Background: The risk of hyperkalemia is elevated in chronic kidney disease (CKD); however, the initial and recurrent risk among older individuals is less clear. Objectives: We set out to examine the initial and 1-year recurrent risk of hyperkalemia by level of kidney function (estimated glomerular filtration rate, eGFR) in older adults (≥66 years old). Design: Population-based, retrospective cohort study Settings: Ontario, Canada Participants: 905 167 individuals (≥66 years old) from 2008 to 2015. Measurements: Serum potassium values Methods: Individuals were stratified by eGFR (≥90, 60-89, 30-59, 15-29 mL/min/1.73 m2) and examined for the risk of incident hyperkalemia (K ≥ 5.5 mEq/L) using adjusted Cox proportional hazards models. The 1-year risk of recurrent hyperkalemia was examined using multivariable Andersen-Gill models. Results: Among a population of 905 167 individuals (15% eGFR ≥ 90, 58% eGFR 60-89, 25% eGFR 30-59, 3% eGFR 15-29) with a potassium measurement, there were a total of 18 979 (2.1%) individuals with hyperkalemia identified. The event rate (per 1000 person-years) and adjusted hazard ratio (HR) of hyperkalemia was inversely associated with eGFR (mL/min; eGFR >90 mL/min: 8.8, referent, 60-89 mL/min: 11.8 HR 1.41; eGFR 30-59: 39.8, HR 4.37; eGFR 15-29: 133.6, 13.65) and with an increasing urine albumin-to-creatinine ratio (ACR, mg/mmol; ACR< 3: 14, referent, ACR 3-30: 35.1, HR 1.98; ACR >30: 93.7, 4.71). The 1-year event rate and adjusted risk of recurrent hyperkalemia was similarly inversely associated with eGFR (eGFR ≥ 90: 10.1, referent, eGFR 60-89: 14.4, HR 1.47; eGFR 30-59: 54.8, HR 4.90; eGFR 15-29: 208.0, HR 12.98). Among individuals with a baseline eGFR of 30 to 59 and 15 to 29, 0.9 and 3.8% had greater than 2 hyperkalemia events. The relative risk of initial and recurrent hyperkalemia was marginally higher with RAAS blockade. Roughly 1 in 4 individuals with hyperkalemia required hospitalization the day of or within 30 days after their hyperkalemia event. Limitations: Limited to individuals aged 66 years and above. Conclusions: Patients with low eGFR are at a high risk of initial and recurrent hyperkalemia. Trial registration: N/A