Methods for the assessment of selection bias in drug safety during pregnancy studies using electronic medical data

Miscarriage after SARS-CoV-2 vaccination: A population-based cohort study

OBJECTIVE

To evaluate the risk of miscarriage following SARS-CoV-2 vaccination, while accounting for the competing risk of induced abortion.

DESIGN

Population-based cohort study.

SETTING

Ontario, Canada.

PARTICIPANTS

Women aged 15-50 years with a confirmed pregnancy at ≤19 completed weeks' gestation.

METHODS

Exposure to first SARS-CoV-2 vaccination, handled in a time-varying manner, was defined as (i) unvaccinated, (ii) remotely vaccinated >28 days before the estimated conception date or (iii) recently vaccinated ≤28 days before conception and up to 120 days after conception.

MAIN OUTCOME MEASURES

The outcome was miscarriage, occurring between the estimated date of conception and up to 19 completed weeks of pregnancy. Fine-Grey hazard models, accounting for the competing risk of induced abortion, generated hazard ratios (aHR), adjusted for socio-demographic factors, comorbidities, and biweekly periods.

RESULTS

Included were 246 259 pregnant women, of whom 34% received a first SARS-CoV-2 vaccination. Miscarriage occurred at a rate of 3.6 per 10 000 person-days among remotely vaccinated women and 3.2 per 10 000 person-days among those recently vaccinated, in contrast to a rate of 1.9 per 10 000 person-days among unvaccinated women, with corresponding aHR of 0.98 (95% confidence interval [CI] 0.91-1.07) and 1.00 (95% CI 0.93-1.08).

CONCLUSIONS

SARS-CoV-2 vaccination was not associated with miscarriage while accounting for the competing risk of induced abortion. This study reiterates the importance of including pregnant women in new vaccine clinical trials and registries, and the rapid dissemination of vaccine safety data.

A potential outcomes approach to defining and estimating gestational age-specific exposure effects during pregnancy

Many studies seek to evaluate the effects of potentially harmful pregnancy exposures during specific gestational periods. We consider an observational pregnancy cohort where pregnant individuals can initiate medication usage or become exposed to a drug at various times during their pregnancy. An important statistical challenge involves how to define and estimate exposure effects when pregnancy loss or delivery can occur over time. Without proper consideration, the results of standard analysis may be vulnerable to selection bias, immortal time-bias, and time-dependent confounding. In this study, we apply the “target trials” framework of Hernán and Robins in order to define effects based on the counterfactual approach often used in causal inference. This effect is defined relative to a hypothetical randomized trial of timed pregnancy exposures where delivery may precede and thus potentially interrupt exposure initiation. We describe specific implementations of inverse probability weighting, G-computation, and Targeted Maximum Likelihood Estimation to estimate the effects of interest. We demonstrate the performance of all estimators using simulated data and show that a standard implementation of inverse probability weighting is biased. We then apply our proposed methods to a pharmacoepidemiology study to evaluate the potentially time-dependent effect of exposure to inhaled corticosteroids on birthweight in pregnant people with mild asthma.

Quantifying the influence of bias in reproductive and perinatal epidemiology through simulation

PURPOSE

The application of simulated data in epidemiological studies enables the illustration and quantification of the magnitude of various types of bias commonly found in observational studies. This was a review of the application of simulation methods to the quantification of bias in reproductive and perinatal epidemiology and an assessment of value gained.

METHODS

A search of published studies available in English was conducted in August 2020 using PubMed, Medline, Embase, CINAHL, and Scopus. A gray literature search of Google and Google Scholar, and a hand search using the reference lists of included studies was undertaken.

RESULTS

Thirty-nine papers were included in this study, covering information (n = 14), selection (n = 14), confounding (n = 9), protection (n = 1), and attenuation bias (n = 1). The methods of simulating data and reporting of results varied, with more recent studies including causal diagrams. Few studies included code for replication.

CONCLUSIONS

Although there has been an increasing application of simulation in reproductive and perinatal epidemiology since 2015, overall this remains an underexplored area. Further efforts are required to increase knowledge of how the application of simulation can quantify the influence of bias, including improved design, analysis and reporting. This will improve causal interpretation in reproductive and perinatal studies.

Association between the timing of asthma diagnosis and medication use during pregnancy

STUDY OBJECTIVE

To assess whether asthma medication use during pregnancy differs in women newly diagnosed with asthma early in pregnancy (first 19 weeks of pregnancy) compared to those newly diagnosed up to 2 years pre-pregnancy.

DESIGN

A retrospective population-based cohort study.

DATA SOURCE

To conduct this study, we used the Quebec Asthma and Pregnancy Database (QAPD) constructed by linking two administrative health databases from the province of Quebec (Canada): the Régie de l'Assurance Maladie du Québec and Maintenance et Exploitation des Données pour l'Étude de la Clientèle Hospitalière databases.

PATIENTS

A cohort comprising pregnant women newly diagnosed with asthma at any time in the 2 years prior to pregnancy or during the first 19 weeks of pregnancy was selected from the QAPD.

MEASUREMENTS AND MAIN RESULTS

We assessed the number of filled prescriptions of inhaled corticosteroids (ICS), ICS/long-acting β2 agonists (LABA), and short-acting β2 agonists (SABA), as well as the number of days' supply of oral corticosteroid (OCS) from the 20^th week of pregnancy until delivery. Poisson regression was used to compare the rates of asthma medication use in women diagnosed pre-pregnancy versus early in pregnancy. The cohort included 1731 women newly diagnosed with asthma pre-pregnancy and 359 women newly diagnosed with asthma early in pregnancy. Women diagnosed early in pregnancy were more likely to use ICS (adjusted rate ratio: 1.9, 95% confidence interval (CI): 1.6-2.3) and SABA (adjusted rate ratio: 2.0, 95% CI: 1.7-2.4) from the 20th week of pregnancy until delivery than those newly diagnosed pre-pregnancy. No significant differences were observed in the use of ICS/LABA [adjusted rate ratio: 0.9, 95% CI: 0.7-1.3] and OCS [adjusted rate ratio: 0.8, 95% CI: 0.6-1.2].

CONCLUSION

The higher use of ICS and SABA observed in women newly diagnosed with asthma early in pregnancy may suggest a more persistent asthma phenotype caused by pregnancy-triggered hormonal changes.

Medicines prescribed for asthma, discontinuation and perinatal outcomes, including breastfeeding: A population cohort analysis

OBJECTIVES

To explore associations between exposures to medicines prescribed for asthma and their discontinuation in pregnancy and preterm birth [<37 or <32 weeks], SGA [<10th and <3rd centiles], and breastfeeding at 6-8 weeks.

METHODS

Design. A population-based cohort study. Setting. The Secure Anonymised Information Linkage [SAIL] databank in Wales, linking maternal primary care data with infant outcomes. Population. 107,573, 105,331, and 38,725 infants born 2000-2010 with information on premature birth, SGA and breastfeeding respectively, after exclusions. Exposures. maternal prescriptions for asthma medicines or their discontinuation in pregnancy. Methods. Odds ratios for adverse pregnancy outcomes were calculated for the exposed versus the unexposed population, adjusted for smoking, parity, age and socio-economic status.

RESULTS

Prescriptions for asthma, whether continued or discontinued during pregnancy, were associated with birth at<32 weeks' gestation, SGA <10th centile, and no breastfeeding (aOR 1.33 [1.10-1.61], 1.10 [1.03-1.18], 0.93 [0.87-1.01]). Discontinuation of asthma medicines in pregnancy was associated with birth at<37 weeks' and <32 weeks' gestation (aOR 1.22 [1.06-1.41], 1.53 [1.11-2.10]). All medicines examined, except ICS and SABA prescribed alone, were associated with SGA <10th centile.

CONCLUSIONS

Prescription of asthma medicines before or during pregnancy was associated with higher prevalence of adverse perinatal outcomes, particularly if prescriptions were discontinued during pregnancy. Women discontinuing medicines during pregnancy could be identified from prescription records. The impact of targeting close monitoring and breastfeeding support warrants exploration.

Timing of Maternal Asthma Diagnosis in Relation to Adverse Perinatal Outcomes

BACKGROUND

It is unclear if asthma diagnosed during pregnancy puts the fetus at a higher risk of poor perinatal outcomes than pre-existing asthma.

OBJECTIVE

To assess if the risks of prematurity, major malformations, and small-for-gestational age (SGA) are higher in women with asthma diagnosed during versus pre-pregnancy.

METHODS

We retrospectively analyzed a cohort of pregnant women aged ≥15 years with and without incident asthma, constructed from health administrative databases. Follow-up began 24 months before pregnancy onset (cohort entry) and ended at delivery. Incident asthma was defined as a first diagnosis among those without asthma in the 8 years before cohort entry. Time was classified into pre-pregnancy and each trimester until delivery. We fit inverse probability weighted Poisson models to estimate marginal relative risks (RRs) for prematurity (delivery <37th week), major malformations, and SGA (birth weight <10th percentile) comparing women with and without asthma, assessing the asthma timing of diagnosis interaction term via a Wald test.

RESULTS

In a cohort of 122,880 deliveries, the increased risk of prematurity, but not SGA, due to incident asthma was higher in those diagnosed during the second (RR, 1.34; 95% confidence interval [CI], 1.08-1.65; Wald P = .05) and third (RR, 1.93; 95% CI, 1.62-2.29; Wald P < .01) trimesters relative to pre-pregnancy (RR, 1.06; 95% CI, 0.98-1.15). A trend toward an increased risk of major malformations was observed in those diagnosed during the first trimester (RR, 1.18; 95% CI, 0.94-1.49; Wald P = .15) than pre-pregnancy (RR, 0.99; 95% CI, 0.92-1.07).

CONCLUSIONS

Asthma diagnosed during, versus before, pregnancy was associated with a greater prematurity risk, suggesting an important role of preconception and prenatal screening.