Association of Sepsis Survivor Subtypes With Long-Term Mortality and Disability After Discharge: A Retrospective...

Determine if previously described sepsis survivor subtypes can be applied outside of their derivation cohort using a parsimonious algorithm. Test the association between subtype and the primary outcome of 3-month mortality, and secondary outcomes of readmission, physical function, and health-related quality of life through 1 year of follow-up.

Retrospective cohort study.

Participants enrolled in the Crystalloid Liberal or Vasopressors Early Resuscitation in Sepsis (CLOVERS) trial, a multisite trial in the United States that enrolled patients with sepsis-induced hypotension.

All participants who were alive on day 28 after enrollment and had nonmissing data for outcome and subtype-defining variables (Charlson Comorbidity Index, length of stay, discharge destination). Participants were retrospectively assigned at time of discharge to one of five previously derived survivor subtypes: low risk, healthy with severe disease, multimorbidity, low functional status, and unhealthy baseline with severe disease.

None.

Of 1563 participants, 1368 were eligible and assigned a subtype. Three-month mortality was 13.1% and varied significantly between subtypes (5.1-45.5%; p < 0.001). In age-adjusted logistic regression, odds ratios for 3-month mortality were 11.1 in the low functional status and 9.7 in the unhealthy baseline with severe illness subtypes, compared with the low-risk subtype ( p < 0.001). Participant subtype was a significant predictor of 6- and 12-month EuroQol 5D five level score and limitations in activities of daily living, but not readmission.

Sepsis survivor subtypes that are readily identifiable at hospital discharge are significantly associated with mortality at 3 months, and patient-important outcomes through 12 months. Using subtypes to predict a patient's risk of adverse outcomes could aid the discharge planning and recovery process.