Tuesday, February 19, 2008

Data Data Everywhere, Nor Any Drop To Drink

A lot of people think the data for publicly reporting the quality and cost of a physician visit doesn't exist.

Which is, of course, nonsense.

Doctors are service professionals who for 95% of their clients bill a third-party. It might be Medicare, Medicaid, a private health insurance company like HIP or CIGNA, or others. Very, very few people - in the USA - pay their own bill.


So we have these huge databases of things doctors billed for. It's called billing data, or administrative data. Health providers use the term in direct reference to this data not being clinical data. As in, it must then be useless data. In the industry, billing data is mud, clinical data is sacred.

The way it works is this: your doctor writes stuff down that will become your chart, your medical record. Someone else then abstracts from that things that can be billed for, and creates a second set of data that tries to describe what the doctor did so a bill can be sent. Your clinical data is not sent to anyone for billing purposes.

In public reporting, as it stands right now, we can get our hands on billing data pretty easily, clinical data is nigh impossible.

People like me involved in the public reporting of health care data hear a lot of providers scorn our use of billing data.

And like I always say, as soon as you give me the clinical data, I'll start using it.

So what's in this unusable billing data?

Office visits. Prescribing antibiotics. Breast exams. Diabetes care. Removing surgical staples.


So we also have huge databases of things doctors didn't bill for. Well, we don't really, but there's a lot of stuff that isn't in the billing data and it's educational.

If a doctor didn't bill for something, I'd say it's a safe bet it didn't get done.

More precisely, given a large enough dataset, where a pattern can be established we can determine staistically significant variations of care delivery.

I've seen a lot of doctor's bills, and they stack a bucketload of stuff in there. You go for one check up visit and look at the paperwork your insurance company sends you. There's at least half a dozen line items on there. It's a very inclusive bill.

Medicare, for example, knows each and every one of its "beneficiaries", let's call them "members". And each member is entitled to certain care from a doctor, who we'll call "service providers".

And each member who doesn't get that care should probably get a call from their service provider to come get their care. If you're my doctor and you don't give me my care, surely my payor (let's call them "payers") should nudge you?

And do you wonder why billing data is electronic but clinical isn't? I don't. Billing data pays the bills, gets the checks cut. Clinical data is good for little more than monitoring clinical services, which we all know no-one wants to do.


So we have huge amounts of billing data. Huge. Tons. We can tell if diabetes patients get billed for diabetes checkups. We can tell if women over 50 get their mammogram every two years. We can tell if little Joey got improperly-prescribed antibiotics for a cold. We can tell if people on antidepressant medication are being properly followed-up with.

We can tell because if these things are happening, someone, somewhere is going to get paid for it.

Surprised? You shouldn't be. This data is up and out there already, but we can't get people to look at it. This Web site lists literally hundreds of quality reports you can go look at. Go look at them. Here's another site chock full of public reports.

So why can't we use this data and hold our heads up high in the street? These are the kinds of things I hear:

"Billing data isn't accurate."

"My patients are sicker."

"My patients don't do what they're told."

"100% is unachievable."

Or, my all time favourite, the one that gets me really, really, super riled up:

"Consumers can't understand the data, or they'll understand it wrong."

Most physicians are doing an awesome job in the face of all kinds of adversity, they're dedicated professionals delivering quality health care. A small percentage are behind the curve. The argument here is how to identify them, and until we get better data that measures how we deliver care, not how we bill for care, we're stuck making highly complex, educated guesses.

Do me, and you, and the rest of us a favour. Use what we make. Go research your health care options. Use the links above and go learn about the hospitals in your region, the doctors in your neighbourhood. And tell them you're doing so.

Push for electronic clinical records. Demand that your health care information be as accessible and as accurate as your banking record or your credit record.

Because the sooner you start using the data we put out, the sooner we can get better data.



Raymond said...

A wonderful post! It is amazing to me the amount of free information out there on health care providers that the majority of people don't seem to use, is hard to understand, or has to be paid for. With the advent of the internet, being able to find detailed quality metrics about your health care providers should be a thing of the past. Sadly, there are all sort of hurdles (many you mention) that ensure this is not the case. We are moving in that direction as a country, but it is a slow pace.

Disclosures and Disclaimers


My employer is compensated through funding to provide analytical research, technology solutions, and Web-based public and private health care performance reports by the State of New York, the State of Illinois, the Centers for Medicare & Medicaid Services, the Agency for Healthcare Research and Quality, the Commonwealth Fund and Bridges to Excellence. I am not being compensated by any of these organisations to create articles for or make edits to this Web site or any other medium; and all posts authored by me are as an individual and do not represent my employer or the agencies I work for.