Kasman · Human reproduction (Oxford, England) 2021 · Retrospective cohort study · n=958,804 pregnancies

Association between preconception paternal health and pregnancy loss in the USA: an analysis of US claims data.

Cited 44 times in the scientific literature.

Level 3 - non-randomized controlled study

Non-randomized retrospective cohort study using an administrative claims database.

PubMed 33336240 · doi:10.1093/humrep/deaa332 · record verified 2026-08-29

What was done

A retrospective cohort study was conducted using a US employer-based insurance claims database from 2007 to 2016, analyzing 958,804 pregnancies. Researchers examined the association between preconception paternal health—including metabolic syndrome (MetS) components, the Charlson Comorbidity Index (CCI), and individual chronic diseases—and pregnancy loss (ectopic pregnancy, miscarriage, and stillbirth), adjusting and stratifying for maternal age and health status.

What was found

Of 958,804 pregnancies (mean paternal age 35.3 ± 5.3 years, mean maternal age 33.1 ± 4.4 years), 22% ended in pregnancy loss. After adjusting for maternal factors, the risk of pregnancy loss increased with higher paternal MetS components compared to men with no components: one component (RR 1.10, 95% CI 1.09–1.12), two components (RR 1.15, 95% CI 1.13–1.17), and three or more components (RR 1.19, 95% CI 1.14–1.24). Increased risk was consistently observed for spontaneous abortion, stillbirth, and ectopic pregnancy, as well as across higher CCI scores and specific chronic conditions. The association between paternal health and the timing of pregnancy loss was weak.

Why it matters

This study highlights that preconception paternal health independently contributes to the risk of pregnancy loss. It suggests that clinical preconception guidance should include optimizing metabolic and general health in men, not just women.

Limits

The study is retrospective and relies on administrative billing claims, which lack direct physiological measurements, behavioral factors, and granular clinical details. The cohort was restricted to commercially insured individuals in the US, limiting generalizability to uninsured, public-payer, or non-US populations.

Cited by