Most Australian women continue to drink during pregnancy despite a national guideline that recommends abstinence. Measures of previous alcohol use were the strongest predictors of compliance. Weekly or binge drinking and previously drinking more than recommended predicted non-compliance with guidelines during pregnancy. Women’s previous compliance with alcohol guidelines, regardless of pregnancy or breastfeeding status at that time, meant they were three and a half times more likely to comply during pregnancy. Contrary to previous research which found pre-pregnancy drinks per drinking day to be a strong predictor of consumption during pregnancy
, this study found the predictive value of quantity of alcohol consumed on a drinking day prior to pregnancy was only applicable when comparing women who drank versus abstainers. An increased quantity of alcohol per drinking day among those who did drink was not itself predictive of guideline compliance in pregnancy. Frequency of pre-pregnancy alcohol use, however, was strongly predictive of such compliance. This supports previous research which found that the frequency, rather than the quantity, of pre-pregnancy alcohol consumption is more useful in predicting alcohol use during pregnancy
[16, 25]. These findings may help to simplify the assessment of women of childbearing age who may be at risk of consuming alcohol if they become pregnant by focusing on how often they drink, rather than how much they usually drink.
By using prospective data before and during pregnancy, this population-based study provided a broadly representative prevalence of pregnant women’s compliance with alcohol guidelines. This is one of the first studies to assess whether the abstinence recommendation in the 2009 guidelines has been adopted by pregnant women. It is reasonable to assume that there may be some bias in this study’s estimates as only women with a recognised pregnancy were included. Considering a larger proportion of women drink during the pre-recognition phase of pregnancy
[18, 20, 26, 27], it is likely that this exclusion criteria may have led to an overestimation of compliance. In contrast to the 72% of women reporting drinking during pregnancy in this study, a report based on the 2010 National Drug Strategy Household Survey (NDSHS) found only 28% of Australian women over 31 reported drinking after pregnancy recognition, while 57% drank during some stage of pregnancy
. It is possible that a proportion of the 72% of non-compliant women in our study were consuming alcohol due to a lack of awareness of the revised alcohol recommendations due to the timing of the survey. However, discrepancy between the current study and the findings from the NDSHS may be partially attributed to a difference in measurement techniques. The ALSWH obtained information at the time of pregnancy, whereas NDSHS used a retrospective recall of the drinking behaviour that occurred in pregnancies within the past 12 months
. The ALSWH utilised a larger sample of pregnant women (N = 837) in a more defined age group (30–36 years) compared with the sample of women in the NDSHS (n = 434) that were relatively comparable in age (31 years or over).
Prior research found that 80% of Australian women were compliant with the 2001 alcohol guidelines which condoned low alcohol intake
, yet this study only found a 28% compliance rate with current guidelines. Given the majority (82%) of drinkers drank at low levels, a higher proportion of this study’s sample would have been classified as compliant with the 2001 alcohol guidelines. Similarly in the UK, where pregnant women are told to avoid alcohol in the first trimester and then limit alcohol to one to two drinks once or twice a week
, only 29% of women in their first trimester complied with the recommendations of early abstinence, whereas 94% of women in later pregnancy adhered to the low alcohol intake recommendation
. It appears that in Australia and the UK pregnant women are far less likely to comply with recommendations for no alcohol intake. In contrast, the US and Canada have maintained strong consistent messages of alcohol abstinence for pregnant women and have found that about 89% and 86% of pregnant women, respectively, complied with alcohol guidelines
[29, 30]. The high proportion of Australian women that continue drinking during pregnancy suggests that there has not been a large scale uptake of the evidence-based recommendation to abstain from alcohol. Previous research supports the notion that guidelines do not necessarily impact drinking behaviour
[10, 12], emphasizing that the creation of guidelines alone is not sufficient in altering population behaviour.
This study confirmed findings that previous alcohol consumption is one of the best predictors of prenatal use of alcohol
[10, 13]. Similarly, a recent Swedish study found that higher pre-pregnancy scores on the Alcohol Use Disorders Identification Test (AUDIT) were predictive of alcohol use during pregnancy
. In addition to the usual forms of alcohol assessment found in the literature (i.e. frequency
[15, 17, 18], and binge status
[14, 19]) this study has taken a novel approach by examining previous compliance to alcohol guidelines. By doing so, the current study was able to show a pattern of non-compliant behaviour.
This study is limited by the age range (30–36 years) of participants. Considering the mean age of Australian mothers is 30 years and there is a national trend of an increase in the age of mothers
, the results are likely to be generalisable to a large proportion of pregnant Australian women. There were missing data in some analyses; however, analyses of bias yielded no significant difference in the outcome of interest due to missing or excluded cases. Self-report may have led to response bias in the under-reporting of alcohol use. However, self-report has been found to be more accurate than physicians’ medical records in identifying prenatal alcohol use
. Furthermore, the confidential nature in using a unique identifying code, as was done in this study, has been found to be equally effective in obtaining a high rate of self-reported alcohol use by pregnant women compared with using a purely anonymous technique
This study was within the confines of a large longitudinal study which led to one of the major limitations. There was a relatively short timeframe between when the 2009 guidelines were introduced and when the surveys were sent out. However, draft guidelines were available and widely publicised as early as 2007. Previous research conducted in late 2008 to early 2009 has shown that health professionals were passing on an abstinence message to pregnant clients, consistent with the 2009 guidelines
. Additionally, participants on average took about three months to return their surveys, with some taking up to 14 months. Seeing as how women were asked about their alcohol use when they were pregnant, rather than asking them to recall their entire pregnancy, it is believed that this study has gathered an accurate measure of drinking during pregnancy at the time the surveys were completed, which occurred under the 2009 guidelines. Whether the guidelines were properly disseminated is a topic for further research but does not limit the fact that the 2009 guidelines were in place when the women were surveyed about their behaviour.
Alcohol behaviours should be assessed before women become pregnant because pre-pregnancy alcohol use and previous compliance with guidelines predict whether Australian women will comply with guidelines during pregnancy. General practitioners (GPs) are ideally suited to assess alcohol intake in women of childbearing age. GPs are the gatekeepers to the Australian healthcare system; 19% of their clients are women of childbearing age (15–45 years) and average consultation times range from 14–15 minutes
. Best practice clinical guidelines suggest that pregnant women, or those who may become pregnant, should be provided with information about potential consequences of prenatal alcohol use in order to make an informed decision
[8, 37]. However, a random sample of Australian health professionals found that only a quarter of providers routinely provided such information
. Awareness and familiarity of, and attitudes towards clinical guidelines have been found to affect health professionals’ adherence to them
It may be necessary for policy makers to implement strategies to effectively disseminate the alcohol guidelines for pregnant women to ensure they are both implemented by the healthcare system and adopted by the general population. Such strategies may include the use of local opinion leaders to address barriers and encourage best practice among health professionals
. Additionally, mass media campaigns could be developed as they have been found to be effective in other public health initiatives such as reducing alcohol-related crashes
 and increasing initiation of and positive attitudes towards breastfeeding
. US authorities have suggested that in addition to mass media campaigns other universal prevention strategies, such as policy-driven warning labels on alcoholic beverages and other strategies to reduce overall consumption for the population, may be useful in helping to prevent alcohol-exposed pregnancies
. Studies from Scandinavian countries have reported that mass media is the number one information source regarding alcohol use in pregnancy for pregnant women
[17, 44]. It has also been found that pregnant women believe a health professional could best communicate this information
 and women are comfortable discussing alcohol use with healthcare providers
. Currently, no mass media campaign or other universal prevention strategies exist in Australia to promote the most recent alcohol guidelines for pregnant women, stressing not only a need for public health promotion but also the importance of healthcare professionals in disseminating this public health message.
Based on the results of this study, GPs may find it useful to initiate a conversation about alcohol use by asking women about their usual alcohol consumption (e.g. when not pregnant) as a lead in to assessing their current alcohol use. If women report usually drinking more than the recommended guidelines or usually drink on a weekly basis, then the GP can use that context to provide them with information about the potential consequences of alcohol use during pregnancy and the national recommendation for abstinence. For women of childbearing age, healthcare providers could offer brief motivational interviewing which has been found to reduce the risk of alcohol exposed pregnancies
. GPs may consider using educational and psychological interventions for their pregnant clients, which have been found to assist pregnant women in abstaining from alcohol