Trends in Risk of Pregnancy Loss Among US Women by Metropolitan Status, 2000-2018

Approximately 20% of pregnancies end in spontaneous loss, an outcome associated with adverse health consequences. Statistically stable estimates of the risk of pregnancy loss are limited for nonmetropolitan populations due to small sample sizes.

This study evaluated the utility of the enhanced Modified Kalman Filter (eMKF) in producing estimates of the risk of pregnancy loss for subgroups of US women with small sample sizes to examine recent trends.

Data from the National Survey of Family Growth (NSFG; 2006-2019) were used to estimate trends in the risk of self-reported pregnancy loss (miscarriage, stillbirth, ectopic pregnancy) among US women (15-44 years) who reported at least one completed pregnancy (excluding induced abortions) conceived during 2000-2018 (n = 17,314 women, 35,988 pregnancies) by metropolitan status and maternal age. The eMKF was used to smooth estimates over groups and time. We compared the relative 95% confidence intervals (95% CIs) of model-based estimates to direct estimates to assess improvements in precision.

Among completed pregnancies conceived during 2000-2018, 21.6% ended in pregnancy loss. Relative 95% CIs for model-based estimates were 33.0% and 53.0% smaller for metropolitan and nonmetropolitan groups, respectively, than direct estimates. After adjustment, the risk of pregnancy loss for women ages 15-44 increased by a relative 1% annually for both metropolitan (risk ratio [RR] 1.01, 95% CI 1.01, 1.02) and nonmetropolitan (RR 1.01, 95% CI 1.00, 1.01) women. The risk of pregnancy loss increased for metropolitan women ages 15-19 (RR 1.01, 95% CI 1.00, 1.01), 20-24 (RR 1.01, 95% CI 1.00, 1.01), 25-29 (RR 1.02, 95% CI 1.01, 1.02), and 30-34 (RR 1.02, 95% CI 1.01, 1.03).

Risk of pregnancy loss increased by a relative 1% annually for women overall, and by 1%-2% annually among subgroups of women ages 15-34 in metropolitan areas. The eMKF provided improvements in estimate precision relative to direct estimates.