Bingo: “Judging medical quality from claims data is like judging a restaurant by looking at its grocery bill.”
Related posts:
- Where Do Claims Go & How Do I Get Paid?
- My take: ER naming rights, grading data, salary disparity
- Avandia and GSK: Trial data
- The physician who disputes second-hand smoke claims
- Are patients looking up quality data before a doctor or hospital visit?
- The data argues against an annual physical, but is it still needed?
- Ketek and shaky trial data
KevinMD.com on Facebook
{ 4 comments }
Certainly makes a nice sound bite, but shouldn’t it be “Judging medical quality from claims data is like judging a restaurant by looking at the bills the restaurant generates and gives to its customers.”
Maybe judging medical quality from claims data is more like judging a mechanic by the bills he generates for his customers.
For one thing, a consumer is more likely to have an ongoing relationship with one primary care mechanic whereas most people tend to frequent more than one restaurant.
And while you can’t judge the technical quality of the work performed by your mechanic merely by looking at the bill, you can certainly check to see if all the scheduled maintenance occurred at your 50,000 mile service visit.
Quality of service is subjective at best and can rarely be judged by administrative data, but the fact that service occurred in a timely fashion, or occurred at all, I don’t understand why that simple measurement of fact is so hard to swallow.
Claims data allows a simple measurement: were you responsible for person A achieving thing B; if so, did person A achieve thing B?
Anyone subjected to a report card of any kind should be given a fair way to dispute the data within of course, it’s deplorable that a physician can’t get through to a payor to correct a mistake, but the answer is not stopping all quality metric reporting.
Hey Jaz, how does claims data tell you about quality? I submit a hospital claim on a patient for UTI or DVT or soemthing and the guy was admitted for a week. So the conclusion is “Gee, Dr Mike is incompetent. It should only take a day or two to discharge such a patient”. Yeah, but what about the home health attendant that needed to be approved? What about the rehab issues, what about the fact that this guy refused to leave? What about having to wait a whole weekend just to get a doopler, V/Q scan, ECHO, etc”
Thats reality in the hospital. And claims data does not reflect that.
“What about having to wait a whole weekend just to get a doopler, V/Q scan, ECHO, etc”
i guess claims data aint so bad in estimating quality.
I was responding to the OP and was talking more about physician office measures, but hospital claims data can be used too, and in fact already is. Luckily, we already have oodles and oodles of quality data from hospitals, so there’s less of a need for administrative data to be used, but still…
I see you’re in NY, so using the NYSDOH SPARCS dataset we can gather a bunch of volume and mortality indicators for a range of conditions and procedures. We can figure out C-Section rates, average hospital charges, VBAC rates, average lengths of stay, all kinds of procedure volume by age and/or gender… a whole realm of indicators that consumers and purchasers find interesting and informative.
Once we have a hospital industry that uses interoperable health records and a functional health information network then we can apply the same kind of data gathering and analysis that gets done in every other modern industry, no-one doesn’t want to work from the chart – it’s just so darn inaccesible. Wherever it’s available it gets used, the HQA / CMS Hospital Compare measures are an example, but the submission process is hell. The constant answer of “you should only use the chart” should be met with “then give it to us”.
Process, outcome, administrative… all provide a slice of a view of a hospital, no one measure or set of measures gives you an accurate picture, but combine quality with cost with patient experience and you start approaching a real, actionable report.
The problem with quality reporting is that the only people driving the reports are the providers, and what they need to see is very different than what the consumers needs to see. What’s relevant to a medical director is not necessarily relevant to my mum.
Each and every measure gets heavily vetted for exclusion data, and each and every measure os open to editing, if you have a concern about a particular measure there are very easy paths to voicing those concerns and having the measure-making group aware.
PS – physician level reporting in the doctor’s office is very different from physician level reporting in the hospital setting, I’m not sure we’re talking about the same thing, but holding a hospital responsible for the quality of it’s physicians doesn’t seem to be a silly idea to me. I have no knowledge of any hospital setting report that lays blame or responsibilty on a specific physician, hospitals tend to be reported out more by system measures. However, in the physician office setting, a primary care provider being held responsible for the general health of his or her patient – and that’s what we consumers get called, we’re labelled “my patient” not “my client” – I’m not at all clear why that’s such a weird idea to see if GPs are doing what we think they should be.
You said:
“Thats reality in the hospital. And claims data does not reflect that.”
I agree. Further, *nothing* we can review after the fact accurately reflects anything, short of 24/7 video monitoring. Nonetheless, we need reports, we need transparency, we need as accurate as possible review of delivery of care and it’s appropriateness. No measurement system, in any setting, in any industry, is perfect. Health care is no different, and we have to report the best we can, not simply sacrifice the good on the altar of the perfect.
Comments on this entry are closed.