Little is known about variation in patterns of recovery among patients discharged alive from hospitalizations for pneumonia.The aim of the is observational cohort study was to characterize the variation in patterns of hospital readmission and survival in the year after discharge for pneumonia in 3 different health systems. The 3 cohorts consisted of (1) the Health and Retirement Study participants enrolled in Fee-for-service Medicare (FFS), (2) Veterans Administration (VA) Healthcare system, and (3) Kaiser Permanente of Northern California (KPNC). The 365-day survival and re-hospitalizations were determined for each cohort. Multinomial logistic regression was used to identify potential contributors to the different patterns. We identified 2731, 23,536, and 39,147 hospitalizations for pneumonia in FFS Medicare, VA, and KPNC, respectively, of whom 88.1%, 92.8%, and 89.7% survived to hospital discharge. The median patient survived to 1 year and was rehospitalized twice in FFS (9.0%), once in VA (14.1%) and KPNC (9.1%). Of the patients who survived the hospitalization, 33.3% (FFS), 30.2% (VA), and 26.8% (KPNC) died during the subsequent year. Of those who survived, 29.8% (FFS), 35.9% (VA), and 46.1% (KPNC) were never rehospitalized. 11.9% (FFS), 11.9% (VA), and 11.7% (KPNC) had greater than 3 hospitalizations. Age, race, gender, comorbidity, ICU use, and hospital length stay collectively explained little (5-7%) of the variation in the recovery pattern. There is significant variation in the year after the hospitalization for pneumonia across individuals, but less so across health systems. There may be important opportunities to better classify these heterogeneous individual-level pathways.