SEARCH

SEARCH BY CITATION

Keywords:

  • Alcohol;
  • low birthweight;
  • meta-analysis;
  • neonatal development;
  • preterm birth;
  • small for gestational age

Abstract

  1. Top of page
  2. Abstract
  3. Introduction
  4. Methods
  5. Results
  6. Discussion
  7. Conclusion and future research
  8. Disclosure of interests
  9. Contribution to authorship
  10. Details of ethics approval
  11. Funding
  12. Acknowledgements
  13. References
  14. Supporting Information

Please cite this paper as: Patra J, Bakker R, Irving H, Jaddoe V, Malini S, Rehm J. Dose–response relationship between alcohol consumption before and during pregnancy and the risks of low birthweight, preterm birth and small for gestational age (SGA)—a systematic review and meta-analyses. BJOG2011;118:1411–1421.

Background  Descriptions of the effects of moderate alcohol consumption during pregnancy on adverse pregnancy outcomes have been inconsistent.

Objective  To review systematically and perform meta-analyses on the effect of maternal alcohol exposure on the risk of low birthweight, preterm birth and small for gestational age (SGA).

Search strategy  Using Medical Subject Headings, a literature search of MEDLINE, EMBASE, CINAHL, CABS, WHOlist, SIGLE, ETOH, and Web of Science between 1 January 1980 and 1 August 2009 was performed followed by manual searches.

Selection criteria  Case–control or cohort studies were assessed for quality (STROBE), 36 available studies were included.

Data collection and analysis  Two reviewers independently extracted the information on low birthweight, preterm birth and SGA using a standardised protocol. Meta-analyses on dose–response relationships were performed using linear as well as first-order and second-order fractional polynomial regressions to estimate best fitting curves to the data.

Main results  Compared with abstainers, the overall dose–response relationships for low birthweight and SGA showed no effect up to 10 g pure alcohol/day (an average of about 1 drink/day) and preterm birth showed no effect up to 18 g pure alcohol/day (an average of 1.5 drinks/day); thereafter, the relationship showed a monotonically increasing risk for increasing maternal alcohol consumption. Moderate consumption during pre-pregnancy was associated with reduced risks for all outcomes.

Conclusions  Dose–response relationship indicates that heavy alcohol consumption during pregnancy increases the risks of all three outcomes whereas light to moderate alcohol consumption shows no effect. Preventive measures during antenatal consultations should be initiated.


Introduction

  1. Top of page
  2. Abstract
  3. Introduction
  4. Methods
  5. Results
  6. Discussion
  7. Conclusion and future research
  8. Disclosure of interests
  9. Contribution to authorship
  10. Details of ethics approval
  11. Funding
  12. Acknowledgements
  13. References
  14. Supporting Information

Many observational studies have been published on the topic of alcohol consumption in pregnant women and its effects on the development of fetus and child. The association of heavy maternal alcohol consumption during pregnancy and various adverse birth outcomes has been well established.1,2 Also, excessive alcohol consumption during pregnancy is associated with adverse postnatal behavioural development.3 However, studies focused on the associations of low to moderate alcohol consumption during pregnancy with birth outcomes showed inconsistent results.4–9 In general, low to moderate maternal alcohol consumption is considered as an average of one alcoholic drink at most per day. Some studies did not find any associations, whereas others found adverse or even beneficial effects. A recent systematic review by Henderson et al.10 also reported no convincing evidence for adverse effects of low to moderate maternal alcohol consumption on pregnancy outcomes, such as miscarriage, stillbirth, fetal growth restriction, prematurity, low birthweight, small for gestational age (SGA) at birth and birth defects including fetal alcohol syndrome. The authors were not able to perform a meta-analysis because of the high heterogeneity in the methods of the various studies used in their systematic review. They suggested that differences in results between studies might be the result of differences in study design and in timing and methods of assessment of maternal alcohol consumption. Also, differences in adjustment for possible confounding factors between the studies may explain inconsistent results. The aim of this systematic review and meta-analysis was to assess the dose–response association of maternal alcohol exposure before and during pregnancy with the risks of low birthweight, preterm birth and SGA.

Methods

  1. Top of page
  2. Abstract
  3. Introduction
  4. Methods
  5. Results
  6. Discussion
  7. Conclusion and future research
  8. Disclosure of interests
  9. Contribution to authorship
  10. Details of ethics approval
  11. Funding
  12. Acknowledgements
  13. References
  14. Supporting Information

Search strategy

We conducted a systematic literature search for potentially relevant original papers using the following electronic databases from January 1980 to the first week of June 2009: MEDLINE, EMBASE, CINAHL, CABS, WHOlist, SIGLE, ETOH and Web of Science. We used the following keywords and medical subject headings to identify relevant articles in electronic databases: (‘alcohol*’ or ‘ethanol’ or ‘light drinking’ or ‘moderate drinking’) AND (‘birthweight’ or ‘low birthweight’ or ‘gestational age’ or ‘small for gestational age’ or ‘preterm*’ or ‘pregnancy outcome’ or ‘pregnancy complication’ or ‘prenatal*’) AND (‘case’ or ‘cohort’ or ‘ratio’ or ‘risk*’ or ‘prospective*’ or ‘follow*’). No language restrictions were applied. Eligible studies were original publications (we excluded letters, editorials, conference abstracts, reviews and comments) of case–control and cohort studies reporting incidence, hazard ratios, relative risks or odds ratios of alcohol consumption in comparison to abstainers. In addition, bibliographies of key retrieved articles, relevant reviews and meta-analyses were hand searched.

The strategy resulted in 1345 hits; of which 90 appeared relevant upon initial inspection. The contents of these abstracts or full-text manuscripts identified during the literature search were reviewed independently by two reviewers to determine whether they met the criteria for inclusion. Articles were considered for inclusion in the systematic review if they reported data from an original study (i.e. no review articles). When there were discrepancies between investigators for inclusion or exclusion, a third reviewer (JR) conducted additional evaluation of the study and discrepancies were resolved in consultation. To be included in our meta-analysis, a published study had to meet the following criteria:

  • 1
     Reported data were from an original study (i.e. no review articles)
  • 2
     Cohort or case–control study in which medically confirmed low birthweight (defined as <2500 g), preterm birth (<37 weeks of gestation) and SGA (<10th percentile of gestational age-adjusted birthweights) were the end points
  • 3
     Reporting of relative risk or odds ratios or hazard ratios (or data to calculate these risks) of low birthweight, preterm birth and SGA associated with alcohol consumption.

Thirty-six studies met all of the inclusion criteria and were included in the meta-analysis. Twenty-four had dose–response information with at least three or more drinking exposure groups and 12 studies had exclusive data on drinker versus no drinker. Four previous systematic reviews10–13 and three meta-analyses14–16 were identified and excluded. Details on study exclusion are given in Figure 1.

image

Figure 1.  Results of systematic review of the relationship between maternal alcohol consumption and low birthweight, preterm birth and small for gestational age (SGA).

Download figure to PowerPoint

Data extraction

All data were independently extracted by means of a standardised protocol. Study characteristics recorded were as follows: title of the study, lead author surname, publication year, source of publication, country of origin, study design (cohort or case–control), characteristics of the study population (e.g. size of the sample; method of sampling; age distribution, average age, and ethnicity), measures of outcome and exposure, duration of follow-up (for prospective cohort studies), confounding factors controlled for by matching or adjustment, and the risk estimates (relative risk or odds ratios or hazard ratios) of birth outcomes studied, compared with abstainers, associated with alcohol consumption and the corresponding confidence intervals. When a range of alcohol intake was given, the midpoint of the range was taken. In cases where open-end for the highest category was given (e.g. 40 + g/day), three-quarters of the length of the immediate previous category range was added to the lower bound and was used as the measure. Where consumption was reported in drinks and not in grams, the gram pure alcohol equivalent (of one drink) explained in the article was used as a conversion factor if stated, and if not, conversion was based on geographical location: for Canada 13.6 g, USA 12 g, UK 8 g and for both New Zealand and Australia 10 g pure alcohol. For all other countries without any clear specifications 12 g pure alcohol was used as an equivalent of one drink.

Information about the level of exposures in each study, the number of cases at each exposure level, the total population at risk at each exposure level, the adjusted estimates of relative risk (RR) compared with abstention for each exposure level, and the corresponding lower and upper 95% confidence intervals (95% CI) of the adjusted RR were obtained.

To ensure accuracy in data abstraction, five included and five excluded studies were randomly chosen to be abstracted independently by a co-author (HI) and the results were compared. Both authors agreed on five of five articles reviewed for inclusion/exclusion, and on 611 of 654 data points abstracted over ten articles. Where disagreements existed, both authors reviewed the materials together until a consensus was reached.

Drinkers versus nondrinkers meta-analysis

In the drinkers versus nondrinkers meta-analysis, the DerSimonian and Laird17 random-effects method was used to combine the natural logarithm of the risk estimates across studies. Where a study provided a dose–response analysis only, the risk estimates for all drinking categories were pooled using the inverse variance weighted method to derive a single estimate. These statistical analyses were completed using the METAN command in stata version 10.1 (StataCorp, College Station, TX, USA).18

Meta-regression of dose–response relationship

Based on previously published research, the associations between maternal alcohol consumption with low birthweight, preterm birth and SGA could be either linear or nonlinear. To be flexible in fitting the best model, we conducted the meta-regression using linear as well as first-order and second-order fractional polynomial regression with powers −2, −1, −0.5, 0, 0.5, 1, 2, 3 to estimate a best-fitting curve to the data. Best-fit curves were assessed using decreased deviance compared with the reference model. Comparisons of curves to determine the best fit were made using a chi-square distribution.19 The first-order and second-order fractional polynomials take the general form shown in equations 1 and 2, respectively:

  • image(1)
  • image(2)

where x is the alcohol exposure level in grams per day, P1 and P2 are the polynomial powers and β1 and β2 are the corresponding coefficients. No intercept term exists because all models have a starting point of Log RR = 0 (RR = 1 at zero consumption). All models were fitted in stata version 10.1, using the Generalized Least Square for Trend estimation (GLST) function.18

Heterogeneity and publication bias

Statistical heterogeneity between studies was assessed using both the Cochrane Q test and the I2 statistic.20 Because all statistical tests for heterogeneity are weak, we also included the 95% CI for I2,21,22 which was calculated based on the method described by Higgins and Thompson.20 Publication bias was assessed by visual inspection of Begg’s funnel plot, the Begg–Mazumdar adjusted rank correlation test23 and the Egger regression asymmetry test for funnel plot.24 The RR estimates were prepooled using the inverse variance weighted method because funnel plot methodology assumes one overall RR per article. Statistically significant publication bias was defined as P < 0.10.

Results

  1. Top of page
  2. Abstract
  3. Introduction
  4. Methods
  5. Results
  6. Discussion
  7. Conclusion and future research
  8. Disclosure of interests
  9. Contribution to authorship
  10. Details of ethics approval
  11. Funding
  12. Acknowledgements
  13. References
  14. Supporting Information

Characteristics of the included studies

We identified 36 observational studies that had met the inclusion criteria as outlined in Figure 1. Twenty-four of the 36 studies had dose–response information (with at least three or more drinking exposure groups) and were the basis of meta-regression analysis. However, all 36 studies were used in a separate meta-analysis of maternal drinking versus maternal nondrinking. For meta-regression analysis, out of 19 studies on low birthweight 15 were cohort studies8,25–38 and four were case–control studies.39–42 Collectively, the 19 studies provided 28 data sets for a total of 277 300 pregnant mothers with 20 582 cases of low birthweight. Similarly, on 14 studies on preterm birth, 12 cohort studies8,25,26,28,29,31,33,34,43–46 and two case–control studies39,47 had 26 data sets and a total of 280 443 pregnant mothers with 12 888 preterm births. Likewise, eight studies provided 17 data sets for a total of 136 949 pregnant mothers with 8679 SGA infants. Six studies25,29–31,37,46 were case–control and the rest were cohort.40,42 For drinker versus nondrinker meta-analysis, in addition to the 24 studies already mentioned, 12 more studies9,48–58 were added. Only one58 of these was a case–control study.

Adjustment for confounders varied between studies. All but seven studies on low birthweight and nine out of 14 studies on preterm birth and all studies on SGA adjusted for confounders (such as smoking, socio-economic status, body mass index). Ascertainment on these birth outcomes was determined through written self-report, interview after birth, outcomes from clinical/medical records, paediatrician examination and hospital delivery record. Tables S1 and S2 summarise the characteristics of the included studies.

Overall, marked heterogeneity was found for all birth outcomes (low birthweight (Q = 122.5, P = 0.006; I2 = 80%, 95% CI 73–85%, P < 0.001); preterm birth (Q = 98.03, P < 0.072; I2 = 89%, 95% CI 84–92%, P < 0.001); SGA (Q = 131.20, P < 0.001; I2 = 92%, 95% CI 88–95%, P < 0.001). Random effects models were used for all subsequent analyses. No significant publication bias was detected.

Drinkers versus nondrinkers meta-analysis

The summary of 28 studies related to low birthweight indicated an overall pooled RR of 1.12 (95% CI 1.04–1.20) among mothers drinking before or during pregnancy. When this analysis was restricted to studies with confounders, the adjusted RR was slightly affected and not significant (RR 1.06, 95% CI 0.99–1.13) (Figure 2). Similarly, the pooled odds ratio of preterm birth between 21 studies was 1.03 (95% CI 0.91–1.16). This effect estimate attenuated (0.93, 95% CI 0.86–1.01) when the analysis was restricted to studies which adjusted for confounders (Figure 3). The pooled odds ratio of SGA among 11 studies was 1.11 (95% CI 0.95–1.30) (Figure 4). The effect estimate on studies that adjusted for confounders resulted in almost no effect (0.99, 95% CI 0.89–1.10).

image

Figure 2.  Relative risks for low birthweight comparing alcohol consumption with no alcohol consumption (28 studies). *The unadjusted pooled relative risk does not include unadjusted estimates from adjusted studies. Unadjusted I2 = 89%, 95% CI 84–93%, P < 0.001; Adjusted I2 = 62%, 95% CI 41–75%, P < 0.001; Overall I2 = 80%, 95% CI 73–85%, P < 0.001.

Download figure to PowerPoint

image

Figure 3.  Relative risks for preterm birth comparing alcohol consumption with no alcohol consumption (21 studies). *The unadjusted pooled relative risk does not include unadjusted estimates from adjusted studies. Unadjusted I2 = 91%, 95% CI 86–94%, P < 0.001; Adjusted I2 = 64%, 95% CI 31–81%, P < 0.001; Overall I2 = 89%, 95% CI 84–92%, P < 0.001.

Download figure to PowerPoint

image

Figure 4.  Relative risks for SGA comparing alcohol consumption with no alcohol consumption (11 studies). *The unadjusted pooled relative risk does not include unadjusted estimates from adjusted studies. Unadjusted I2 = 93%, 95% CI 83–97%, P < 0.0001; Adjusted I2 = 82%, 95% CI 65–91%, P < 0.0001; Overall I2 = 92%, 95% CI 88–95%, P < 0.0001.

Download figure to PowerPoint

Dose–response meta-analyses

A total of 44 first-degree fractional polynomial models were examined (eight first-order models and 36 second-order fractional polynomials) for both low birthweight and preterm birth. Overall among pregnant mothers for low birthweight, the best second-degree model with powers 0.5 and 1 (and function β1x5 + β2x) fitted significantly better than the linear and first-degree models (P < 0.001). For preterm birth and SGA, the best fitting model was the second-degree fractional polynomials with powers 0.5 and 0.5 (and function β1x.5 + β2x.5lnx) (P < 0.001).

Figures 5–7 show an overall dose–response relationship between alcohol consumption and risk of low birthweight, preterm birth and SGA, respectively. Compared with abstainers, the risk of low birthweight with alcohol consumption was not apparent until more than 10 g/day or an average of about one drink per day (based on US conversions) but linearly associated thereafter up to 120 g/day to a maximum of 7.48 (95% CI 4.46–12.55), indicating a steeper slope after 10 g/day. The risk becomes two-fold only after 52 g/day (equivalent to an average of four to five drinks per day) (also see Table S3). Relative to nondrinking mothers, alcohol consumption of less than 19 g/day, or an average of about 1.5 drinks per day, was not associated with a risk of preterm birth. However, at an average of three drinks (36 g/day), the risk of having a preterm birth is 23% more likely than in nondrinking mothers (RR 1.23, 95% CI 1.05–1.44) (Table S4). Similarly, compared with abstinent mothers, maternal drinking up to 10 g/day was not associated with the risk of SGA. With an average of three or more drinks a day, the risk of having a SGA infant increases (see Table S5).

image

Figure 5.  Meta-analysis 19 studies* showing the dose–response relationship between maternal alcohol consumption and low birthweight. *For the information supporting this figure, please see the details in Table S3.

Download figure to PowerPoint

image

Figure 6.  Meta-analysis of 14 studies* showing the dose–response relationship between maternal alcohol consumption and preterm birth. *For the information supporting this figure, please see the details in Table S4.

Download figure to PowerPoint

image

Figure 7.  Meta-analysis of eight studies* showing the dose–response relationship between maternal alcohol consumption and SGA. *For the information supporting this figure, please see the details in Table S5.

Download figure to PowerPoint

In a sensitivity analysis, we looked into the type of studies, i.e. case–control versus cohort. As a result, we repeated the analyses separately for case–control and cohort studies. We observed that study type did affect the risk relation with increasing volume of alcohol exposure with preterm birth. But for the other two pregnancy outcomes, it did not have much affect. Although there were similar risk patterns up to an average of one drink per day, a linearly increasing dose–response relationships existed, thereafter, among case–control studies15 and a model very similar to the main analysis, in the cohort studies (see Tables S3–S5).

The second sensitivity analysis compared risks of both pregnancy outcomes on prepregnancy (i.e. until pregnancy is known) and during pregnancy. Consumption during prepregnancy was associated without a risk of low birthweight, preterm birth and SGA up to 30 g/day (an average of 2.5 drinks/day), 50 g/day (an average of about four drinks/day) and 18 g/day (an average of 1.5 drinks/day), respectively. On the other hand, compared with the main analysis, risk estimates changed a little among studies during pregnancy or consumption during different trimesters of the pregnancy period (see Tables S3–S5).

The final sensitivity analysis, performed using studies that adjusted for confounders (at least smoking as one of the confounders), resulted in models remarkably similar to those in the main analysis (see Tables S3–S5).

Discussion

  1. Top of page
  2. Abstract
  3. Introduction
  4. Methods
  5. Results
  6. Discussion
  7. Conclusion and future research
  8. Disclosure of interests
  9. Contribution to authorship
  10. Details of ethics approval
  11. Funding
  12. Acknowledgements
  13. References
  14. Supporting Information

This systematic review and meta-analysis indicates a non-linear association between maternal alcohol consumption and the risks of low birthweight, preterm birth and SGA. The risk of low birthweight and SGA with alcohol consumption increased linearly in mothers who consumed an average of one drink or more per day. Similarly, in mothers who consumed more than three alcoholic drinks per day, the risk of having a preterm birth was increased by 23%.

Methodological considerations

In total, 36 observational studies were identified in this meta-analysis. We analysed a large data set with, depending on the outcome measure, 277 300 or 280 443 pregnant mothers with 20 582 children with low birthweight, 12 888 preterm birth and 8679 SGA infants.

Most studies in this meta-analysis adjusted their multiple regression analyses for possible confounders, i.e. smoking, socio-economic status, body mass index and ethnicity. After adjusting for these confounding factors there is presumably still residual confounding, because of inaccuracy in measuring these confounders or not adjusting for other important, possible unmeasured, confounders. These unmeasured confounders may be mainly lifestyle-related and socio-economic-related factors. The potential for residual confounding is also reflected by the larger effect estimates for the unadjusted estimated models than the adjusted models. The majority of studies in this meta-analysis had data at low to moderate alcohol consumption levels compared with heavy consumption levels, making the dose–response curves of maternal alcohol consumption and low birthweight, preterm birth and SGA (Figures 5–7) more stable at the low to moderate levels and more variable at the higher consumption levels. Also, we observed marked heterogeneity of the identified studies on all outcomes. This could be because of methodological or actual differences between the studies. To address this heterogeneity, we used random effects models for the pooled effect estimates analyses. The use of these random effects models explains why we are able to perform this meta-analysis compared with the systematic review of Henderson et al.10 Also, because of a different inclusion period we were able to add four large studies25,31,40,41 performed worldwide on this topic, and finally, we performed a dose–response meta-regression as well as a drinker versus nondrinker meta-analysis. Another important limitation of meta-analyses is the presence of publication bias. Such bias occurs when research that appears in the published literature is systematically not representative of the population of completed studies.59 We did not, however, observe any publication bias in our analyses.

Different definitions of birth outcomes did not occur. All identified studies used the common criteria for defining low birthweight (<2500 g), preterm birth (<37 weeks of gestation) and SGA (below the 10th percentile). Low birthweight analyses in the included studies were adjusted for gestational age at birth. Additionally, different collection methods, of alcohol consumption levels and timing of consumption, used in the included studies may also have influenced our results. Some studies used averaged alcohol consumption habits because of repeated assessment, whereas others only collected this information once during pregnancy, or even postnatally. This latter method may have introduced recall bias. Women with adverse pregnancy outcomes may under-report their actual alcohol consumption,60 whereas women with good pregnancy outcomes may not under-report. The high levels of alcohol consumption reported retrospectively could be more similar in magnitude to the levels reported for the postpartum period, and hence influence their retrospective recall.61 For example, when interviewed retrospectively about alcohol consumption on a typical week, many mothers do not either recognise or probably forget the exact consumption and end up reporting their postpartum intake. This can cause overestimation of the effect estimates. Besides, misclassification error of the alcohol exposure may have occurred because of the use of self-reporting questionnaires or postpartum interviews.62,63 If under-reporting was present in all categories of alcohol consumption, the effect estimates would have been underestimated. However, if mothers with heavy alcohol consumption selectively under-reported their average number of drinks, the differences between no alcohol consumption and the lower categories of alcohol consumption would have been overestimated. This misclassification could be averted by using more objective measures of alcohol consumption levels. Unfortunately, the use of current biomarkers to assess alcohol consumption levels, including carbohydrate-deficient transferrin and γ-glutamyltransferase, seem to be inappropriate for the assessment of light to moderate alcohol consumption levels.64 But for heavy maternal drinking (average 42.5 grams absolute alcohol ingested daily), fatty acid ethyl esters extracted from meconium are found to be a reliable biomarker.65,66 Using meconium, a Canadian research group67 showed that they could objectively detect babies exposed to excessive maternal drinking of alcohol in pregnancy. Finally, a potential limitation in the studies is the use of the average number of alcoholic drinks per day or per week without taking into account the patterns of alcohol consumption. The explanation of our findings may lie in patterns of drinking (see O’Leary et al.31, for an exception). Several previous studies reported harmful effects of more concentrated drinking patterns or binge drinking on fetal and postnatal development.68,69 One recent study70 suggests that ignoring the pattern and frequency may in some circumstances completely mask the association (e.g. language delay, behavioural problems at early age). To deal with this, their study has proposed a new method of classification that reflects real-life drinking patterns.

Maternal alcohol consumption and low birthweight, preterm birth and SGA

This analysis adds weight to previous findings that light to moderate alcohol consumption during pregnancy does not increase the risks of low birthweight, preterm birth and SGA. The results are similar to findings in a recently published systematic review on low to moderate prenatal alcohol exposure and pregnancy outcomes.10 Henderson et al.10 suggested that small amounts of alcohol appeared to have a small protective effect on birthweight, and found either no effect or a reduction in risk of prematurity with the consumption of up to 72 g alcohol per week. These results should be interpreted with caution, because no information on drinking patterns was taken into account. Consumption of small amounts of alcohol concentrated within a few days may be harmful. The authors provided a possible explanation for this finding with the ‘healthy-drinkers effect’, in which women with poorer obstetric history or prognosis are more prone to abstain from alcohol consumption during pregnancy.

This meta-analysis also shows that for alcohol consumption levels a cut-off value of the average number of alcoholic drinks may exist at which alcohol consumption may lead to adverse effects on birth outcomes. According to our study this cut-off lies between 1 and 1.5 averaged alcoholic drinks per day, which means approximately 10–18 g of alcohol per day. Previous studies also suggested effects of maternal alcohol consumption on postnatal growth and development. It was shown that the rate of postnatal growth is reduced in children who were prenatally exposed to alcohol.68,69 Postnatal weight, length and head circumference were negatively affected at least from 14 to 21 years of age because of alcohol exposure during pregnancy.71–74 However, a more recent longitudinal study showed that moderate maternal alcohol consumption during pregnancy was not associated with either weight or head circumference at the age of 5 years.75 Also, inconsistent results were found on behavioural development and cognitive processing in children prenatally exposed to alcohol.76,77 A study by Faden and Graubard78 showed a higher activity level, a greater difficulty in following instructions and eating problems among offspring exposed to alcohol during pregnancy. Furthermore, binge drinking during pregnancy was shown to be associated with increased odds for the appearance of psychiatric disorders.3 Previous studies did not show consistent associations of binge drinking during pregnancy with several outcomes, except for neurodevelopmental outcomes.2 Whether light to moderate alcohol consumption is related to postnatal growth and development still needs to be examined. The effects of alcohol consumption are dependent on the absorption and metabolism in the mother and the fetus. This may be partially genetically determined. Therefore, the effects of alcohol consumption in specific groups of women should still be studied.

Conclusion and future research

  1. Top of page
  2. Abstract
  3. Introduction
  4. Methods
  5. Results
  6. Discussion
  7. Conclusion and future research
  8. Disclosure of interests
  9. Contribution to authorship
  10. Details of ethics approval
  11. Funding
  12. Acknowledgements
  13. References
  14. Supporting Information

The results of this meta-analysis indicate that heavy alcohol consumption during pregnancy increases the risk of low birthweight and preterm birth whereas light alcohol consumption may not affect these neonatal outcomes. Preventive measures could be initiated and promotion of a healthy lifestyle could be optimised in antenatal care. In this way current awareness of the risks of certain lifestyle factors may increase and the adverse effects may subsequently decrease. Most important, the harmful effects of heavy alcohol consumption, in even the preconception period, should be acknowledged and emphasised. Future research should be focused on the associations of low to moderate alcohol consumption with postnatal growth and development before new public health strategies can be developed.

Contribution to authorship

  1. Top of page
  2. Abstract
  3. Introduction
  4. Methods
  5. Results
  6. Discussion
  7. Conclusion and future research
  8. Disclosure of interests
  9. Contribution to authorship
  10. Details of ethics approval
  11. Funding
  12. Acknowledgements
  13. References
  14. Supporting Information

JP and JR conceived the study, conducted the underlying systematic reviews and supervised all aspects of its implementation and led the writing. JP and HI also contributed to the methodology and quantitative analysis of the study. JP, RA, SM and VWVJ were involved with data interpretation, critical revisions of the paper and provided approval for its publication.

Funding

  1. Top of page
  2. Abstract
  3. Introduction
  4. Methods
  5. Results
  6. Discussion
  7. Conclusion and future research
  8. Disclosure of interests
  9. Contribution to authorship
  10. Details of ethics approval
  11. Funding
  12. Acknowledgements
  13. References
  14. Supporting Information

This work was financially supported by a small contribution from the Global Burden of Disease (GBD) Study to the last author. Also, we received support from NIAAA (Alcohol- and Drug-Attributable Burden of Disease and Injury in the US; contract # HHSN267200700041C). In addition, support to the Centre for Addiction and Mental Health (CAMH) for salaries of scientists and infrastructure has been provided by the Ontario Ministry of Health and Long Term Care. The views expressed here do not necessarily reflect those of the Ministry of Health and Long Term Care.

Acknowledgements

  1. Top of page
  2. Abstract
  3. Introduction
  4. Methods
  5. Results
  6. Discussion
  7. Conclusion and future research
  8. Disclosure of interests
  9. Contribution to authorship
  10. Details of ethics approval
  11. Funding
  12. Acknowledgements
  13. References
  14. Supporting Information

We would like to thank the core group of the Comparative Risk Assessment for alcohol within the GBD 2005 Study for their support and comments on the general methodology and an earlier version of this paper.

References

  1. Top of page
  2. Abstract
  3. Introduction
  4. Methods
  5. Results
  6. Discussion
  7. Conclusion and future research
  8. Disclosure of interests
  9. Contribution to authorship
  10. Details of ethics approval
  11. Funding
  12. Acknowledgements
  13. References
  14. Supporting Information

Supporting Information

  1. Top of page
  2. Abstract
  3. Introduction
  4. Methods
  5. Results
  6. Discussion
  7. Conclusion and future research
  8. Disclosure of interests
  9. Contribution to authorship
  10. Details of ethics approval
  11. Funding
  12. Acknowledgements
  13. References
  14. Supporting Information

Table S1. Characteristics of 24 observational studies with dose–response data on maternal alcohol consumption and risk of low birthweight, preterm birth and SGA.

Table S2. Characteristics of 12 observational studies included only in the drinker versus nondrinker meta-analysis.

Table S3. The associations of low birthweight with alcohol consumption according to different exclusion criteria (dose–response data on 19 studies).

Table S4. The associations of preterm birth with alcohol consumption according to different exclusion criteria (dose–response data on 14 studies).

Table S5. The associations of SGA with alcohol consumption according to different exclusion criteria (dose–response data on eight studies).

FilenameFormatSizeDescription
BJO_3050_sm_TableS1.pdf395KSupporting info item
BJO_3050_sm_TableS2.pdf27KSupporting info item
BJO_3050_sm_TableS3.pdf57KSupporting info item
BJO_3050_sm_TableS4.pdf76KSupporting info item
BJO_3050_sm_TableS5.pdf73KSupporting info item

Please note: Wiley Blackwell is not responsible for the content or functionality of any supporting information supplied by the authors. Any queries (other than missing content) should be directed to the corresponding author for the article.