Bill Keller: Pay For Performance/ Doctors and the Bandwagon

Bill Keller

There is something to be said for a liberal arts education. Bill Keller, a former editor of the New York Times, now a freelance writer examines in today's Times the whole notion of "pay for performance"  (P4P) as it applies to doctors. This is a response to a recent New York City effort to institute changes which are designed to reward doctors for delivering good care rather than for the number of procedures they do. 

Mr. Keller is not a physician, has never worked in a hospital, and in fact the only way he can know about what drives costs in our medical system would be from reading, talking to people and thinking, the sorts of things a liberal education is suppose to facilitate. Mr. Keller is the son of an oil company executive and Mr. Keller is a graduate of Pomona college. Other than that, he's been in the newspaper business.

Which is not to say being a doctor automatically qualifies you to comment on what drives health care costs and what could reign them in: Atul Gawande is a practicing surgeon, who also has the ear of the New Yorker editors and he has made lots of suggestions over the years in his articles which fit the characterization Mr. Keller applies to the P4P bandwagon:  great in theory, bad in practice. When it comes to analyzing medical care systems, the surgeon Gawande is a naif. A naif with the ear of The President of the United States and others, but wrong most of the time because, while he may be a good surgeon, he is clueless about systems. He is trying to do fresh thinking, but he often winds up simply thinking simplistically. 

Pay for performance has a nice meritocratic ring to it:  Let's pay for value. But all efforts to figure out what good is, what you value,  and then assign a code number to that and then put that number in a column on a spreadsheet,  have been frustrated by the complexity of human behavior.  

The main problem is trying to go to a bottom line and it's the problem with judging teachers by the performance of their students on standardized tests: The teachers have less control over the outcome than the parents,  the socioeconomic status of the students, their friends, the aspirations and resources of the kids taking the tests.

In the case of the doctors, proposals to reward ER doctors for the time it takes to clear a patient out of the ER run up against all the people who stand in the way once the doctor has done his work. The doctor cannot control the transport staff ("the escort service") or the nurses on the floors, or the people cleaning out the ward rooms or a myriad of others.

Citing a variety of studies, Keller notes there really is no good data to suggest American doctors order more tests, do more procedures or spend more of the system's resources on their patients than doctors in Europe. 

The major difference is what American doctors and hospitals get for their efforts: The $5 aspirin pill, the $100 fee for cleaning a hospital room, the $2500 MRI.  As readers of the Phantom will recall, the Phantom has murmured all along there are many procedures which are vastly overpaid--colonoscopy and MRI being the two most egregious examples. Doctors don't like to make enemies of their colleagues by saying this, but you have to come to a Chris Rock moment, eventually, where you say, "There it is! I said it!"  Okay, we all know it's true but we are not willing to speak its name in public.

Is there really any reason a young woman 5 years out of medical school, who has no on call, works no nights, no weekends, works 4 days a week, pays small malpractice premiums, should make $500,000 as a dermatologist for cutting out lesions which a technician with 6 months training could do?  Should the "aesthetic" dermatologist who sells potions from her office store, does laser removal of lacey varicose veins, who never has to face a patient with a potentially lethal disease,  make $500,000 a year?

Should a man who spent 4 years in college, 4 years in medical school, 3 years in fellowship, but who learned colonoscopy and endoscopy in about 9 months be able to charge enough for a colonoscopy to be able to work 4 mornings a week and make $350,000 a year, doing a procedure a technician could learn in a year? 

Why does a MRI cost $75 in Europe and $2500 in the USA? The MRI can be done in Silver Spring, Maryland, read by the radiologist in Mumbai and the hospital in Maryland gets $2500, but the hospital where the doctor in Mumbai does his day job gets $45 for the same test.

The list goes on. 

The problem with P4P is it depends on developing some meaningful metrics. And as Howard Colvin said about statistics, it just ruined the job of being a police. It did more than ruin police work for the police; it ruined police work for the citizen. Before being judged by the number of arrests, the number of tickets written, the number of drug busts made, the cop on the beat, strolled down the sidewalk, and stopped to talk with the woman on the stoop, got to know her, and she got to know him. And when something "went down" in the neighborhood, a rape, a shooting, the police could go to that woman, who knew his name and she knew his name and he could find out what happened. When statistics replaced policing, the policeman became part of an occupying force and nobody would tell him anything. 

No snitching.

The trouble with a system run by organization men is they often have no feel for the real work of the police, or the doctor or the mailman.

There was a mailman in my building who delivered mail to 175 offices in my building alone. One day he delivered a letter from one of my patients, correctly, to me but the the letter had the wrong name on it. She meant it for me, but was talking to someone at her office as she addressed it, and she absent mindedly wrote the name of her co worker rather than mine. She got the rest of the address correct, the name of the building, (but no office number), the street address, town, state, zip code. 

So how did the mailman know the letter was meant for me?  

"That letter from Donnelly Construction," the mailman said. "You got that same letter from that same lady, in the same Donnelly envelop every Friday for two years. I knew her handwriting, and I knew her name and I knew she just got the name wrong, but that was her weekly letter."  

He was right of course. But he begged me not to say anything about it. I wanted to write him a letter of commendation. "Oh, no! Don't do that. That'd cause me a world of trouble. I should have returned that letter to sender. Addressee unknown."

The bosses want every worker to be interchangeable. His memory, his attention to detail, his ability to pick out one letter from the roughly 3500 pieces of mail he delivered in our building every day--and he had five other buildings of similar size on his route--would be considered a negative value. He had not followed the rules. His large brain was something they had not accommodated  their system for.

The Phantom is not saying there is no way to assess quality of care in medicine. He is just saying, the metrics suggested will not do it. The assessment would require people who actually know what good is to be making judgments and you could not just plug anyone with a certain degree on their wall into the judgment panels. And you could not look to U.S. News and World Reports for an indication of excellence.

You'd have to do a lot of thinking and listening to get worthwhile, meaningful, considered and educated judgments. 

And, so far, there isn't a puff of smoke on the horizon to suggest anyone has actually got serious about that.