Sepsis-Associated In-Hospital Births

Abstract

This cross-sectional study explores cases and risk factors of sepsis among birthing mothers in Mississippi from 2016 to 2025.

Introduction

Maternal sepsis is a life-threatening condition resulting from a dysregulated immune response to infection during pregnancy, childbirth, or after abortion.¹ Sepsis is responsible for an estimated 11% of maternal deaths globally.² In the US, sepsis and infections were the third leading cause of maternal mortality in 2024.³ Given the high mortality associated with this condition,⁴ we examined risk factors for maternal sepsis. We hypothesized that sepsis is associated with severe outcomes among birthing mothers in Mississippi, a US state with the third highest maternal mortality rate.⁵

Methods

We analyzed 2016-2025 statewide hospital discharge data from Mississippi residents aged 10 to 49 years. The University of Mississippi Medical Center Institutional Review Board approved this population-based cross-sectional study and waived informed consent because deidentified data were used. We followed the STROBE reporting guideline.

The outcome was sepsis-associated in-hospital births. Covariates were sociodemographic, obstetric, and clinical characteristics. Self-reported race categories were Black, White, and other (American Indian or Alaska Native, Asian, Native Hawaiian or Other Pacific Islander, and multiracial; Hispanic or Latino ethnicity) and analyzed to provide a demographic overview, identify possible high-risk populations, and adjust for upstream socioeconomic vulnerabilities. Race and ethnicity categories were collapsed due to small counts.

Methodological details and codes used for selecting delivery hospitalizations, sepsis cases, and comorbidities are provided in the eMethods and eTables 1 to 4 in Supplement 1. We performed independent, 2-tailed t tests to compare continuous variables; used joinpoint regression to evaluate the maternal sepsis trendline; and conducted logistic regression to obtain adjusted odds ratios (AORs). All tests were 2-sided, with significance set at P < .05. We conducted analyses with Joinpoint Regression, version 5.4.0.0 (National Cancer Institute) and SAS, version 9.4 (SAS Institute).

Results

We identified 197 sepsis cases (0.06%) among 320 179 hospital births in Mississippi during the study period. Of these cases, the mean (SD) maternal age was 27.3 (6.4) years; 125 mothers (63.5%) had Medicaid coverage; 116 (58.9%) identified as Black, 67 (34.0%) as White, and 14 (7.1) as other races; and 111 (56.3%) were nonmetropolitan residents (Table).

Table. Rates and Associations of Sepsis During Delivery Hospitalizations in Mississippi, 2016-2025.

Characteristic	Patients, No. (%)		AOR (95% CI)a	P valueb	Sepsis rate per 10 000 birthsc
	With sepsis	Without sepsis
Total number	197	319 982	NA	NA	6.2
Maternal risk factors
Age group, y
10-24	77 (39.1)	117 767 (36.8)	1 [Reference]	NA	6.5
25-34	88 (44.7)	168 560 (52.7)	0.85 (0.63-1.17)	.32	5.2
35-49	32 (16.2)	33 655 (10.5)	1.45 (0.93-2.21)	.09	9.5
Raced
Black	116 (58.9)	134 431 (42.0)	1.74 (1.28-2.39)	.001	8.6
White	67 (34.0)	159 501 (49.9)	1 [Reference]	NA	4.2
Othere	14 (7.1)	26 050 (8.1)	1.15 (0.61-2.01)	.65	NA
Primary expected payer
Medicaid	125 (63.5)	166 645 (52.1)	1.52 (1.09-2.13)	.01	7.5
Private insurance	57 (28.9)	131 347 (41.0)	1 [Reference]	NA	4.3
Otherf	15 (7.6)	21 990 (6.9)	1.68 (0.91-2.93)	.08	NA
Residence statusg
Metropolitan area	86 (43.7)	150 036 (46.9)	1 [Reference]	NA	5.7
Nonmetropolitan area	111 (56.3)	169 946 (53.1)	1.11 (0.84-1.48)	.46	6.5
Study period
Pre–COVID-19 pandemic	81 (41.1)	135 534 (42.3)	1 [Reference]	NA	6.0
COVID-19 pandemic	57 (28.9)	95 912 (30.0)	0.96 (0.68-1.34)	.79	5.9
Post–COVID-19 pandemic	59 (30.0)	88 536 (27.7)	1.07 (0.76-1.49)	.71	6.7
Comorbidity statush
Chronic comorbidity	105 (53.3)	106 812 (33.4)	2.17 (1.63-2.88)	<.001	9.8
Hypertension	72 (36.6)	67 399 (21.1)	2.04 (1.51-2.73)	<.001	10.7
Obesity	51 (25.9)	47 580 (14.9)	1.81 (1.30-2.49)	<.001	10.7
Diabetes	22 (11.2)	24 949 (7.8)	1.43 (0.89-2.20)	.12	8.8
Obstetric comorbidity index, mean (SD)	1.5 (2.0)	0.9 (1.4)	NA	<.001	NA
Mode of delivery
Cesarean birth	121 (61.4)	120 774 (37.7)	2.32 (1.73-3.12)	<.001	10.0
Vaginal birth	76 (38.6)	199 208 (62.3)	1 [Reference]	NA	3.8
Outcomes
Birth status
Preterm birth: <37 wk	106 (53.8)	45 811 (14.3)	5.90 (4.42-7.90)	<.001	23.1
Term birth: ≥37 wk	91 (46.2)	274 171 (85.7)	1 [Reference]		3.3
Resource utilization
Length of stay, mean (SD), d	10.4 (12.9)	2.5 (1.9)	NA	<.001	NA
Hospital charges, mean (SD), $	134 844.40 (193 520.20)	22 876.20 (21 410.00)	NA	<.001	NA

Abbreviations: AOR, adjusted odds ratio; NA, not applicable.

^a Regression models for comorbidities were adjusted for demographic characteristics (race, age, payer, and residence); the rest of the models were adjusted for demographic characteristics and chronic comorbidity (hypertension, obesity, and diabetes).

^b P values reflect the results from logistic regression analyses and t tests (continuous variables).

^c Crude rates for each group were calculated as sepsis cases divided by the group total per 10 000 in-hospital obstetric deliveries. Rates were not calculated when the count was fewer than 20 (ie, for other races and other payers).

^d Race was self-reported and obtained from medical records. Due to small counts, racial categories were aggregated into 3 groups.

^e Other races include American Indian or Alaska Native, Asian, Native Hawaiian or Other Pacific Islander, and multiracial. Hispanic or Latino ethnicity was included in the other races group (eMethods in Supplement 1). Rates for other races were not calculated due to small counts.

^f Other payers include Medicare, TRICARE or CHAMPUS, ChampVA, self-pay, and charity.

^g Metropolitan area has an urban core of more than 50 000 residents, whereas a nonmetropolitan area has fewer than 50 000 residents.

^h Chronic comorbidity was defined as the presence of at least 1 target condition: hypertension, obesity, or diabetes. The overall comorbidity status was assessed with the obstetric comorbidity index, a weighted summary score of preexisting illnesses and advanced maternal age (eMethods in Supplement 1).

The overall sepsis rate was 6.2 cases per 10 000 births; the rate was highest in the group aged 35 to 49 years and had distinct subgroup variations (Table). Specifically, the rate was higher for Black than White mothers (8.6 cases per 10 000 births vs 4.2 cases per 10 000 births) and for those with Medicaid than private insurance (7.5 cases per 10 000 births vs 4.3 cases per 10 000 births). Although the observed sepsis rate fluctuated annually, the modeled trendline was nearly flat with a nonsignificant increase of 0.29 (95% CI, −5.77 to 6.30) (Figure).

Figure. Bar and Line Graphs of Maternal Sepsis Cases and Rates in Mississippi, 2016-2025.

A and B, The highest sepsis rate (7.8 cases per 10 000 births) occurred in 2021 during the COVID-19 peak, while the highest number of cases (26) occurred in 2016. The rate surged twice: once in 2016 and again in 2021, when 11 of the 22 COVID-19–associated sepsis cases were reported. Aside from the 2021 increase, however, the sepsis rate remained low during the pandemic. These dynamics indicate that public health measures aimed at infectious disease control during pregnancy, such as COVID-19 vaccinations, helped reduce the overall risk of maternal sepsis during the pandemic. C and D, The excessively high chronic comorbidity rate (13.8 cases per 10 000 births) and cesarean birth rate (13.7 cases per 10 000 births) among Black mothers with sepsis, highlighting the importance of these risk factors. Rates for the other races group were not calculated due to small counts.

Risk factor analyses indicated that 53.5% (105 of 197) of mothers with sepsis also had hypertension, obesity, or diabetes and had higher odds of experiencing at least 1 of these conditions (AOR, 2.17; 95% CI, 1.63-2.88) than mothers without sepsis. Additionally, mothers with sepsis had markedly higher cesarean delivery rates (61.4% vs 37.7%) and increased odds of undergoing the procedure (AOR, 2.32; 95% CI, 1.73-3.12) than mothers without sepsis. Race-stratified analyses further indicated that the burdens of chronic illness (59.5% [69 of 116]) and cesarean delivery (62.9% [73 of 116]) were disproportionately elevated in Black mothers with sepsis (Figure). The most frequent potential infectious sources were urinary tract infections (52 cases) and pneumonia (49 cases); the top pathogens coded were Escherichia coli (31 cases) and SARS-CoV-2 infection (22 cases). Five of the 11 mothers who died from sepsis had COVID-19.

Severe complications were highly prevalent in the sepsis cohort, including respiratory distress syndrome (65 [33.0%]), acute kidney failure (46 [23.4%]), and septic shock (46 [23.4%]). Sepsis accounted for 18.3% (43 of 235) of maternal respiratory ventilations performed and 32.4% (11 of 34) of all delivery-associated hospital deaths during the study period, yielding a 5.6% (11 of 197) case fatality rate. The sepsis cohort had an excessively high prevalence of preterm births (53.8% vs 14.3%) and higher odds of preterm births (AOR, 5.90; 95% CI, 4.42-7.90) than the nonsepsis cohort. Moreover, 35 mothers with sepsis (17.8%) had stillbirths. Sepsis-associated hospital charges were higher than nonsepsis-associated charges ($134 844.40 vs $22 876.20), with total sepsis charges reaching $26 564 353.30.

Discussion

Although sepsis is a rare delivery complication, it is associated with excessive maternal morbidity and mortality, severe neonatal outcomes, and steep hospital charges. In Mississippi, key factors associated with maternal sepsis were Black race, Medicaid insurance, hypertension, and cesarean delivery.

Study limitations are its small sample size; its cross-sectional design, which lacks power to detect potential reverse causality between cesarean births and infections; and the implications of the COVID-19 pandemic for sepsis mortality and trends. Despite these limitations, we built a detailed epidemiological profile of Mississippi’s delivery-associated sepsis and pinpointed the need for creating a statewide prevention plan focused on identifying high-risk patients and treating infections promptly.

Cesarean delivery is a major risk factor for sepsis; proactive maternal care should target Mississippi’s excessive 37.9% surgical birth rate.^6,7 Our research supports the notion that sepsis is a complex, multifactorial condition that often emerges at the intersection of infectious triggers and preexisting chronic illnesses.^8,9 To address the entire spectrum of risk factors for sepsis and maternal deaths, Mississippi must intensify the fight against its entrenched and ongoing chronic disease epidemic.

Supplement 1.

eMethods

eTable 1. Codes Used to Identify Births in Hospital Discharge Data

eTable 2. Diagnostic and Procedure Codes for Sepsis and Severe Complications

eTable 3. Diagnostic Codes for Clinical Conditions and Outcomes

eTable 4. Diagnostic Codes Used in the Obstetric Comorbidity Index

eReferences

Supplement 2.

Data Sharing Statement