Trends in mortality from cancer and sepsis in the United States, 1999 to 2023: An analysis of CDC WONDER data

Abstract

Cancer remains the second leading cause of death in the United States (U.S.), with increasing incidence among younger adults. Sepsis, a life-threatening organ dysfunction caused by infection, frequently complicates cancer due to disease-related and treatment-induced immunosuppression. Understanding cancer–sepsis mortality trends is essential for identifying high-risk populations and improving prevention and clinical management strategies. We analyzed U.S. mortality data from the Centers for Disease Control and Prevention Wide-ranging Online Data for Epidemiologic Research Multiple Cause of Death database to identify deaths among adults aged ≥ 25 years in which both cancer (International Classification of Diseases, Tenth Revision C00–D48) and sepsis (International Classification of Diseases, Tenth Revision A41) were listed on the same death certificate, representing concomitant cancer–sepsis mortality. Age-adjusted mortality rates per 100,000 population were calculated and temporal trends were assessed using Joinpoint regression. Statistical significance was set at P < .05. From 1999 to 2023, a total of 20,108,555 U.S. adults aged 25+ died from cancer complicated by sepsis. From 1999 to 2023, the average annual percentage change (AAPC) decreased by 1.09% (95% confidence interval, −1.33–−0.84%). Rates fell from 417.6 per 100,000 in 1999 to 322.0 in 2023, showing a steady decline until 2018 (annual percent change [APC]: −1.32%), a brief rise from 2018 to 2021 (APC: 2.17%), and a significant drop from 2021 to 2023 (APC: −3.65%). Mortality was highest in men (AAPC: −1.33%) and Black or African American individuals (AAPC: −1.68%), as well as in nonmetropolitan areas (AAPC: −0.60%) and the Southern U.S. (AAPC: −1.02%). Most age groups experienced significant declines, particularly middle-aged adults, while the youngest (25–34) and oldest (85+) age groups showed minimal long-term change. All racial and ethnic groups showed improvement, although American Indian or Alaska Native trends remained largely flat until a notable decline (APC: −8.26% from 2021 to 2023). Most deaths occurred in hospitals (38%) or at home (35%), with fewer in nursing homes, hospices, or other settings. Concomitant cancer–sepsis mortality declined overall in the U.S. from 1999 to 2023, but substantial disparities persist by sex, race/ethnicity, geography, and urbanization, reflecting inequities in access to oncology care, critical care, and timely sepsis management.

1. Introduction

From 1999 to 2023, trends in United States (U.S.) sepsis and cancer-related mortality have shown both improvements and new challenges. Cancer remains the second leading cause of death, but thanks to better treatments and decreases in smoking, age-adjusted cancer death rates have dropped, resulting in nearly 4 million lives saved since 1991.^[1,2] However, current epidemiologic analyses of trends indicate a rising incidence of cancer among young adults, particularly colorectal, pancreatic, and endometrial cancers, which could lead to shifts in risk patterns and later diagnoses in these groups.^[3,4]

Sepsis, which is life-threatening organ dysfunction caused by a dysregulated host response to infection, exhibits a more complex mortality trend. Age-adjusted sepsis mortality rates from 1999 to 2019 remained relatively stable at around 76 to 77 deaths per 100,000 people.^[5] However, the coronavirus disease 2019 (COVID-19) pandemic completely disrupted this pattern: between 2019 and 2021, sepsis mortality increased by over 30%, with a large portion attributable directly to severe acute respiratory syndrome coronavirus 2 infection (6). Ongoing disparities persist, with higher mortality rates among men, non-Hispanic Blacks, American Indian/Alaska Natives, and residents of the southern U.S.^[6] Elevated death rates are strongly related to the number of failing organs and the presence of comorbidities.^[7]

These converging forces highlight the 2-pronged challenge facing public health in the U.S.: maintaining momentum in cancer prevention and care while addressing the ongoing and emerging burden of sepsis, especially in the context of emerging infectious diseases. This in-depth analysis aims to provide insight into the incidence, prevalence, and mortality rates of cancer and sepsis. This area has been historically under-researched. By addressing disparities among different groups, measures can be implemented to prevent future issues. Accordingly, this study examines long-term trends in concomitant cancer–sepsis mortality, defined as deaths in which both cancer and sepsis are recorded on the same death certificate, using nationally representative Centers for Disease Control and Prevention (CDC) Wide-ranging Online Data for Epidemiologic Research (WONDER) multiple cause of death data.^[8,9]

2. Methodology

2.1. Study setting and population

We analyzed mortality data from the CDC WONDER Multiple Cause of Death database.^[9], identifying deaths among U.S. adults aged ≥ 25 years between 1999 and 2023 in which both cancer (International Classification of Diseases, Tenth Revision C00–D48) and sepsis (International Classification of Diseases, Tenth Revision A41) were listed as contributing causes of death. This approach captures concomitant cancer–sepsis mortality, rather than cancer-only or sepsis-only deaths. This study did not require approval from the local institutional review board because it used a de-identified, government-issued public use dataset and adhered to the Strengthening the Reporting of Observational Studies in Epidemiology guidelines for reporting.^[10]

2.2. Statistical analysis

To examine national trends in concomitant cancer–sepsis mortality, we calculated crude and age-adjusted mortality rates (AAMRs) per 100,000 people, broken down by sex, race/ethnicity, census region, and urbanization status, with 95% confidence intervals (CIs) covering 1999 to 2023. Crude mortality rates were obtained by dividing the number of cancer- and sepsis-related deaths by the corresponding U.S. population of each year. AAMRs were derived by standardizing mortality rates to the 2000 U.S. standard population to the 2000 U.S. population.^[11] The JoinPoint Regression Program (Joinpoint V 5.2.0.0, National Cancer Institute) was employed to estimate the annual percent change (APC) in AAMR with 95% CIs.^[12] This method detects significant changes by fitting log-linear regression models during periods of temporal variation, revealing increases or decreases in APCs based on the deviation of the slope from zero: with significance set at P < .05, using 2-tailed t-tests. Data suppressed in the CDC WONDER database for confidentiality were excluded from the analysis.

3. Results

From 1999 to 2023, a total of 20,108,555 deaths in which both cancer and sepsis were listed on the same death certificate among adults aged 25 and older. Throughout the study period, males had higher AAMRs than females. Among racial and ethnic groups, Black or African American individuals consistently showed the highest AAMRs. Nonmetropolitan areas experienced greater mortality rates than metropolitan areas. Regionally, the Southern U.S. reported the highest AAMRs, followed by the Midwest, Northeast, and West, respectively (Table 1)

Table 1.

Demographic characteristics of cancer and sepsis-related mortality in the U.S. from 1999 to 2023.

Outcome		Deaths	Population	AAMR (95% CI)	AAMR 1999 (95% CI)	AAMR 2023 (95% CI)	AAPC (95% CI)
Variable
Overall		20,108,555	5,163,131,262	362.8 (362.0–363.64)	417.6 (416.6–418.5)	322 (321.6–322.6)	−1.08* (−1.33–−0.84)
Sex	Female	9,700,202	2,672,090,011	309.3 (308.3–310.3)	348.6 (347.5–349.8)	279.1 (278.2–279.9)	−0.96* (−1.17–−0.75)
	Male	10,408,353	2,491,041,251	439.8 (438.4–441.2)	524.7 (523–526)	379.1 (378–380)	−1.32* (−1.59–−1.05)
Race/ Ethnicity	American Indian or Alaska Native	2367	37,910,158	337.7 (326.7–348.7)	342.4 (328.4–356.4)	303.4 (295.1–311.8)	−0.50 (−1.58–0.58)
	NH White	15,829,796	3,528,912,956	365.8 (364.9–366.7)	410.8 (409.7–411.8)	335.1 (334.3–335.9)	−0.86* (−1.099–−0.62)
	NH Black or African American	2,409,595	602,204,655	476.7 (464.7–470.8)	582.8 (279–586.7)	386.2 (383.8–388.6)	−1.68* (−1.96–−1.40)
	Hispanic or Latino	1,209,441	702,028,728	269.9 (267.3–272.6)	305.1 (301.3–308.9)	241.7 (239.9–243.6)	−0.99* (−1.23–−0.75)
	Asian or Pacific Islander	496,483	281,715,534	223.9 (220.5–227.2)	261.6 (256.5–266.6)	194.8 (192.6–197.1)	−1.20* (−1.50–−0.91)
Census Region	Northeast	3,865,907	948,225,258	359.3 (357.5–361.1)	429.6 (427.5–431.7)	301.7 (300.2–303.3)	−1.47* (−1.73–−1.21)
	Midwest	4,597,130	1,110,999,093	371.1 (369.4–372.8)	420.6 (418.7–422.6)	334.9 (334.4–336.4)	−0.94* (−1.19–−0.69)
	South	7,695,533	1,916,380,109	378.0 (376.7–379.4)	429.9 (428.3–431.5)	339 (337.9–340.1)	−1.015* (−1.28–−0.74)
	West	3,949,985	1,187,526,802	330.5 (328.9–332.2)	379.8 (377.8–381.8)	279 (295.7–298.4)	−1.01* (−1.26–−0.77)
Urbanization†	Metropolitan	14,117,743	3,795,213,822	361.8 (360.9–362.7)	417.7 (416.6–418.8)	328.5 (327.7–329.3)	−1.35* (−1.44–−1.25)
	Nonmetropolitan	3,243,240	678,634,169	390.6 (388.6–392.6)	417.7 (415.5–419.9)	384.2 (382.4–386.1)	−0.59* (−0.69–−0.49)

AAMR = age-adjusted mortality rate, AAPC = average annual percent change, CI = confidence interval, NH = non-Hispanic, U.S. = United States.

^*Statistically significant at P < .05.

^†Urbanization analyses available through 2020 due to data completeness.

3.1. Annual trends for cancer and sepsis mortality

During the study period from 1999 to 2023, there was a significant decline in the average annual percentage change (AAPC: −1.09%, 95% CI: −1.33–−0.84).

The AAMR for cancer and sepsis mortality declined over the study period, decreasing from 417.6 in 1999 (95% CI: 416.6–418.5) to 322.0 in 2023 (95% CI: 321.3–322.7). This overall decline included a steady reduction from 1999 to 2018 (APC: −1.32%, 95% CI: −1.37–−1.27), a temporary increase between 2018 and 2021 (APC: 2.17%, 95% CI: 0.33–4.04), and a further decline from 2021 to 2023 (APC: −3.65%, 95% CI: −5.18–−2.10) (Fig. 1).

Figure 1.

AAMR trends for concomitant cancer–sepsis mortality, defined as deaths with both cancer and sepsis listed on the death certificate, among U.S. adults aged ≥ 25 years, 1999 to 2023. AAMR = age-adjusted mortality rate, APC = annual percent change, CI = confidence interval, U.S. = United States.

3.2. Trends for cancer and sepsis mortality, stratified by demographics

3.2.1. Sex-wise analysis

Throughout the study period from 1999 to 2023, the AAMRs for males consistently stayed higher than those for females.

Throughout the study period, the AAPC declined for both sexes, with −0.96% (95% CI: −1.18–−0.75) for females and −1.33% (95% CI: −1.60–−1.05) for males. In 1999, the AAMR for males was 524.7 (95% CI: 523.0–526.4), significantly higher than the rate for females, which was 348.6 (95% CI: 347.5–349.8). By 2023, both rates had decreased, with males at 379.1 (95% CI: 378.0–380.2) and females at 279.1 (95% CI: 278.2–279.9), showing an overall decline in mortality for both sexes over the study period. From 1999 to 2018, females saw a significant decline in AAMR (APC: −1.21%, 95% CI: −1.27–−1.16). This was followed by a slight yet statistically significant increase from 2018 to 2021 (APC: 1.87%, 95% CI: 0.31–3.45). Between 2021 and 2023, the trend shifted again, showing a notable decrease in mortality (APC: −2.77%, 95% CI: −4.24–−1.27). Detailed annual counts and mortality rates stratified by sex are provided in Table S1.

Similarly, males exhibited a sustained decline in mortality from 1999 to 2018 (APC: −1.59%, 95% CI:−1.64–−1.53), followed by an upward trend between 2018 and 2021 (APC: 2.35%, 95% CI: 0.37–4.37). From 2021 to 2023, mortality rates decreased again (APC: −4.22%, 95% CI:−6.10–−2.32), although the CI suggests this reduction may not be statistically meaningful (Fig. 2).

Figure 2.

AAMR trends for concomitant cancer–sepsis mortality, defined as deaths with both cancer and sepsis listed on the death certificate, across sexes among U.S. adults aged ≥ 25 years. AAMR = age-adjusted mortality rate, APC = annual percent change, CI = confidence interval, U.S. = United States.

3.2.2. Age group-wise analysis

During the study period from 1999 to 2023, the AAMRs showed a decreasing trend across most age groups. Among young adults aged 25 to 34 years, a nonsignificant decline was observed (AAPC: −0.39%, 95% CI: −1.46–0.68). In contrast, early middle-aged adults aged 35 to 44 years experienced a statistically significant decrease (AAPC: −1.04%, 95% CI: −1.69–−0.39), as did middle-aged adults aged 45 to 54 years (AAPC: −1.35%, 95% CI: −1.62–−1.07) and late middle-aged adults aged 55 to 64 years (AAPC: −1.30%, 95% CI: −1.52–−1.07). A decline was also seen among older adults aged 75 to 84 years, with a significant decrease (AAPC: −1.07%, 95% CI: −1.24–−0.89). For young-old adults aged 65 to 74 years, a slight increase was noted (AAPC: −1.51%, 95% CI: −1.81–−1.20), though it was not statistically significant. Meanwhile, in the oldest age group (85 years and older), the AAMR remained relatively stable over time with a minimal, nonsignificant decline (AAPC: −0.29%, 95% CI: −0.58–0.00). Complete age-stratified mortality counts and rates are shown in Table S2.

Among individuals aged 25 to 34 years, the AAMR declined from 1999 to 2006 (APC: −1.96%, 95% CI: −2.89–−1.02), followed by relative stability from 2006 to 2018 (APC: −0.04%, 95% CI: −0.56–0.49). Rates then increased between 2018 and 2021 (APC: 5.93%, 95% CI: −1.70–14.15) and decreased again from 2021 to 2023 (APC: −6.03%, 95% CI: −12.56–1.00), though neither change was statistically significant.

Among individuals aged 35 to 44 years, the AAMR declined markedly from 1999 to 2010 (APC: −2.23%, 95% CI: −2.49–−1.97), followed by a slower downward trend from 2010 to 2018 (APC: −0.64%, 95% CI: −1.26–−0.02). Mortality subsequently increased between 2018 and 2021 (APC: 6.65%, 95% CI: 1.88–11.63) before declining sharply from 2021 to 2023 (APC: −6.98%, 95% CI: −11.03–−2.75).

Among individuals aged 45 to 54 years, the AAMR decreased steadily from 1999 to 2015 (APC: −1.11%, 95% CI: −1.17–−1.06) and more rapidly from 2015 to 2018 (APC: −3.61%, 95% CI: −4.99–−2.21). This was followed by an upward shift between 2018 and 2021 (APC: 3.11%, 95% CI: 1.44–4.81) and continued elevation through 2021 to 2023 (APC: −6.18%, 95% CI: −7.61–−4.73).

Among individuals aged 55 to 64 years, mortality rose during 1999 to 2007 (APC: −2.09%, 95% CI: −2.27–−1.90), then declined gradually from 2007 to 2018 (APC: −0.77%, 95% CI: −0.89–−0.65). Between 2018 and 2021, the trend reversed with a modest increase (APC: 1.65%, 95% CI: 0.12–3.21), followed by a subsequent decline from 2021 to 2023 (APC: −5.30%, 95% CI: −6.79–−3.79).

Among individuals aged 65 to 74 years, the AAMR declined steadily from 1999 to 2018 (APC: −1.86%, 95% CI: −1.93–−1.79), followed by a moderate increase between 2018 and 2021 (APC: 2.37%, 95% CI: 0.15–4.65). The trend then reversed, showing a marked decline from 2021 to 2023 (APC: −3.86%, 95% CI: −5.90–−1.78).

Among those aged 75 to 84 years, mortality declined modestly between 1999 and 2007 (APC: −0.81%, 95% CI: −0.94–−0.67) and more sharply from 2007 to 2018 (APC: −1.51%, 95% CI: −1.60–−1.41). Between 2018 and 2021, this downward trajectory was interrupted by an increase in AAMR (APC: 1.48%, 95% CI: 0.25–2.73), followed by a subsequent decline from 2021 to 2023 (APC: −3.41%, 95% CI: −4.53–−2.28).

Among individuals aged 85 years and older, the AAMR decreased from 1999 to 2018 (APC: −0.75%, 95% CI: −0.94–−0.57), after which mortality rose modestly during 2018 to 2023 (APC: 1.50%, 95% CI: 0.19–2.84) (Fig. 3).

Figure 3.

AAMR trends for concomitant cancer–sepsis mortality, defined as deaths with both cancer and sepsis listed on the death certificate, across different age groups, among U.S. adults aged ≥ 25 years, 1999 to 2023. AAMR = age-adjusted mortality rate, U.S. = United States.

3.2.3. Race-wise analysis

Across the study period from 1999 to 2023, Black or African American individuals consistently exhibited the highest AAMRs, followed by Non-Hispanic White, Non-Hispanic American Indian or Alaska Native, Hispanic or Latino, and Non-Hispanic Asian or Pacific Islander populations. Detailed race-specific mortality counts and rates are presented in Table S3.

From 1999 to 2023, a statistically significant decline in AAPCs was observed across all racial and ethnic groups. The largest reduction was seen among Black or African American individuals (AAPC: −1.68%, 95% CI: −1.96–−1.40), followed by Asian or Pacific Islander (AAPC: −1.21%, 95% CI: −1.50–−0.91), Hispanic or Latino (AAPC: −1.00%, 95% CI: −1.24–−0.76), and White individuals (AAPC: −0.86%, 95% CI: −1.10–−0.63). Although a decrease was also noted among American Indian or Alaska Native individuals, it was not statistically significant (AAPC: −0.50%, 95% CI: −1.58–0.58).

Among American Indian or Alaska Native individuals, the AAMR remained relatively stable from 1999 to 2015 (APC: 0.03%, 95% CI: −0.26–0.33), followed by a slight, nonsignificant decline between 2015 and 2018 (APC: −3.60%, 95% CI: −9.38–2.55) and a modest, nonsignificant increase from 2018 to 2021 (APC: 5.32%, 95% CI: −0.69–11.69). From 2021 to 2023, mortality declined substantially (APC: −8.26%, 95% CI: −13.53–−2.67).

Among Asian or Pacific Islander individuals, the AAMR decreased steadily between 1999 and 2018 (APC: −1.43%, 95% CI: −1.51–−1.34), then rose from 2018 to 2021 (APC: 2.63%, 95% CI: 0.48–4.83), before declining again from 2021 to 2023 (APC: −4.71%, 95% CI: −6.65–−2.74).

Among Black or African American individuals, the AAMR declined from 1999 to 2018 (APC: −2.05%, 95% CI: −2.11–−1.99), followed by an increase between 2018 and 2021 (APC: 3.22%, 95% CI: 1.18–5.29), before reversing sharply from 2021 to 2023 (APC: −5.31%, 95% CI: −7.21–−3.37).

Among White individuals, mortality decreased steadily from 1999 to 2018 (APC: −1.11%, 95% CI: −1.16–−1.06), rose between 2018 and 2021 (APC: 1.98%, 95% CI: 0.27–3.72), and then declined again from 2021 to 2023 (APC: −2.68%, 95% CI: −4.29–−1.04).

Among Hispanic or Latino individuals, the AAMR declined from 1999 to 2018 (APC: −1.24%, 95% CI: −1.30–−1.17), increased between 2018 and 2021 (APC: 4.93%, 95% CI: 3.16–6.74), and subsequently fell sharply from 2021 to 2023 (APC: −7.15%, 95% CI: −8.68–−5.60) (Fig. 4).

Figure 4.

AAMR trends for concomitant cancer–sepsis mortality, defined as deaths with both cancer and sepsis listed on the death certificate, across different race groups, among U.S. adults aged ≥ 25 years, 1999 to 2023. AAMR = age-adjusted mortality rate, APC = annual percent change, CI = confidence interval, U.S. = United States.

3.3. Trends for cancer and sepsis mortality, stratified by region

3.3.1. Urbanization-wise analysis

Throughout the study duration, AAMRs remained consistently higher in nonmetropolitan regions than in metropolitan areas. Annual mortality counts and rates by urbanization category are provided in Table S4.

Between 1999 and 2020, both metropolitan and nonmetropolitan areas demonstrated a decline in AAMR, with a more substantial decrease observed in metropolitan areas (AAPC: −1.35%, 95% CI: −1.45–−1.26) compared to nonmetropolitan areas (AAPC: −0.60%, 95% CI: −0.70–−0.50).

In metropolitan areas, the AAMR showed a consistent and statistically significant decline over the entire study period from 1999 to 2020, with an APC of−1.35% (95% CI: −1.45–−1.26).

In nonmetropolitan areas, the AAMR also declined steadily from 1999 to 2020, with an APC of−0.60% (95% CI: −0.70–−0.50), reflecting a modest but statistically significant reduction over the study period (Fig. 5).

Figure 5.

AAMR trends for concomitant cancer–sepsis mortality, defined as deaths with both cancer and sepsis listed on the death certificate, across urbanization categories, among U.S. adults aged ≥ 25 years, 1999 to 2023. AAMR = age-adjusted mortality rate, APC = annual percent change, CI = confidence interval, U.S. = United States.

3.3.2. Census-wise

Between 1999 and 2023, all 4 U.S. census regions demonstrated a statistically significant decline in AAMRs. The most pronounced decrease was observed in the Northeast (AAPC: −1.48%, 95% CI: −1.73–−1.22), followed by the Midwest (AAPC: −0.94%, 95% CI: −1.19–−0.69), the South (AAPC: −1.02%, 95% CI: −1.29–−0.74), and the West (AAPC: −1.02%, 95% CI: −1.26–−0.77). Region-specific annual mortality counts and rates are shown in Table S5.

From 1999 to 2018, the AAMR in the Northeast region declined steadily (APC: −1.62%, 95% CI: −1.67–−1.57), followed by a modest increase between 2018 and 2021 (APC: 0.51%, 95% CI: −1.31–2.36), before dropping sharply from 2021 to 2023 (APC: −3.05%, 95% CI: −4.85–−1.21).

In the Midwest, mortality decreased consistently from 1999 to 2018 (APC: −1.13%, 95% CI: −1.18–−1.08), rose during 2018 to 2021 (APC: 1.97%, 95% CI: 0.17–3.81), and then declined again from 2021 to 2023 (APC: −3.44%, 95% CI: −5.16–−1.68).

In the South, the AAMR declined steadily from 1999 to 2018 (APC: −1.27%, 95% CI: −1.32–−1.21), followed by an increase between 2018 and 2021 (APC: 2.81%, 95% CI: 0.83–4.83), and then a marked decline from 2021 to 2023 (APC: −4.21%, 95% CI: −6.04–−2.34).

In the West, mortality rates decreased from 1999 to 2018 (APC: −1.31%, 95% CI: −1.37–−1.26), rose during 2018 to 2021 (APC: 2.41%, 95% CI: 0.63–4.22), and subsequently fell again between 2021 and 2023 (APC: −3.25%, 95% CI: −4.90––1.56) (Fig. 6).

Figure 6.

AAMR trends for concomitant cancer–sepsis mortality, defined as deaths with both cancer and sepsis listed on the death certificate, across census categories, among U.S. adults aged ≥ 25 years, 1999 to 2023. AAMR = age-adjusted mortality rate, APC = annual percent change, CI = confidence interval, U.S. = United States.

3.3.3. Place of death

Out of a total of 20,108,555 deaths, 38.40% occurred in inpatient medical facilities, 34.75% at home, 11.90% in nursing homes or long-term care facilities, 8.05% in hospice facilities, 4.05% in other locations, 2.55% in outpatient or emergency departments, 0.15% were dead on arrival, and 0.05% had an unknown place of death (Fig. S1).

4. Discussion

The U.S.’ 2-decade trend in sepsis and cancer mortality shows a general improvement amid intermittent significant setbacks and variation among subgroups. The overall decline aligns with anecdotal evidence of advances in cancer treatment, sepsis recognition, and critical care management in population studies. However, the temporary peak in the mid-2010s reflects global reports of stalled progress caused by new infections, antimicrobial resistance, and health disparities.^[13,14]

Mechanistic understanding highlights the complex crosstalk between sepsis and cancer. Malignancy subjects patients to immune dysregulation, either through tumor immunosuppression or secondary effects of chemotherapy and radiotherapy.^[15,16] Sepsis triggers systemic inflammation and organ injury, increasing the risk of death in previously compromised cancer patients. Sandwiched between dysregulated host responses, therapeutic toxicities, and comorbidities, this dual burden represents one of the most dangerous intersections in medicine today.^[17]

Sex differences characterize sepsis and cancer mortality. Consistent with European and American historical data, men generally fare worse than women.^[18,19] Sex hormone regulation of immunity also plays a role, with estrogen shown to enhance innate and adaptive immunity, while testosterone has immunosuppressive effects.^[20] Behavioral factors also contribute; men are more likely to delay medical care, present with disease already advanced, and have higher exposures to alcohol and tobacco, which increase the risk and severity of sepsis.^[21,22] The narrowing of this gap later in life is partly due to higher rates of cancer screening among men and more aggressive sepsis treatment in this group.^[23] Conversely, the brief peak in both sexes before the pandemic aligns with evidence linking peaks in COVID-19 cases, hospitalization burden, and reduced elective cancer treatments to increased susceptibility.^[24,25]

Age stratification showed the expected pattern of increasing mortality with age, but the rate and direction of change varied greatly between groups. Teenagers experienced a moderate decrease in mortality, due to reports of earlier diagnosis of solid tumors and hematologic tumors, improved infection prevention, and aggressive intensive care treatment.^[26] In contrast, middle-aged adults saw a significant decline until the late 2010s, when mortality began to rise again. This peak is likely linked to the growing recognition of the impact of comorbid metabolic diseases, excess weight, opioid-induced immunosuppression, and disruptions in preventive care during public health emergencies.^[4]

The oldest generation bears most of the costs of mortality. Despite overall decreases, individuals aged 75 and older experience high mortality rates, consistent with earlier reports that frailty, polypharmacy, and reduced physiological reserve limit the benefits of improved oncologic and intensive care.^[27] Older patients are also less likely to receive aggressive treatment for sepsis or curative cancer therapy due to concerns about therapy intolerance and comorbidities.^[28] The consistent mortality rate in the oldest age group reflects biological vulnerability, as well as the underuse of intensive care unit-based therapies and a growing focus on comfort-oriented care at the end of life.^[29]

Differences by race and ethnicity are among the strongest trends in sepsis and cancer mortality. As in previous years, Black or African American groups experienced the highest mortality rates during the study period.^[30] Several factors contribute to this ongoing disparity: higher baseline rates of comorbidities such as diabetes and hypertension,^[31] delays in cancer treatment and screening, and structural barriers to early detection of sepsis and access to critical care resources.^[32] Notably, although overall decreases were observed, the short-term acceleration in mortality during the late 2010s was more pronounced in Black populations, likely due to pandemic-related disruptions affecting cancer therapy and access to critical care. Conversely, Asian and Pacific Islander and Hispanic groups, with lower baseline rates, reversed these trends over the same period. These findings align with reports indicating that factors such as immigration status, language barriers, and low healthcare utilization increased delays in treatment.^[33] Mortality trends among White populations were inconclusive but transient, consistent with system strain. The relative stability seen in American Indian or Alaska Native groups reflects challenges related to underfunded chronic care and geographic barriers to oncology and sepsis services.^[34]

Urbanization also influenced mortality patterns. Higher rural mortality is associated with issues observed across rural America, such as fewer oncologists to see patients, fewer critical care beds per capita, and longer transport times to higher-level facilities.^[35] With increased treatment of sepsis and cancer, rural hospitals lack specialist services and sepsis response orders, leading to longer delays in reducing mortality. Urban areas show greater decreases in mortality, partly due to better access to tertiary cancer centers and the adoption of sepsis bundle standards.^[36] These gaps suggest an overlap between geographic and socioeconomic disparities in outcomes.

Regional stratification confirms this. The South also had consistently worse mortality, as expected from the dominant literature linking this region to higher poverty, increased rates of chronic disease, and lower insurance coverage.^[37] The Midwest experienced similar but less severe trends, consistent with comparable socioeconomic stress across regions but greater network interconnectedness of cancer networks in some states.^[38] The Northeast saw the largest declines, as expected from earlier adoption of comprehensive cancer centers, public health interventions, and higher baseline healthcare spending.^[39] The West showed remarkable improvement, possibly due to immigration-related barriers to treatment causing a short-term decline during an epidemic year.^[40]

Lastly, understanding death patterns is essential for developing care strategies. High hospital mortality reflects the severity of cancer-related sepsis and the level of care required. Excess deaths at home indicate cultural and institutional denial of the need for urgent hospitalization. Increased deaths in hospices and nursing homes highlight broader trends toward integrating palliative care. Both situations illustrate how improvements in survival are primarily offset by the focus on end-of-life care, which significantly influences cancer–sepsis outcomes.

4.1. Limitations

Certain limitations of this study must be acknowledged based on existing literature. National death certificates are vulnerable to misclassification bias because sepsis often goes uncoded or is undercoded when it occurs as a cancer complication. Second, residual confounding by socioeconomic status, access to care, and treatment variability could not be controlled for, even when stratified by sex, race, age, urbanization, and geographic location. Third, excluding clinical variables such as cancer stage, treatment, or infection site reduces the interpretability of disease-specific mortality patterns. Additionally, temporal patterns during the COVID-19 period will inevitably be affected by pandemic-related disruptions to healthcare delivery rather than by genuine epidemiologic change. Despite these biases, the country-representative population database strongly suggests long-term mortality trends, with increasing generalizability to diverse U.S. subpopulations. In addition, we were unable to examine sex-by-race/ethnicity interactions, as such cross-classified stratified analyses were not prespecified and would require additional modeling beyond the scope of the present study.

4.2. Conclusion

Concomitant cancer–sepsis mortality, defined as deaths in which both cancer and sepsis were recorded on the death certificate, has declined overall in the U.S. over the past 2 decades, but marked disparities persist by sex, race/ethnicity, urbanization, and census region. Men, Black individuals, residents of rural areas, and people in the South continue to experience higher mortality, likely due to ongoing issues with access to care, comorbidities, and healthcare system structures. While some of the slower decline in deaths can be attributed to improvements in the quality of sepsis and cancer care, addressing structural disparities through targeted interventions is essential to further reduce mortality and achieve equitable outcomes across all population groups.

Acknowledgments

No third-party services or individuals contributed to this manuscript who are not listed as authors. No AI software was used in the preparation of this manuscript.

Author contributions

Conceptualization: Iffat Ambreen Magsi.

Data curation: Iffat Ambreen Magsi, Ahmad Kamal, Rumaisa Kunwar.

Formal analysis: Iffat Ambreen Magsi, Muhammad Suhaib Hanif, Ahmad Kamal, Mustafa Amin, Rumaisa Kunwar.

Funding acquisition: Iffat Ambreen Magsi, Muhammad Suhaib Hanif, Maira Hamid, Ahmad Kamal, Mustafa Amin.

Investigation: Iffat Ambreen Magsi, Muhammad Suhaib Hanif, Maira Hamid, Ali Msheik, Ahmad Kamal, Pinkey Kumari, Mustafa Amin, Rumaisa Kunwar.

Methodology: Iffat Ambreen Magsi, Maira Hamid, Ali Msheik, Pinkey Kumari, Mustafa Amin, Rumaisa Kunwar.

Project administration: Iffat Ambreen Magsi, Muhammad Suhaib Hanif, Ehsanullah Alokozay, Maira Hamid, Ali Msheik, Pinkey Kumari, Mustafa Amin, Rumaisa Kunwar.

Resources: Iffat Ambreen Magsi, Muhammad Suhaib Hanif, Maira Hamid, Ali Msheik, Ahmad Kamal, Pinkey Kumari, Mustafa Amin, Rumaisa Kunwar.

Software: Iffat Ambreen Magsi, Muhammad Suhaib Hanif, Ehsanullah Alokozay, Maira Hamid, Ahmad Kamal, Pinkey Kumari, Mustafa Amin, Rumaisa Kunwar, Aman Advani.

Supervision: Iffat Ambreen Magsi, Muhammad Suhaib Hanif, Ehsanullah Alokozay, Maira Hamid, Ali Msheik, Ahmad Kamal, Pinkey Kumari, Waleed Tariq, Aman Advani.

Validation: Muhammad Suhaib Hanif, Ehsanullah Alokozay, Pinkey Kumari, Waleed Tariq, Aman Advani.

Visualization: Ehsanullah Alokozay, Ali Msheik, Ahmad Kamal, Waleed Tariq, Aman Advani.

Writing—original draft: Ehsanullah Alokozay, Ali Msheik, Ahmad Kamal, Waleed Tariq.

Writing—review & editing: Ehsanullah Alokozay, Waleed Tariq.