One-year incidence of newly diagnosed dementia after inpatient sepsis hospitalization

BACKGROUND: Sepsis survivors face increased long-term cognitive risk, but contemporary U.S. population-level estimates of claims-ascertained dementia remain limited. OBJECTIVE: To estimate 1-year incidence of newly documented claims-ascertained dementia after sepsis hospitalization and describe variation by age, sex, payer, and sepsis severity. METHODS: We conducted a retrospective cohort-study using a nationally representative U.S. claims database with linked medical, pharmacy, and enrollment data from 1 January 2020 through 31 March 2025. Adults aged 18 years or older with an index sepsis hospitalization, no prior dementia diagnosis, and complete 1-year post-discharge claims observability were included. Newly documented dementia was defined using a claims-based algorithm requiring atleast 2 outpatient claims separated by 30 days or atleast 1 inpatient claim. Exact Clopper-Pearson 95% confidence intervals (CIs) were calculated. RESULTS: Among 35,689 sepsis survivors (mean-age, 62.4 years; 51.5% male; 42.5% Medicare; 19.7% with severe sepsis or septic shock), 1,072 received a newly documented dementia diagnosis within 1 year (3.00%; 95% CI, 2.83%-3.19%). Incidence increased from 0.60% among patients aged 18-64 years to 2.83% among those aged 65-74 years and 8.24% among those aged 75 years or older. Median-time to diagnosis was 112 days (interquartile range, 39-223 days). Incidence was similar by sex and higher with severe sepsis (3.62% vs 2.85%). Sensitivity analyses yielded estimates of 2.19%-4.29%. CONCLUSIONS: Approximately 3% of insured adult sepsis survivors with complete 1-year claims observability received a new dementia diagnosis within 1-year, with risk concentrated among older adults. Older sepsis survivors may warrant closer cognitive follow-up, though these findings do not establish screening benefit or sepsis-related dementia risk.