Saturday, October 13, 2007

Ranking Doctors Using Claims Data Legal

I’ve never liked the idea of health plans using claims data to steer members to doctors. I don't think you can use data to judge human interaction in the patient-physician setting, what actions followed, or the quality of the physician.

But the use of claims data by health plans to steer patients to "quality" doctors is happening more and more , and you ought to be aware of what’s going on out there. Health plans are mining claims data with sophisticated algorithms to rank doctors for quality and economic performance. Doctors in New York State, Washington State, and Connecticut have objected, and have filed law suits, saying rankings focus primarily on cost, not quality.

To address the doctor backlash, George Washington University School of Public Health and the Robert Wood Johnson Foundation did a legal analysis. The analysis indicated rankings are legal under state laws. Their experts maintain physician rankings are critical to a broader movement to measure and publicly report on physician and hospital performance and to help improve the quality and value of health care (Source: Study Underscores Legality of Physician Ranking and Public Reporting Systems When Used to Improve Care Quality, Oct 9, 2007 - Washington, D.C.)

More health plans are rating physicians on quality and efficiency measures and are designing insurance products classifying physicians in tiers. Plans are given incentives to members to select certain high performing physicians. The GW legal analysis concludes transparency is critical when designing a tiering process.

The process should include physicians in the rating process and should be open and visible to physicians and allow their input. Classifying doctors based on the quality and efficiency of services is legal, and so is publishing information regarding health care quality. But undertaking these efforts in the dark could and should lead to physician backlash. Proprietary software should be open to inspection.

Standards should be developed by state insurance regulators in the case of insured plans, the Centers for Medicare and Medicaid Services in the case of Medicare and Medicaid plans, and by the U.S. Department of Labor in the case of ERISA-governed plans.

To me, much of this comes down to who do you trust – hard data from health plans, state and federal government, or subjective opinions of physicians. We shall see if these quality-directed efforts result in better or less costly care – or if they simply justify the economic existence of health plans and add to the bureaucratic red tape already strangling medical practices, or if published data alters the physician-patient relationship.

For doctors, avoiding judgments on quality based on data mining is likely a losing proposition To health plan executives and federal bureaucrats, this aphorism still holds, “In God we trust. All others use data.”
This is for your information. Your comments are invited.
The George Washington School of Public Health and the Robert Wood foundation have released an analysis saying that health plans can legally use sophisticated algorithms to mine claims data to steer health plan members to physicians whom the data indicates practice quality and economic medicine. The GW and RWJ experts maintain physician rankings are critical to a broader movement to measure and publicly report on physician and hospital performance and to help improve the quality and value of health care. The process, the analysis adds, should be transparent and should invite physician input.
For full story, see


Howard Luks MD said...

Another nice piece Richard.

In the future I envision a situation where the industry not only tiers the physician ---but they will then combine that with their "cost" ranking--- to establish a punitive co-pay to see each doc. Once the patient becomes financially responsible for their decision then I think the situation will prove to be a problem for many practices.

Right now I imagine most patients are intelligent enough to see this for what it is. Similar to Care Managers, many patients will hate the intrusion they feel the insurance company is making into their private decisions.
We'll see.

Dr. Bonis said...

Hard data is just a help, but not the way to knowledge (and even less to wisdom).

Trusting in hard data and mathematic models to take decisions in a complex and fuzzy domain as are medical decisions and doctor-patient relationships is not new. Remember for exampel the story of McNamara and USA-Vietnam War.

In my opinion hard data is useful to detect trends or to compare groups (aggregation), but not individuals. In the outcome of healthcare (quality) there is a complex combination of factors (doctos, nurses, the own system, the patient socioeconomic status, adherence to treatments, environmental factors...). That complexity can be only afforded by a combination of hard and soft data (soft data that is inside the head of the clinicians).

The risk of ranking docs (and traslate this in money rewards) based in hard data is the selection of patients. If I am a doc and want to obtain no claims and good results I will try to select the patients that better respond to my treatments (the less problematic, the more literated, the less comorbid...). There are several ways to trick hard data anaylisis (this is known as statistic hacking and most doctors have done). The "Inverse Care Law" described by Hart in 1971 in its biggest expression!!!

Even there is a "diagnosis hacking", it happends when you make a medical decision about a patient (to admit in the hospital for example) and AFTER THAT you look for a diagnosis to justify your decision (constitutional syndrome or inestable chronic hearth failure of unkown origin are classics).

ObGynThoughts said...

Hard data about what? The problem is how to measure quality. For example: percentage of patients over 40 that get mammograms. Measures not only how often physicians order them, but also compliance by patients. Still only a tiny fragment of what a physicins does during an annual visit. I am sure the cost of prescribed medications is a factor, the cheaper, the better, I assume. The best medication nevertheless is the one most appropriate for this condition in this individual patient at this specific time. Hard to "measure". It is pretty tricky to get "quality" right. HMOs WILL take the shortcut of weighing cost more than anything. In my experince, this is another sleazeball move by insurances looking to pay physicians less. I will tell my patients to stay away from those physicians that rank high on the HMO lists, since I suspect that they will shortchange patients and over-emphasize cost.

Richard L. Reece, MD said...

I think you may be right. You tell the patient what the doctor
"costs", then you punish the patient if the doctor is high-priced. This may drive down quality and be considered intrusive, and those reasons might not work.

Florence said...

I fully match with whatever thing you have written.