The association between race/ethnicity and peripartum hysterectomy and the risk of perioperative complications

Research output: Contribution to journalArticlepeer-review


Objective: To compare perioperative outcomes by patient race/ethnicity. Methods: A retrospective cohort study identified 7 331 638 childbirth hospitalizations for women aged 12–55 years in the USA between 2004–2014. Peripartum hysterectomy, in-hospital mortality, perioperative complications, length of stay, and cost of hysterectomy data were analyzed using SAS. Results: Among childbirth hospitalizations (52.9% white, 13.5% black, 23.0% Hispanic, 5.2% Asian, and 5.4% other), peripartum hysterectomy occurred in 6619. The incidence of peripartum hysterectomy was 90.3 (95% confidence interval [CI] 87.7–93.0) per 100 000 hospitalizations, and higher for black (111.0, 95% CI 104.5–117.4), Hispanic (104.9, 95% CI 99.1–110.8), and Asian women (119.6, 95% CI 109.1–130.2) compared to whites (75.7, 95% CI 72.8–78.5). After adjustment, Hispanic women had an 18% higher odds of undergoing peripartum hysterectomy (odds ratio [OR] 1.18, 95% CI 1.08–1.29; P=0.004) than white women. Non-white women had a 2–3-fold higher odds of in-hospital mortality (ORblack 2.76, 95% CI 1.44–5.30; ORHispanic 1.99, 95% CI 1.04–3.82; ORAsian+other 2.44, 95% CI 1.11–5.40. Black and Asian/other women were more likely to undergo blood transfusions. Conclusion: Women of color have higher rates of peripartum hysterectomy and experience higher rates of poor perioperative outcomes and mortality.

Original languageEnglish (US)
Pages (from-to)57-66
Number of pages10
JournalInternational Journal of Gynecology and Obstetrics
Issue number1
StatePublished - Oct 1 2020


  • Ethnicity
  • Perioperative complications
  • Peripartum hysterectomy
  • Pregnancy
  • Race

ASJC Scopus subject areas

  • Obstetrics and Gynecology


Dive into the research topics of 'The association between race/ethnicity and peripartum hysterectomy and the risk of perioperative complications'. Together they form a unique fingerprint.

Cite this