Loop Diuretic Use in the Months and Years Preceding a Heart Failure Diagnosis: A Case-Control Study

Authors

David Knorek, Geisinger Health System, Danville, PA
Steven Steinhubl, Scripps Translational Science Institute, La Jolla, CA
Christopher deFilippi, Inova Heart and Vascular Institute, Falls Church, VAFollow
Kenney Ng, IBM Research, Cambridge, MAFollow
Roy Byrd, IBM Research, Cambridge, MAFollow
Zahar Daar, Geisinger Health System, Danville, PAFollow
Walter Stewart, Sutter Health, Walnut Creek, CA

Publication Date

8-10-2017

Keywords

cardiovascular disease, clinical practice patterns/guidelines, chronic disease

Abstract

Background: Loop diuretics are routinely prescribed for patients with heart failure (HF) to minimize symptoms due to fluid overload, but relatively little is known about prescribing patterns in individuals preceding a formal HF diagnosis. We want to understand the prescribing of loop diuretics in primary care patients without a HF diagnosis and to determine if their use signals an increased probability of a future HF diagnosis.

Methods: A nested case-control study of 1,288 new-onset (incident) HF cases and 10,319 matched controls was completed with electronic health record data extracted between 2001 and 2010 from primary care practices affiliated with the Geisinger Health System. New loop diuretic prescriptions, and associated orders < 36 months before and 12 months after an incident HF diagnosis in cases and a comparable time in controls, were evaluated.

Results: Loop diuretic use was significantly greater (11.3% vs 3.5%, P < 0.001) in future HF cases compared with controls 2 to 3 years preceding a HF diagnosis. Their use progressively increased in cases such that by 3 to 5 months preceding a HF diagnosis, 27.1% of cases versus 7.5% of controls (P < 0.001) had been prescribed a loop diuretic. Edema (ICD-9 code 782.3) and essential hypertension (ICD-9 code 401.x) were the most commonly associated diagnoses for starting a loop diuretic in both eventual HF cases as well as in controls. By multivariate analysis, being prescribed a loop diuretic was one of the strongest clinical predictors of an eventual diagnosis of HF (odds ratio: 2.44, 95% confidence interval: 2.10–2.83, P < 0.001).

Conclusion: In primary care, loop diuretic use is not uncommon in the months and years preceding a formal incident HF diagnosis. These results suggest there is substantial opportunity to improve the diagnosis and earlier treatment of individuals with HF.

Recommended Citation

Knorek D, Steinhubl S, deFilippi C, Ng K, Byrd R, Daar Z, Stewart W. Loop diuretic use in the months and years preceding a heart failure diagnosis: a case-control study. J Patient Cent Res Rev. 2017;4:163.