National Drug Strategy
National Drug Strategy

Intergovernmental committee on Drugs working party on Fetal Alcohol Spectrum Disorders


Fetal Alcohol Spectrum Disorders in Australia: An Update

June 2012

4.2 Patterns of maternal alcohol consumption and risk to the fetus

prev pageTOC |next page

Table of contents

It is generally accepted that the principal determinant of functional deficit is the dose and frequency of alcohol consumption but alcohol dependence is not required for alcohol-related problems to occur in the prenatally exposed child (Jacobson and Jacobson 1999). Expression of the full clinical features of FAS results from large amounts of alcohol consumed during early pregnancy. The mothers often have a history of either chronic heavy alcohol use or frequent intermittent heavy alcohol use.

It is important to recognize that not all children exposed to high levels of alcohol in utero will be negatively affected or affected to the same degree, indicating that expression of the anomalies requires the presence of other ‘component’ factors. A number of component factors have been identified, including the pattern and quantity of alcohol, the stage of development of the fetus at the time of exposure (the first 3-6 weeks of embryonic development identified as a critical time for the full teratogenic effects to occur) and sociobehavioural risk factors such as poverty, smoking, maternal age of 30 years or more and increasing parity (O'Leary 2004).

The lack of a universally accepted definition for a standard drink adds a level of complexity when comparing studies across countries. For instance, in the UK the term ‘unit’ is used rather than standard drink and reflects 8 grams of alcohol. A standard drink is 10 g alcohol in Australia, France, Hungary, Ireland, New Zealand, Poland and Spain; it is 12g Denmark, Italy and South Africa; 13.6g in Canada; and 14g in the USA and Portugal. To further complicate our ability to compare studies is the variation in the definition of ‘binge’ drinking according to the study. The conventional definition of binge drinking for women is five or more standard drinks per occasion, equating to 50g or more per occasion (Strandberg-Larsen et al. 2008; O'Leary and Bower 2009). However, a UK study defined binge drinking as four or more UK units per occasion (32g or more) (Sayal et al. 2009), equivalent to three Australian standard drinks or higher.

4.2.1 Binge drinking and risk to the fetus

A weekly binge pattern of alcohol consumption (5+ drinks/occasion) during pregnancy is reported to contribute 80 percent of functionally impaired infants (Jacobson and Jacobson 1999). However, a systematic review of the research on binge drinking and fetal outcomes found no convincing evidence of adverse effects, except possibly in relation to neurodevelopment (Gray and Henderson 2006; Henderson et al. 2007). However, these authors noted that methodological issues (discussed in section 5.3) precluded them from determining the impact on the fetus from occasional binge drinking.

Since the systematic review by Henderson et al. (Henderson et al. 2007), there have been a number of studies to support an increased risk to the fetus from exposure to a binge pattern of alcohol consumption. Studies using the conventional definition of binge drinking (5+ drinks/occasion) have demonstrated increased risk of stillbirth (Strandberg-Larsen et al. 2008) and language delay following occasional (weekly or less frequently) binge drinking in late pregnancy (O'Leary et al. 2009). Increased risk of hyperactivity and inattention were reported in a UK study following four or more UK units per occasion during pregnancy (Sayal et al. 2009). An Australian study found increased risk of anxiety/depression and aggressive behaviour following prenatal exposure to 3-4 standard drinks/occasion and no more than 70g/week (i.e. one to two drinking sessions per week). However, this pattern of drinking was defined as ‘moderate’ drinking in the Australian study (O'Leary et al. 2010).

4.2.2 Effects of low to moderate alcohol exposure in pregnancy

In contrast to the evidence of increased risk of fetal effects and poor pregnancy outcomes from heavy prenatal alcohol exposure, the effect from low to moderate levels of prenatal alcohol exposure remains unclear. Debate about the exact nature of the relationship between prenatal alcohol exposure and fetal effects is ongoing (Nathanson et al. 2007; O'Brien 2007). Whether there is a threshold effect below which there is no harm to the developing child is yet to be fully determined (Henderson et al. 2007; O'Leary and Bower 2011; Andersen et al. 2012).

A number of reviews of the literature on alcohol use and pregnancy outcomes have been conducted and are summarised in Table 4.4. Most research findings do not support a relationship between low levels of alcohol consumption and fetal growth abnormality, preterm birth, stillbirth, malformations, abnormal neurodevelopment and leukaemia.

The exception relates to an increase in risk of spontaneous abortion/miscarriage reported by both the Makarechian and the Henderson reviews (Makarechian et al. 1998; Henderson et al. 2007). Henderson, however, cautions about this finding as two of the five studies that showed an increased risk due to alcohol use had significant limitations. In one of these studies the women were also heavy smokers and the results of two of the studies were only of borderline statistical significance (Henderson et al. 2007). However, a large cohort study published after the Henderson review found that even 2–3˝ alcoholic drinks per week during early pregnancy increased the risk of spontaneous abortion, indicating that the fetus is particularly susceptible to alcohol exposure in early pregnancy. The authors found no increased risk of fetal death after 16 weeks of pregnancy at low levels of exposure (Andersen et al. 2012). Moderate levels of prenatal alcohol exposure have also been reported to increase the risk of neonatal asphyxia (defined as 2-4 drinks per week) (Meyer-Leu et al. 2011) and infant mortality (an average of 4+ drinks per week).

A number of other epidemiological studies examining the relationship between the pattern of maternal drinking and fetal effects have been published between 2009 and 2011, after the reviews documented in Table 4.4 (O'Leary and Bower 2011). The studies are from a range of
countries including Australia, the USA, the UK, Ireland and one from Denmark. As with the evidence from the systematic reviews, there is no strong evidence to support an increased risk to the fetus from low levels of prenatal alcohol exposure (O'Leary and Bower 2011). Further
research and new techniques investigating the effect of low levels of prenatal alcohol exposure are required.

Authors Number of studies
Outcomes Reference Group Alcohol Categories Results
Abel and Hannigan
Not specifiedLow birth weight and/or
AbstainersAverage drinks per day; 6
categories: 0, >0 and <0.5, 0.5-
1, 1.5-2 , 2-3
Evidence of a threshold relationship with a decrease in birth weight following an
average of 2 or more drinks per day
Polygenis et al.
7Malformations2 or less drinks per weekMore than 2 drinks per week
to 2 drinks per day
(24-168 g per week)
Makarechian et al.
8Spontaneous abortion, stillbirth,
premature birth
2 or less drinks per weekMore than2 drinks per week
to 2 drinks per day
(20-140 g per week)
↑ Spontaneous Abortion§
↓ Stillbirth§*
N/S Premature Birth
Testa et al. (2003)9Infant Development
AbstainersLess than1 drink per day **
1-1.99 drinks per day
2+ drinks per day
USA studies
N/S ↑ less than1drink per day
European studies
Protective less than 1drink per day
N/S less than 1drink per day
Gray and
Henderson (2006);
Henderson et al.
Number of studies
included varied by
Miscarriage (8 studies), Stillbirth
(5 studies),
Growth (7 studies), Birthweight
(19 studies),
Preterm Birth (16 studies),
Malformations (6 studies)
Neurodevelopment (7 studies)
Abstainer OR infrequent
(Less than 6g per week)
Up to 12g per day
( Less than 84g per week)
Miscarriage 5 of 8 studies ↑ §
Stillbirth N/S
Others each had only one study reporting
significant ↑
Some findings indicated a protective effect
Swedish National
Institute of Public
(Holmgren 2009)
6Cognitive and socio-emotional
Abstainer or infrequent
Low to moderate -1-4 glasses
per week (12-48 g); <1 glass
per week; binge and max per
Moderate prenatal exposure (approx <1/2
drink per day) reduced attention. <1 glass
per week associated with high total
strengths and difficulties scores in girls but
not higher doses.
Dolan et al. (2010)14Attention – cognitive performance
Abstainers or infrequent
drinkers (< 3 drinks per
Average ounces absolute
alcohol; binge drinking (5 or
more per occasion); average
number of drinking occasions
per month; average number of
drinking occasions per
occasion; exposed
dysmorphic; exposed nondysmorphic
No specific component of CPT
consistently associated with prenatal
alcohol exposure. Trends indicate more
commission and omission errors in
exposed than comparison group.
Latino-Martel et al.
19 total, 9 with dose
ranging from 0.5drinks
per week to >7 drinks
per week
Leukaemia- acute lymphoblastic
leukaemia (ALL); acute myeloid
leukaemia (AML)
Abstainers0.5 drinks per week to >0.7
dinking drinks per week
No increased odds of ALL. Increased odds
of AML in: i) analysis of binary exposure,
ii) analyses restricted to AML diagnosed
0-4 years iii) with increased odds with an
increase of one drink per week.
Patra et al. (2011) (Review article published in addition to those in O’Leary and Bower (2011))36 total, 24 with dose –
response information.
Number of studies
included varied by
Low birthweight (28 studies)
Preterm birth (21 studies)
Small for gestational age (11
AbstainersAverage number of alcoholic
drinks per day. One drink was
estimated to be 12 g pure
alcohol unless otherwise
No effect for low birthweight and SGA up
to an average of 10/g per day (approx 1
No effect for preterm birth up to an
average of 18g/day (approx 1.5 drinks),
thereafter increasing risk.
Odendaal et al.
(2009) (Review article published in addition to those in O’Leary and Bower (2011))
13 total, 3 with
combined effect of
alcohol and tobacco
Preterm birth (4 studies alcohol
alone, 1 combined effect of
alcohol and tobacco)
Low birthweight (2 combined
effect of alcohol and tobacco
AbstainersBinary: drank alcohol or notIncreased odds of preterm labour among
women who smoked and drank. This was
more than the sum of the effects of either
smoking or drinking.

a adapted from O'Leary, C. and C. Bower (2011). "Guidelines for pregnancy: What's an acceptable risk, and how is the evidence (finally) shaping up?" Drug and Alcohol Review with additional reviews by Patra et al 2011 and Odendaal 2011 included.
§ Significant results
*The authors suggest that while the significant decrease may either be due to a true beneficial effect of alcohol or it may be a result of the ‘healthy drinker effect’ in which women with a poor obstetric history are more likely to abstain from drinking alcohol
**Quantity of alcohol for a standard drink is not mentioned
N/S non significant.

prev pageTOC |next page