Hospital Heart Failure Case Volumes Predict Care Quality, Not Outcomes

January 31, 2018

DALLAS — Lower HF case volumes tended to predict less adherence to HF quality-of-care measures, especially use of device therapies, yet their association with 30-day and 6-month clinical outcomes was weak at best, a new analysis shows.[1]

For example, hospitals with the lowest case volume were less likely than highest-volume centers to discharge HF patients with atrial fibrillation (AF) on oral anticoagulants, to put in or prescribe an implantable defibrillator (ICD), or measure LV ejection fraction with echocardiography, but nevertheless they didn't have significantly higher mortality or readmissions at 30 days or 6 months.  

The findings, based on 125,595 patients admitted with HF at 342 hospitals participating in the Get With The Guidelines (GWTG) Heart Failure program, were published January 29 in Circulation, with lead author Dr Dharam J Kumbhani (University of Texas Southwestern Medical Center, Dallas).

Although some guidelines and reimbursement schemes continue to consider case volume to be an index of care quality that translates into improved outcomes, Kumbhani told theheart.org | Medscape Cardiology, the current analysis suggests that the relationship between volume and outcomes is more complex.

"The real emphasis should be on understanding what the things are that higher-volume hospitals do better" and then "implementing them in other hospitals even if they are lower volume," he said.

"This is not to say that there is absolutely zero relationship between higher volume and outcomes, but I think the relationship is very modest."

Even among the GWTG hospitals in the current analysis, "there were clearly certain process measures that higher-volume hospitals did better than lower-volume hospitals." And higher volume, Kumbhani observed, probably does correlate with better HF care quality for some interventions, such as some device therapies, PCI, and surgical procedures.

"Extremely Powerful"

The current study suggests that "among hospitals that have joined a quality-improvement initiative, volume doesn't seem to matter as much as it does in the absence of a quality-improvement initiative," Dr Karen Joynt Maddox (Washington University School of Medicine, St Louis, MO) said in an interview.

"That, I think, is extremely powerful. That tells me that quality-improvement initiatives may particularly help low-volume centers," said Joynt Maddox, who isn't connected with the analysis.

For LV assist device management or complex surgeries, for example, "there is no doubt that volume is important," she said. But "measuring volume is a surrogate for quality. And if we can measure quality, we should just measure quality, because I would rather go to a high-quality low-volume center than a low-quality high-volume center."

Joynt Maddox also noted that HF case volume in the current study did correlate with some quality measures, such as use of ICDs and cardiac resynchronization therapy (CRT). That's not surprising, but such treatments would take time to confer better outcomes.

"The expectation that those quality measures would translate into outcomes at 30 or 90 days, or even a year? That's not how they work," she said.

"Those process measures matter because they influence people's 2-, 5-, and 10-year survival." So quality measures can matter "even if you don't you don't see any difference in short-term outcomes."

GWTG centers in the current analysis covering 2005 to 2014 had a mean annual HF case volume of 89; the median was 77.3. Those with volumes in the lowest quartile (5 to 38 cases) compared with the highest quartile (123 to 457 cases) generally were less likely to meet care-quality measures.

Table 1. Odds Ratio for Meeting Quality Measures by HF Case Volume, First (Lowest) vs Fourth (Highest) Quartile

Care Quality Measure Odds Ratio (95% CI)a Q1 vs Q4 P Value
ACE inhibitors or ARBs at Discharge 0.67 (0.47–0.95) 0.025
β-Blocker at discharge 0.60 (0.43–0.83) 0.002
Aldosterone antagonist at discharge 0.81 (0.62–1.06) 0.13
Anticoagulation for AF 0.59 (0.47–0.74) <0.001
LV function measured 0.26 (0.17–0.41) <0.001
ICD implanted or prescribed 0.41 (0.28–0.58) <0.001
CRT device implanted or prescribed 0.42 (0.23–0.78) 0.006
Appropriate discharge instructions 0.58 (0.39–0.87) 0.008
ARB = angiotensin receptor blocker; Q1 = first quartile; Q4 = fourth quartile.

aAdjusted for sociodemographic and demographic features, medical history (anemia, ischemic disease, cerebrovascular events, diabetes, hyperlipidemia, hypertension, chronic obstructive pulmonary disease or asthma, peripheral vascular disease, and renal insufficiency), smoking, features at baseline (systolic blood pressure, heart rate, serum sodium, serum urea nitrogen, and ejection fraction), and hospital characteristics (region and teaching vs nonteaching hospital).

 

In an analysis treating case volume as a continuous variable rather than by quartiles, volume was slightly but significantly related to 6-month outcomes. The odds of 6-month readmission from any cause was 0.98 (95% CI, 0.97–1.00; P=0.025) and for 6-month mortality was 0.98 (95% CI, 0.97–0.99; P=0.001) for every increase of 50 HF cases per year.

But analysis of case volume by quartiles showed no such significant effect.

Table 2. Odds Ratio for Outcomes 1st (Lowest) vs 4th (Highest) Heart-Failure Case Volume Quartile

Outcome Odds Ratio (95% CI)a Q1 vs Q4 P Value
In-hospital mortality 1.15 (0.92–1.43) 0.23
30-d readmission 1.04 (0.94–1.15) 0.47
30-d mortality 1.05 (0.92–1.19) 0.45
6-mo readmission 1.04 (0.96–1.12) 0.38
6-mo mortality 1.06 (0.99–1.14) 0.11
Q1 = first quartile; Q4 = fourth quartile

a30-day and 6-month outcomes apply to patients discharged (hospital survivors). Adjusted for sociodemographic and demographic features, medical history (anemia, ischemic disease, cerebrovascular events, diabetes, hyperlipidemia, hypertension, chronic obstructive pulmonary disease or asthma, peripheral vascular disease, and renal insufficiency), smoking, features at baseline (systolic blood pressure, heart rate, serum sodium, serum urea nitrogen, and ejection fraction), and hospital characteristics (region and teaching vs nonteaching hospital).

 

"If you just focus on volume, you're missing the bigger picture: that you have the ability to deliver outstanding care, quality care, at low- and medium-volume hospitals," Kumbhani said. Once the differences between low- and high-volume centers are understood, you can then select the "highest-yield" care processes and improve them at the lower-volume centers.

Circulation is the flagship journal of the American Heart Association (AHA), which also launched the GWTG initiative. GWTG has been supported by Amgen Cardiovascular, Medtronic, GlaxoSmithKline, Ortho-McNeil, and the AHA Pharmaceutical Roundtable. Kumbhani reports receiving honoraria from the American College of Cardiology and consulting for Somahlution; disclosures for the other authors are in the report. Joynt Maddox reports that she has no relevant financial relationships.

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

Comments

3090D553-9492-4563-8681-AD288FA52ACE
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.

processing....