AFFILIATIONS
1 Ottawa Hospital Research Institute, Ottawa, Ontario, Canada, University of Ottawa, Ottawa, Ontario, Canada
2 University of Toronto, Toronto, Ontario, Canada
3 ICES uOttawa, Ottawa, Ontario, Canada, Ottawa Hospital Research Institute, Ottawa, Ontario, Canada
4 University of Calgary, Calgary, Alberta, Canada
5 University of Ottawa, Ottawa, Ontario, Canada
6 Hamilton Health Sciences, Hamilton, Ontario, Hamilton, Ontario, Canada, McMaster University, Hamilton, Ontario, Hamilton, Ontario, Canada, McMaster University, Hamilton, Ontario, Hamilton, Ontario, Canada
7 ICES uOttawa, Ottawa, Ontario, Canada, Ottawa Hospital Research Institute, Ottawa, Ontario, Canada
8 Ottawa Hospital Research Institute, Ottawa, Ontario, Canada, University of Ottawa, Ottawa, Ontario, Canada
9 University of Ottawa, Ottawa, Ontario, Ottawa, Ontario, Canada
10 Bruyère Health Research Institute, Ottawa, Ontario, Canada
11 ICES uOttawa, Ottawa, Ontario, Canada, University of Ottawa, Ottawa, Ontario, Canada
12 Duke University School of Medicine, Durham, NC, Durham, North Carolina, United States, Duke University School of Medicine, Durham, NC, Durham, North Carolina, United States, Duke University School of Medicine, Durham, NC, Durham, North Carolina, United States
13 University of Ottawa, Ottawa, Ontario, Canada
14 University of Ottawa, Ottawa, Ontario, Canada
ABSTRACT
Background: Previous studies using hospitalization databases in the United States have reported higher pregnancy-associated venous thromboembolism (VTE) incidence among Black compared with White individuals. This disparity has often been attributed to differences in the prevalence of VTE risk factors, some of which are socially determined (e.g., obesity) and others primarily genetic (e.g., sickle cell trait). However, VTE events are diagnosed and managed in both inpatient and outpatient settings, and incidence estimates may differ depending on how comprehensively these events are captured. Our prior work has identified higher 30-day hospitalization rates following diagnosis of pregnancy-associated VTE among Black birthing individuals compared to White individuals. This finding therefore raises the possibility that case ascertainment could influence observed racial differences in pregnancy-associated VTE incidence.
Objective: To determine whether estimates of racial differences in pregnancy-associated VTE incidence vary according to VTE case ascertainment algorithms.
Methods: We conducted a population-based cohort study using linked administrative health data in Ontario, Canada, analyzed at ICES. Individuals with live or stillbirths between April 2012 and March 2022 were followed from pregnancy onset to 6 weeks postpartum. Exposure was maternal racial category (Asian, Black, or White) captured on the antenatal serum screening requisition. The primary outcome was pregnancy-associated VTE, identified using validated diagnostic and imaging procedure codes. Adjusted risk ratios (aRRs) were calculated after accounting for age, gestational age, obstetrical comorbidities, number of livebirths, mode of delivery, body mass index, and the frequency of healthcare use in the 5 years prior to pregnancy (hospitalizations, emergency department encounters, outpatient visits). We compared adjusted relative risks for pregnancy-associated VTE incidence by racial group using 3 case identification algorithms: 1) inpatient and outpatient diagnoses, 2) inpatient diagnoses only, and 3) a requirement for a second VTE diagnosis code (either inpatient or outpatient).
Results: Among 305,326 birthing individuals (17.6% Asian, 6.5% Black, 75.9% White), 1,793 developed VTE during pregnancy and the puerperium (5.9 per 1,000 individuals). Compared with White individuals, VTE incidence was lower among Asian (aRR 0.48, 95% CI 0.41–0.57) and Black individuals (aRR 0.73, 95% CI 0.61–0.89) using case ascertainment that included both inpatient and outpatient diagnoses. Findings were consistent when a second VTE diagnosis code was required. On the other hand, restricting to inpatient-only databases resulted in Black individuals having a numerically higher, non-statistically significant difference in VTE incidence compared with White individuals (aRR 1.36, 95% CI 0.72–2.56).
Conclusion: In a population-based cohort study with inpatient and outpatient case ascertainment, Asian and Black birthing individuals had lower risks of pregnancy-associated VTE compared with White individuals, while restricting to inpatient-only diagnoses led to different results in Black individuals. Racial differences in pregnancy-associated VTE incidence were sensitive to method of case ascertainment. The impact of VTE case identification algorithm on observed racial differences in VTE epidemiology requires further investigation.