The Potential Association Between Prenatal Cannabis Use and Congenital Anomalies
The Potential Association Between Prenatal Cannabis Use and Congenital Anomalies
复制标题
DOI：
10.1097/adm.0000000000000639
复制
发表时间：
2020-12-01
影响因子：
5.5
通讯作者：
Corsi, Daniel J.
中科院分区：
文献类型：
--
作者：
Corsi, Daniel J.
C ongenital anomalies include birth defects and fetal malformations that may be detected prenatally, at birth, or in later childhood. 1, 2 In Canada, congenital anomalies surveillance collects data on anomalies occurring among live births and stillbirths, however data are limited and incomplete on congenital anomalies in terminated pregnancies following prenatal diagnosis. Congenital anomalies are birth outcomes of particular concern due to their strong association with infant morbidity and mortality, implications for quality of life in children and increased use of health services. Congenital anomalies are heterogeneous and may be related to chromosomal abnormalities or other structural and functional defects occurring during intrauterine development. Although many causes and risk factors for non-genetic congenital anomalies are established (eg, Zika virus 3 and rubella infection, 4 and exposure to teratogenic medications such as thalidomide), 5 the causes and risk factors for many anomalies remain unknown. Research to identify causal factors is of public health importance and may provide opportunities to intervene and reduce the population burden of congenital anomalies. In this issue of the Journal of Addiction Medicine, Reece and Hulse present an analysis of cannabis consumption and rates of congenital anomalies. 6 Using data at the provincial/territorial level in Canada and advanced geospatial methods, the authors demonstrate a positive correlation between rates of cannabis use, and rates of congenital anomalies overall, as well as a correlation with rates of several specific anomalies (ie, cardiovascular defects, orofacial clefts, Down syndrome, and gastroschisis). These findings are notable given the increasing social acceptance of cannabis use and the recent legalization of cannabis for recreational use in Canada. This work also presents an opportunity to highlight some of the methodological challenges which exist when trying to identify risk factors for congenital anomalies and in the use and interpretation of aggregate-level data.What Can We Learn From Aggregate Data? There are several methodological issues and challenges in interpreting the study as presented. Despite the impressive geospatial analyses, the underlying data in the study are aggregate, and as such, require careful interpretation and inference that is appropriate for this type of study design. As an illustration, a simple model for these data is presented in the Figure, which displays Pearson correlations coefficients between rates of anomalies and cannabis consumption at the provincial level. These results suggest that the association may not be as robust or consistent across groups of anomalies as implied by the more sophisticated analyses, particularly for rates of total anomalies and rates of age-adjusted gastroschisis, which were not statistically significantly correlated with cannabis use. Moreover, the correlations between provincial rates of anomalies and cannabis use were similar in direction and magnitude to those for tobacco smoking (Fig. 1). Aggregate or ecological data have well-known limitations when trying to make inferences at the individual level. In this context, although provincial-level cannabis use correlates with province-wide rates of congenital anomalies, it is not possible, given these data, to infer that individual birth outcomes are causally related to cannabis use. The conventional interpretation is that ecological studies provide relatively weak evidence for causality. However, this limitation could be overcome by using time series data or examining changes in the rates of congenital anomalies and the rates of cannabis use. Unfortunately, such data are …
