Excess Polypharmacy a Risk in Heart Failure, but Is Illness the Whole Reason?

September 19, 2018

NASHVILLE, Tennessee — About one-fourth of 947 adult patients with heart failure were concurrently taking at least 10 different medications in an observational analysis of data from the United States covering 2003 to 2014.

Those "hyperpolypharmacy" (HPP) patients in the sample took an average of 12 different meds, whereas non-HPP patients, who were on fewer than 10 agents, took an average of 5.5.

"The most striking difference was in the noncardiovascular agents," Peter J. Kennel, MD, told theheart.org | Medscape Cardiology. Although both groups were taking a spectrum of cardiovascular (CV) meds, non-CV meds in the HPP group "accounted for a large proportion of what makes them hyperpolypharmacy patients."

The average daily non-CV medication count was 6.6 for HPP patients and 2.3 for the remainder, said Kennel, from Weill Cornell Medicine, New York City, here at the Heart Failure Society of America 22nd Annual Scientific Meeting.

In adjusted analysis, presumed predictors of adverse events related to HPP, such as older age and functional and cognitive impairment, did not correlate with number of meds, nor did type of third-party coverage or lack of coverage.

Predictors of HPP that emerged as significant included, not surprisingly, a greater number of comorbidities, but also lower attained level of education and household income, leading Kennel and his colleagues to conclude that nonmedical factors were contributing to HPP in this cohort.

The list of CV medications patients with heart failure might take is long, Kennel said, and they tend to have comorbidities, so they could be at special risk for HPP and its demonstrated hazards, which include increased risk for disability and hospitalization.

But that's speculative. The current analysis, based on data from HF patients in the community from the National Health and Nutrition Examination Survey (NHANES), did not actually demonstrate that HF patients in particular are harmed by polypharmacy, he said.

"Still, even after controlling for lifestyle factors, there's something about low education level and low income that drives people into hyperpolypharmacy, which is obviously dangerous," Kennel said. "We think it might have something to do with how people access care."

People with less education or income, he proposed, might not have consistent access to a primary care physician as a gateway to specialists. "They may not see a cardiologist, and they have less contact with outpatient doctors, so they seek their care more often in emergency rooms."

If they don't have a "medical home" for continuity of care, "they go to an emergency room and get meds thrown at them," he said. Although there are likely other reasons for their findings, "we think that's the driver."

Table. Differences Between HPP and Non-HPP Patients in a NHANES Sample of Adults With Heart Failure
Parameter HPP, n = 242 Non-HPP, n = 705 P Value
Female, % 54 48 .07
High-school education or less, % 93 8 .01
Mean comorbidities, n 5.9 4.3 <.001
Mean total meds, n 11.9 5.5 <.001
Mean HF medications, 2.8 1.9 <.001
Mean non-HF CV meds, n 2.5 1.3 <.001
Mean non-CV meds, n 6.6 2.3 <.001
HPP = hyperpolypharmacy, defined as at least 10 concurrent medications.


HPP patients took significantly more of each of the following: beta blockers, ACE inhibitors or angiotensin-receptor blockers, aldosterone antagonists, vasodilators, diuretics, lipid-modifying agents, antiplatelets, antiarrhythmics, calcium-channel blockers, antianginals in general, and blood-pressure-lowering meds in general.

Non-CV meds made up 55% of the total in HPP patients but only 42% in non-HPP patients. HPP patients, compared with non-HPP patients, took significantly more opioids, nonopioid analgesics, psychiatric meds (such as antidepressants), antidiabetics, acid blockers, agents for thyroid disorders, bronchodilators, and mineral or vitamin supplements.

"If they go to an emergency department over and over again, there should be a doctor who takes a close look at their meds and does a thorough reconciliation of medications and supplements currently being taken," he said.

Table. Significant Predictors of HPP in a NHANES Sample of Adults With Heart Failure*
Predictor Odds Ratio (95% Confidence Interval) P Value
Highest education: high school or below 1.74 (1.01–2.99) .04
Income <$20,000 1.70 (1.01–2.85) .04
Number of comorbidities (higher vs lower) 1.19 (1.12–1.27) <.001
Number of outpatient contacts in previous year    
  4–9 vs 0–3 2.20 (1.23–3.96) .01
  ≥10 vs 0–3 3.01 (1.73–5.21) <.001
Number of hospitalizations in previous year, ≥3 vs <3 1.70 (1.27–2.30) .001

*NHANES database adjusted for lifestyle and other factors; further adjusted for socioeconomic and demographic variables, comorbidities, conditions of aging, and healthcare utilization.


The analysis was supported by the National Institutes of Health. Kennel had no disclosures.

Heart Failure Society of America (HFSA) 22nd Annual Scientific Meeting: Abstract 080. Presented September 16, 2018.

Follow Steve Stiles on Twitter: @SteveStiles2. For more from theheart.org | Medscape Cardiology, follow us on Twitter and Facebook.


Comments on Medscape are moderated and should be professional in tone and on topic. You must declare any conflicts of interest related to your comments and responses. Please see our Commenting Guide for further information. We reserve the right to remove posts at our sole discretion.