Thread by @benharder

Two noteworthy findings by Wang @rkwadhera @DLBHATTMD in @JAMACardio:
1) Hospitals in the @usnews top 50 have better mortality rates than other hospitals in AMI, CABG & HF. Better HCAHPS, too. Not too surprising, but certainly reassuring.
2) But...
jamanetwork.com/journals/jamac…

1/

2) But those same top hospitals do no better than others on readmission.

Why? Well, we have to ask ourselves whether readmission is a valid indicator of quality. And if it is, is it valid in all patient cohorts, or only certain ones?
2/

As Wang @rkwadhera @DLBHATTMD understatedly write:

“The observed discrepancy between readmissions and other performance measures raises concern that readmissions may not be an adequate metric of hospital care quality.”
3/

They add: “This disconnect...highlights the ongoing uncertainty as to whether readmissions rates are an adequate surrogate for quality of care, particularly for cardiovascular conditions, such as HF.”
Indeed, we @usnews do not use readmission in our Heart Failure ratings. Why?
4/

We’ve found that readmission correlates weakly, if at all, with quality* in medical conditions such as HF and COPD.

In surgical episodes, by contrast, we find readmission consistently correlates with quality*.

(*By “quality,” I mean...
5/

*By “quality,” I mean the latent variable that determines the @usnews Procedures & Conditions ratings. See usnews.com/static/documen…, Tables 5, 7, 9, 13, 17, 19 & 21.

6/

Unlike readmission, mortality correlates with quality in every cohort we @usnews have examined, whether medical or surgical.

With the exception of AAA repair, we haven’t yet investigated endovascular procedures using these methods. So I can’t shed light on, for example, PCI.
7/

Those who attended my talk @USNHoT this month heard me discuss (my slide shown here) two candidate outcome measures we are investigating for use in the @usnews Cardiology & Heart Surgery rankings and other specialty rankings.

Neither candidate measure was readmission.

8/

Regarding CABG readmission, it’s worth noting the measure specifications & risk adjustment we use differ from @CMSGov specs & adjustment. While our measure correlates with CABG quality, I’m not shocked @rkwadhera @DLBHATTMD found the CMS measure doesn’t correlate as expected.

9/

I plan to speak more about @usnews’s use of outcome measures at the ACC Quality Summit #ACC19 in New Orleans on March 14 (4pm local).

Thank you @ACCinTouch @ACCCVQuality for inviting me to be part of an important panel.
10/

I also look forward to addressing, in my May 21 lecture at @RadiologyACR, the selection of outcome metrics used by @usnews. Measure selection is really crucial & differentiates valid ratings from invalid.

(Thank you @RichDuszak for inviting me to speak.)

#RSNA2018 #RSNA18
11/11

But wait, there’s more!

“Admission to a top-ranked hospital for neurology or neurosurgery was associated with lower 30-day risk-standardized mortality but higher readmission after ischemic stroke.”

Again, nonsurgical care (mostly) & readmits=fishy.
mdedge.com/clinicalneurol…
12/11