Causes of In-hospital and Post discharge Mortality Among Patients Hospitalized with Laboratory-Confirmed Influenza, Influenza Hospitalization Surveillance Network, 2014–2015

BACKGROUND: Influenza results in an estimated 12,000–56,000 deaths annually in the USA. While in-hospital deaths are well characterized, less is known about deaths that occur after discharge among those hospitalized with influenza. METHODS: We identified patients hospitalized with laboratory-confirmed influenza who died during hospitalization or within 30 days after discharge during the 2014–2015 influenza season for 11 Influenza Hospitalization Surveillance Network sites. We matched cases to the National Center for Health Statistics Electronic Death Registration System and abstracted cause and location of death from death certificates. We compared clinical characteristics between those who died during hospitalization and those who died after hospital discharge using χ(2) tests. RESULTS: Among 795 patients with laboratory-confirmed influenza who died, 370 (47%) died during hospitalization, and 425 (53%) died within 30 days after discharge. Eighteen (2%) were 0–17 years and 652 (82%) were ≥65 years. Common causes of death listed in any position on the death certificate included influenza (35%), other respiratory causes (50%), cardiovascular disease (37%), and sepsis (15%). Among those who died after discharge, 207 (49%) died within 7 days, 86 (20%) within 8–14 days, and 132 (31%) within 15–30 days post discharge. Patients who died after discharge were more likely to be ≥65 years (88 vs. 74%) or admitted from a nursing home (48 vs. 36%), but were less likely to be admitted to an intensive care unit (30 vs. 68%) or receive a pneumonia diagnosis (46 vs. 62%) than patients who died during hospitalization (all P < 0.001). There were no significant differences in sex, race, underlying conditions, vaccination rates, or time from symptom onset to hospitalization. Patients who died in hospital were more likely to have influenza listed as a cause of death (55 vs. 21%, P < 0.01). CONCLUSION: Over half of deaths among patients hospitalized with laboratory-confirmed influenza occurred after discharge. Patients who died after discharge were older and less likely to have influenza listed as a cause of death. Deaths that occur after an influenza-related hospitalization represent an important and under-characterized contribution to the burden of seasonal influenza. DISCLOSURES: E. J. Anderson, AbbVie: Consultant, Consulting fee; NovaVax: Research Contractor, Research support; Regeneron: Research Contractor, Research grant; MedImmune: Research Contractor, Research grant and Research support. W. Schaffner, Pfizer: Scientific Advisor, Consulting fee; Merck: Scientific Advisor, Consulting fee; Novavax: Consultant, Consulting fee; Dynavax: Consultant, Consulting fee; Sanofi-pasteur: Consultant, Consulting fee; GSK: Consultant, Consulting fee; Seqirus: Consultant, Consulting fee