Continued Potassium Supplementation Use Following Loop Diuretic Discontinuation in Older Adults: An Evaluation of a "Prescribing Cascade Relic"

Background: Loop diuretics often prompt hypokalemia and subsequent potassium supplementation, constituting a prescribing cascade, whereby a new medication is initiated to manage a drug-induced adverse event. However, loop diuretics are often used sporadically and discontinued for periods of time, after which continued potassium supplementation can cause hyperkalemia. We sought to assess the occurrence and risk factors of this "prescribing cascade relic".

Objectives: To quantify the presence and identify the risk factors of continued potassium supplementation use following loop diuretic discontinuation among older adults experiencing a loop diuretic-potassium prescribing cascade.

Methods: We used Medicare Fee-For-Service data between 2011 and 2018 to identify adults aged ≥66 years, who initiated a loop diuretic, followed by potassium supplementation within 30 days, and who had no diagnosis of hypokalemia or use of potassium supplementation in the previous 360 days. To quantify the occurrence of this prescribing cascade relic, we identified the daily treatment status of each patient for 150 days following potassium supplementation initiation. We then calculated the proportion of person-days under potassium supplementation alone and percent of patients with >30 cumulative days under potassium supplementation alone in patients experiencing the loop diuretic-potassium prescribing cascade. We also assessed the risk factors associated with this prescribing cascade relic.

Results: We identified 284,369 loop diuretic initiators, of which 66,451 (23.4%) initiated potassium supplementation within 30 days, most often (n=62,929; 94.7%) on the same day as loop diuretic initiation. Overall, 20,445 (30.8%) experienced the prescribing cascade relic, of which 4,814 (7.2%) had a prescribing cascade relic for >30 cumulative days during the 150-day follow-up. In the 150 days following potassium supplementation initiation, 4.0% of person-days was for potassium supplementation alone. Older age (OR; 95% CI 1.30; 1.24-1.37), female (1.16; 1.12-1.21), higher daily dose of loop diuretics (1.17; 1.13-1.21), diagnosis with heart failure (1.23; 1.18-1.28), and higher Charlson Comorbidity Index score (1.19; 1.11-1.27) were the risk factors of this prescribing cascade relic.

Conclusions: Our findings suggested that among patients who experienced the prescribing cascade, nearly one—third continued to receive potassium supplementation following loop diuretic discontinuation. Clinicians should be aware of the potential implications of this prescribing cascade relic to avoid unnecessary harm (e.g., hyperkalemia, cardiac arrhythmia) to patients.