/reports_research/W/157844 (accessed 20 May 2012).
17. Monteiro M. The road to a World Health Organization global strategy for reducing the harmful use of alcohol. Alc Res Health 2011;34(2):257-260.
18. Chang G. Screening and brief intervention in prenatal settings. Alcohol Res Health 2005;55:80-84.
19. Naimi TS, Brewer RD, Mokdad A, et al. Binge drinking among US adults. JAMA 2003;289(1):70-75. [http://dx.doi.org/10.1001%2Fjama.289.1.70]
20. Razvodovsky YE. Contribution of alcohol in accident related mortality in Belarus: a time series approach. J Inj Violence Res 2012;4(2):58-64.
Full text views: 7208