Association between stillbirth and risk factors known at pregnancy confirmation

Abstract

Context: Stillbirths account for almost half of US deaths from 20 weeks' gestation to 1 year of life. Most large studies of risk factors for stillbirth use vital statistics with limited data.

Objective: To determine the relation between stillbirths and risk factors that could be ascertained at the start of pregnancy, particularly the contribution of these factors to racial disparities.

Design, setting, and participants: Multisite population-based case-control study conducted between March 2006 and September 2008. Fifty-nine US tertiary care and community hospitals, with access to at least 90% of deliveries within 5 catchment areas defined by state and county lines, enrolled residents with deliveries of 1 or more stillborn fetuses and a representative sample of deliveries of only live-born infants, oversampled for those at less than 32 weeks' gestation and those of African descent.

Main outcome measure: Stillbirth.

Results: Analysis included 614 case and 1816 control deliveries. In multivariate analyses, the following factors were independently associated with stillbirth: non-Hispanic black race/ethnicity (23.1% stillbirths, 11.2% live births) (vs non-Hispanic whites; adjusted odds ratio [AOR], 2.12 [95% CI, 1.41-3.20]); previous stillbirth (6.7% stillbirths, 1.4% live births); nulliparity with (10.5% stillbirths, 5.2% live births) and without (34.0% stillbirths, 29.7% live births) previous losses at fewer than 20 weeks' gestation (vs multiparity without stillbirth or previous losses; AOR, 5.91 [95% CI, 3.18-11.00]; AOR, 3.13 [95% CI, 2.06-4.75]; and AOR, 1.98 [95% CI, 1.51-2.60], respectively); diabetes (5.6% stillbirths, 1.6% live births) (vs no diabetes; AOR, 2.50 [95% CI, 1.39-4.48]); maternal age 40 years or older (4.5% stillbirths, 2.1% live births) (vs age 20-34 years; AOR, 2.41 [95% CI, 1.24-4.70]); maternal AB blood type (4.9% stillbirths, 3.0% live births) (vs type O; AOR, 1.96 [95% CI, 1.16-3.30]); history of drug addiction (4.5% stillbirths, 2.1% live births) (vs never use; AOR, 2.08 [95% CI, 1.12-3.88]); smoking during the 3 months prior to pregnancy (<10 cigarettes/d, 10.0% stillbirths, 6.5% live births) (vs none; AOR, 1.55 [95% CI, 1.02-2.35]); obesity/overweight (15.5% stillbirths, 12.4% live births) (vs normal weight; AOR, 1.72 [95% CI, 1.22-2.43]); not living with a partner (25.4% stillbirths, 15.3% live births) (vs married; AOR, 1.62 [95% CI, 1.15-2.27]); and plurality (6.4% stillbirths, 1.9% live births) (vs singleton; AOR, 4.59 [95% CI, 2.63-8.00]). The generalized R(2) was 0.19, explaining little of the variance.

Conclusion: Multiple risk factors that would have been known at the time of pregnancy confirmation were associated with stillbirth but accounted for only a small amount of the variance in this outcome.

Figure. Study Enrollment and Inclusion in Case-Control Analysis

Details of the study design, methods, and recruitment have been reported. In brief, recruitment was through 59 hospitals in 5 geographic regions defined by state and county lines that included the State of Rhode Island and Bris-tol County, Massachusetts; DeKalb County, Georgia; Galveston and Brazoria Counties, Texas; Bexar County, Texas; and Salt Lake County, Utah. Analysis was restricted to participants with a complete or partial maternal interview and chart abstraction. Within this group, comparisons were made overall (all stillbirths compared with all live births) as well as for 2 subgroups (all stillbirths compared with all live births among deliveries at ≥24 weeks’ gestation and antepartum stillbirths compared with all live births among deliveries of nonanomalous singletons at ≥24 weeks’ gestation). Actual numbers of participants are shown; for weighted sample sizes, see Tables 2 through 5.