Association between pregnancy affected by vaginal bleeding and women's mortality: A cohort study

Abstract

Objective: To investigate the association between vaginal bleeding (VB) in pregnancy and women's mortality, using VB-unaffected pregnancies, terminations and miscarriages as comparators. Design: Observational cohort study. Setting: Nationwide registries of Denmark linked at an individual level. Population or Sample: 1 354 181 women and their 3 162 317 pregnancies (1979–2017), including 70 835 VB-affected pregnancies and comparators: 2 236 359 VB-unaffected pregnancies ending in childbirth; 589 697 terminations; and 265 426 miscarriages. Methods: We followed pregnancies until the earliest date of woman's death, emigration or end of data. Main outcome measures: All-cause and cause-specific mortality rates per 10 000 person-years (PY) and hazard ratios (HRs) with 95% confidence intervals (CIs) adjusted using Cox proportional hazards regression for age, calendar year, pre-existing chronic conditions and socio-economic factors. Results: There were 2320 deaths from any cause among women following VB-affected pregnancy (mortality rate 15.2, 95% CI 14.6–15.9 per 10 000 PY); 55 030 deaths following VB-unaffected pregnancy (mortality rate 12.7, 95% CI 12.6–12.8); 27 500 deaths following a termination (mortality rate 21.9, 95% CI 21.6–22.1), and 10 865 deaths following a miscarriage (mortality rate 19.2, 95% CI 18.8–19.6). For comparison of VB-affected versus VB-unaffected pregnancies, associations with all-cause (HR 1.14, 95% CI 1.09–1.19), natural causes (HR 1.15, 95% CI 1.09–1.22) and non-natural causes (HR 1.27, 95% CI 1.08–1.48) mortality were attenuated in a sensitivity analysis of pregnancies recorded in 1994–2017 (HR 1.00, 95% CI 0.90–1.12, HR 0.98, 95% CI 0.85–1.14 and HR 1.04, 95% CI 0.72–1.51, respectively). Contrasts with remaining comparators did not suggest increased risks of all-cause, natural or non-natural mortality causes. Conclusion: We found no evidence of an increased risk of women's mortality following VB-affected versus VB-unaffected pregnancy, termination or miscarriage.