Maternal alcohol-use disorder is associated with increased risk of sudden infant death syndrome and infant death from other causes
- Section of Social Medicine, Department of Public Health, University of Copenhagen, Copenhagen K, Denmark
- Correspondence to: Dr Katrine Strandberg-Larsen, Section of Social Medicine, Department of Public Health, University of Copenhagen, Øster Farimagsgade 5, Post Box 2099 Copenhagen K 1014, Denmark;
Implications for practice and research
Women with an alcohol-use disorder should be identified and offered supportive antenatal care services and treatment.
Alcohol-use disorder is associated with lifestyle and parenting styles that contribute to an increased risk of infant death; the prevention of and effective management of alcohol abuse has the potential of reducing infant deaths, particularly sudden infant death syndrome (SIDS).
Future research should include alcohol-related diagnosis in fathers and compare associations between maternal and paternal alcohol-related diagnosis and infant mortality in order to disentangle environmental factors from direct intrauterine effects of alcohol consumption on infant and child development.
Previous studies on prenatal exposure to alcohol and infant survival have reported conflicting results. This is in part because outcome definitions vary across studies, for example, deaths in the perinatal period, within the first week, month or year of life and including all causes of deaths or deaths limited to SIDS.1 ,2 Studies focusing on all causes of infant mortality have suggested that prenatal exposure to at least four to six alcoholic drinks a week is associated with increased risk.1 ,3 ,4 Findings from studies limited to SIDS, although less conclusive, suggest an association between heavy maternal alcohol use and SIDS.
O'Leary and colleagues undertook a register linkage study accessing multiple data sources in Western Australia from 1983 to 2005 to examine the association between alcohol-use disorder and infant mortality, not classified as SIDS and SIDS as a separate cause. Mortality was compared in two cohorts of children: children of mothers with an alcohol-related diagnosis (n=21 841) and a random sample of children of mothers without an alcohol-related diagnosis (n=56 054). An alcohol-related diagnosis was recorded from 10 years of age and onwards and was used as a proxy for alcohol-use disorder, indicating heavy alcohol consumption. The timing of the alcohol-related diagnosis in relation to pregnancy was categorised into five groups: during pregnancy, ≤1 year prepregnancy, ≤1 year postpregnancy, >1 year prepregnancy and >1 year postpregnancy.
There were 598 infant deaths and 303 cases of SIDS across both cohorts: 1.1% compared with 0.6% of infant deaths excluding SIDS, and 0.8% compared with 0.2% of deaths attributed to SIDS when comparing the alcohol-related diagnosis cohort with the comparison cohort, respectively. The elevated risk of infant death, not classified as SIDS, was highest and more than double when the mother was diagnosed during pregnancy. There was more than a three-fold risk of SIDS in the alcohol-related diagnosis cohort, which was eight times greater when mothers were diagnosed within 1 year postpartum. Around 4% of infant deaths not classified as SIDS and 25% of infant deaths classified as SIDS can be attributed to a history of a maternal alcohol-related diagnosis.
O'Leary and colleagues have highlighted that the increased risk of SIDS and other infant deaths is likely to be attributable to direct intrauterine effects of heavy alcohol exposure as well as environmental risk factors involving lifestyle, social circumstances or impaired parental capacity associated with alcohol abuse. An approach to disentangle the direct intrauterine and environmental effects would be to include alcohol-related diagnosis of fathers and compare maternal and paternal associations between alcohol-related diagnosis and infant mortality. The father, in many families, is an integral part of the child care environment. If the increased risk of infant mortality is caused by environmental risk factors, for example, bed sharing, that are more frequent among people with an alcohol-use disorder, similar associations would be expected. Having a record of an alcohol-related diagnosis in either the administrative health registers or drug/alcohol office data seems to be a proxy, with a high specificity for an alcohol-use disorder. However, as discussed by the authors, the sensitivity is likely to be low. It is a well-recognised challenge to identify pregnant women with heavy alcohol consumption. An alcohol-related diagnosis recorded within the year postpartum is very likely to be an indicator of an unidentified alcohol abuse during pregnancy. Women with heavy alcohol consumption are less likely to access antenatal care services which may limit the chances of detecting and offering support for these mothers-to-be. Moreover, some women with an alcohol-use disorder have the ability to conceal their alcohol abuse. The study by O'Leary et al identified that mothers with an alcohol-related diagnosis represent a high-risk population, due to the likelihood of heavy alcohol consumption during pregnancy and also because they are more often single, have a cosubstance use disorder or mental disorder and smoke during pregnancy. The increased risk of infant mortality, particularly SIDS, may be attributable to this high-risk profile or excessive alcohol levels. A previous study identified that an association between alcohol consumption and infant mortality was unchanged when SIDS or injury-related deaths were excluded from the analyses.1 Findings in this study supported that alcohol affects other causes of infant mortality, and is consistent with O'Leary and colleagues’ findings. Women with an alcohol diagnosis, recorded from age 10 and more than 1 year postpartum, represent a high-risk group and their infants have lower survival rates within the first year of life. More efforts should be made to identify these women during pregnancy or at least at time of birth.