Apparently my first blog post, the mask icon and the name, “The Lone Economist” didn’t make it clear what this blog is about. So, let me be more explicit.
“The Lone Economist” is an allusion to the 1950’s TV show, “The Lone Ranger”. For those of you too young to remember, the Lone Ranger, was a masked vigilante who rode a magnificent white horse named “Silver” while dispensing justice to the rousing soundtrack of the William Tell Overture.
When I was a kid, I loved watching this show even though it never made any sense. There may have been some origin episode that explained everything, but if there was, I never saw it. Here was a guy who had no visible means of support and apparently lived outdoors, but managed somehow to always wear a clean, pressed shirt. How he could afford to feed his horse, much less himself, was a mystery.
So why did I choose to associate my blog with the Lone Ranger? Well for starters, I’ve always wanted a job where I ride up on my big white horse and tell a bunch of people what their problem is. And I get to refer to myself in the third person!
And then there’s the masked vigilante thing. You see, unlike the Lone Ranger, the Lone Economist has a day job. In keeping with the western theme, I’m a hired gun. Moneyed interests pay me to do their bidding. Nowadays we’re called consultants and expert witnesses.
Although I’m paid to give my honest opinion, that’s seldom what my clients really want. They want my useful opinion and that’s what I have learned to give them. I have never lied, mind you, in any professional capacity as a consultant or as an expert witness, but I don’t volunteer that I’m not being asked the right questions either.
I‘ve found that complete honesty and candor can be a liability at times. For example, several years ago I was hired to testify for a hospital that was being sued by a doctor for wrongful termination. The doctor was a brain surgeon who alternated with another brain surgeon in the hospital’s surgical on-call rotation. That meant that every other week this doctor was the first surgeon that would be called if there was a patient who required emergency brain surgery. These are mostly head trauma cases and, not surprisingly, the mortality rate for these patients is pretty high. So when a patient dies, one does not think to immediately call a medical malpractice attorney.
Needless to say, a place in a hospital’s on-call rotation is very lucrative and not one most doctors will give up easily. When I asked why this doctor’s privileges had been revoked, the hospital’s lawyer just said he had a reputation for being “rough”. My job was to show that the hospital was well within its rights to terminate this individual.
Most states collect a lot of data from their hospitals and make them public. This hospital’s state was no exception. With those data in hand I set out to test the lawyer’s implicit hypothesis concerning Dr. Rough.
Please note that “Dr. Rough” is not his real name. Like “Dragnet” (another 1950’s TV show reference!), the names have been changed to protect the innocent, specifically me. I don’t want to get sued. If there is an actual Dr. Rough somewhere, that is not the person I am referring to.
The first step was to calculate the doctor’s surgical mortality rate, which is easy. It’s the number of surgical patients who die divided by the number of surgeries. Every surgeon knows exactly what his or her mortality rate is. But a high rate is not necessarily an indication of a bad surgeon.
Surgeons with high mortality rates invariably say that their patients are sicker than other surgeons’ patients. “All the really bad cases are sent to me because everybody knows that I am the best” is a typical surgeon’s lament. From my experience, this statement rings true in two different ways. It is true that many of the best surgeons in the medical profession have high mortality rates because they treat the sickest patients. And it is also true that surgeons are an extremely conceited lot.
Here’s a riddle for you: how many surgeons does it take to screw in a light bulb? Answer: None, surgeons are too good to screw in light bulbs.
So why do I bring this up, you might ask? Well, besides an opportunity to insult surgeons, it highlights a common problem in the use of performance statistics for healthcare providers.
In statistical terms, this problem is called omitted-variables bias or confounding. All the differences in the mortality rates are being attributed to the identities of the surgeons and none are being attributed to the differences between their patients. Some patients are young and vibrant. Others are old and fragile.
One obvious solution is to add more variables to the statistical model. If all the relevant variables are included, then none are omitted. But human health is too complex to completely capture in one or even a dozen variables. So, no matter how many variables we add to the model we can never be sure we have eliminated all the confounding.
The only surefire way to eliminate confounding is to randomize, i.e. randomly assign patients to doctors. That is the basis for all randomized clinical trials. It’s how we differentiate causation from mere correlation.
The more I thought about it though, the more I realized that we didn’t have an omitted-variables bias in this instance. It would have been impossible for the hospital to take in only severe head trauma cases in odd numbered weeks and only minor cases in the even numbered ones. By including Dr. Rough and the other brain surgeon in its on-call rotation, the hospital had unintentionally conducted a randomized clinical trial.
And what was Dr. Rough’s mortality rate? 24% versus the other brain surgeon’s rate of 8%. That amounted to an odds ratio of 3.0 and was well past the threshold for statistical significance. This covered several years and dozens of deaths. For every 30 of Dr. Rough’s patients who died, 20 would not have if the other surgeon had treated them. It would have been criminal if the hospital hadn’t revoked Dr. Rough’s privileges.
So, I presented my findings to the hospital’s lawyer expecting high fives and pats on the back, but instead I got “Are you crazy?!! I can’t have you testify under oath about this! I had no idea it was this bad! We’ll get sued into oblivion!” I’m paraphrasing and cleaning this up a bit.
Anyway, I was told to stop billing time to the case immediately and the hospital paid Dr. Rough a million dollars to go away quietly. I had found scientific evidence that Dr. Rough was in effect a serial killer. The only difference between his rampages and those of Jeffrey Dahmer and Ted Bundy was that his were unintentional, perfectly legal and performed in an operating room. I got the old heave-ho and he got a million bucks. The absurdity of the situation left me stupefied.
You might wonder why I didn’t go to the authorities. Well, which authorities? I couldn’t go to the police. There was no criminality here. Dr. Rough was just a bad doctor. There are a lot of bad doctors out there. His was just another face in the crowd.
What about the state medical board? Couldn’t they take his license away? Not with the evidence I had. Although scientific, it was still only a point estimate. His incompetence might have actually killed only 18 or maybe 23. I just didn’t know for sure. And I couldn’t say exactly which 20 of the 30 patients he had killed either. To take someone’s license away, the accusation and the evidence need to be specific.
More to the point, I had a professional and contractual duty to my client, the hospital, not to put them in legal jeopardy. Malpractice attorneys sue for money and Dr. Rough didn’t have much. The hospital, on the other hand, had very deep pockets. It would have suffered the full brunt of all the legal claims.
Hospitals can’t win. They get sued and lose a lot of money when they do the wrong thing. And they get sued and lose a lot of money when they do the right thing.
But back to the main topic, you see my dilemma, right? I can’t talk about this stuff openly. I’ve seen things. I know things. Things that would test one’s faith in the basic goodness of lawyers.
Sorry, I couldn’t resist. No one in their right mind has faith in the basic goodness of lawyers. Wow, I’ve insulted two professions in one post. I’ve exceeded my quota!
And then there are my government agency clients. I don’t know how they would react to the news that I think Obamacare is doomed and the indiscriminate expansion of Medicare would be a disaster.
History is replete with heroes who bravely spoke truth to power in the full light of day. The Lone Economist will not be one of them. Remember John the Baptist? He spoke truth to power and the power spoke back. The Lone Economist does not want his head served on a platter.
So, like my childhood idol, I’ll conduct my lonely crusade anonymously from behind a mask.