Addressing the smoking-hypertension paradox in pregnancy: insight from a multiethnic US birth cohort.

Background: Smoking during pregnancy has been associated with reduced risk of a spectrum of hypertensive (HTN) disorders, known as the "smoking-hypertension paradox." Objective: We sought to test potential epidemiologic explanations for the smoking-hypertension paradox. Methods: We analyzed 8510 pregnant people in the Boston Birth Cohort, including 4027 non-Hispanic Black and 2428 Hispanic pregnancies. Study participants self-reported tobacco, alcohol, cannabis, opioids, or cocaine use during pregnancy. We used logistic regression to assess effect modification by race/ethnicity, and confounding of concurrent substances on hypertensive disorders or prior pregnancy. We also investigated early gestational age as a collider or competing risk for pre-eclampsia, using cause-specific Cox models and Fine-Gray models, respectively. Results: We replicated the paradox showing smoking to be protective against hypertensive disorders among Black participants who used other substances as well (aOR: 0.61, 95% CI: 0.41, 0.93), but observed null effects for Hispanic participants (aOR: 1.14, 95% CI: 0.55, 2.36). In our cause-specific Cox regression, the effects of tobacco use were reduced to null effects with pre-eclampsia (aOR: 0.81, 95% CI: 0.63, 1.04) after stratifying for preterm birth. For the Fine-Gray competing risk analysis, the paradoxical associations remained. The smoking paradox was either not observed or reversed after accounting for race/ethnicity, other substance use, and collider-stratification due to preterm birth. Conclusions: These findings offer new insights into this paradox and underscore the importance of considering multiple sources of bias in assessing the smoking-hypertension association in pregnancy.