Among U.S. deaths with diabetes recorded as a non-underlying cause, the underlying-cause profile is shifting from coronary-dominated mortality toward broader multimorbidity, with declining circulatory disease shares and increasing shares for nervous system diseases, mental and behavioral disorders, and malignant neoplasms between 1999 and 2024.
Key Findings
Results
Deaths with diabetes recorded as a non-underlying cause (DM-nonUCD) increased substantially over the study period, occurring in 4,794,136 total deaths.
Annual DM-nonUCD counts increased from 141,371 in 1999 to 246,426 in 2024.
DM-nonUCD was defined as deaths with ICD-10 codes E10-E14 in multiple-cause fields, excluding deaths with diabetes as the underlying cause.
Data were drawn from CDC WONDER Multiple Cause of Death records spanning 1999 to 2024.
Both type 1 diabetes (E10) and type 2 diabetes (E11) subtypes were analyzed separately.
Results
Circulatory diseases were the leading underlying cause of DM-nonUCD deaths overall but declined as a proportional share over time.
Circulatory diseases accounted for 2,294,799 deaths (47.87% of total DM-nonUCD).
The circulatory disease share declined from 60.51% in 1999 to 44.48% in 2024.
The age-adjusted mortality rate (AAMR) for circulatory disease as an underlying cause showed a decreasing trend with an average annual percent change (AAPC) of -1.01% (95% CI, -1.21 to -0.88).
Malignant neoplasms were the second leading cause at 725,592 deaths (15.13%), and respiratory diseases third at 449,837 deaths (9.38%).
Results
Within circulatory causes, ischemic heart disease AAMR declined while hypertensive disease AAMR increased substantially.
Ischemic heart disease AAMR declined with an AAPC of -2.04% (95% CI, -2.23 to -1.92).
Hypertensive disease AAMR increased with an AAPC of 3.21% (95% CI, 2.94 to 3.56).
These opposing trends indicate a shift in the composition of circulatory disease deaths among DM-nonUCD decedents.
Analyses were performed using Joinpoint trend methodology.
Results
The proportional share of nervous system diseases as an underlying cause among DM-nonUCD deaths increased markedly over the study period.
The share of nervous system diseases increased from 2.67% in 1999 to 7.56% in 2024.
This represents nearly a threefold increase in proportional share across the 25-year study period.
Similar shifts were observed in both the type 1 (E10) and type 2 (E11) diabetes subtype analyses.
Results
Mental and behavioral disorders increased substantially as an underlying cause among DM-nonUCD deaths.
The share of mental and behavioral disorders increased from 1.05% in 1999 to 4.41% in 2024.
This represents more than a fourfold increase in proportional share over the study period.
Both E10-nonUCD and E11-nonUCD subtypes showed similar increasing trends for this cause category.
Results
Malignant neoplasms showed a modest increase as a proportional underlying cause among DM-nonUCD deaths.
The share of malignant neoplasms as an underlying cause increased from 14.01% to 16.30% over the study period.
Malignant neoplasms were the second leading underlying cause overall, accounting for 725,592 deaths (15.13%).
This trend was observed across the full 1999-2024 study window.
Results
Subtype analyses for type 1 and type 2 diabetes showed similar cause-distribution shifts, with declining circulatory shares and increasing nervous system and mental and behavioral shares.
Type 1 diabetes was identified using ICD-10 code E10 and type 2 diabetes using E11 in multiple-cause fields.
Both E10-nonUCD and E11-nonUCD showed declining proportions of circulatory diseases as underlying causes.
Increasing shares of nervous system diseases and mental and behavioral disorders were observed in both subtypes.
The parallel trends across subtypes suggest the cause-distribution shift is not specific to one diabetes type.
Conclusions
The authors concluded that DM-nonUCD represents a substantial proportion of deaths in which diabetes was recorded on the death certificate, and that the underlying-cause profile is shifting from coronary-dominated mortality toward broader multimorbidity.
The study used CDC WONDER Multiple Cause of Death data, which captures all conditions listed on the death certificate, not just the underlying cause.
The findings support 'surveillance and health-system planning that consider competing causes beyond ischemic heart disease.'
The shift toward multimorbidity reflects patterns including nervous system diseases, mental and behavioral disorders, and hypertensive disease.
The 25-year surveillance window (1999-2024) allowed detection of long-term trend changes using Joinpoint regression.
What This Means
This research examined over 4.7 million U.S. deaths between 1999 and 2024 in which diabetes was listed on the death certificate but was not identified as the primary (underlying) cause of death. The study found that while the number of such deaths grew from about 141,000 in 1999 to over 246,000 in 2024, the types of diseases listed as the primary cause of death changed significantly. Heart disease—particularly coronary artery disease—used to dominate as the main cause of death in people with diabetes, but its share dropped from about 61% to 44% over this period. Meanwhile, conditions like nervous system diseases (such as dementia), mental and behavioral disorders, high blood pressure complications, and cancer became increasingly common as the recorded primary causes of death.
These shifts suggest that people with diabetes are living longer and are now more likely to die from a wider range of conditions rather than primarily from heart attacks. This reflects a broader pattern of 'multimorbidity'—having multiple serious health conditions at the same time—which is increasingly common as the population ages and medical treatments for heart disease improve. Notably, high blood pressure-related deaths increased sharply among this group even as overall cardiovascular deaths declined, and nervous system diseases nearly tripled in their proportional share. These trends were seen in both type 1 and type 2 diabetes.
This research suggests that public health monitoring and healthcare planning for people with diabetes need to look beyond heart disease and account for this expanding range of serious conditions. Surveillance systems, clinical care models, and resource allocation should increasingly consider conditions like dementia, mental illness, and hypertension-related complications as important contributors to mortality in people with diabetes.
Xiao Y, Zhou Q, Yan X, Gan L, Chen T, Zhang K, et al.. (2026). Shifting competing underlying causes of death among U.S. decedents with diabetes recorded as a non-underlying cause, 1999-2024.. Frontiers in endocrinology. https://doi.org/10.3389/fendo.2026.1957392