Guest Post from Barbara Duck: http://ducknetweb.blogspot.com/
The Rise of the Quants, Again! This Time In US Healthcare- Taming the MACRA Beast of Quantitated Madness For A Lot of Things That Are Probably Just Not True
I’ve been watching this for years with the way healthcare is changing in the US today and we certainly need and do use constructive and good measurements in what the industry does, but just like in the financial world, we’re starting to cross some borders here into madness where mathematical models with “no” proof of concept even are producing numbers that don’t compute and host environments that are hostile to the future of providing good care. In other words, it’s time to take to take a step back and look at where we are and it’s not pretty nor productive in a lot of business areas today.
It was back in 2009 I started noticing the correlations of what was happening in the financial world and compared it to what is taking place and healthcare and when you do a bit of research and reading on this topic, it’s not hard to figure out that some of the healthcare models are not much more than some of the financial models that have been tweaked a bit to try and perform the same linear methodologies that are either flawed from the start, or that contain proprietary computer code that developers have been selling with the tag line “trust me”. “Trust me” is becoming a huge problem as proprietary software and algorithms are all locked up in what is called a “compiler”, in other words nobody without proper credentials can see what the magic sauce is.
There are good reasons to have some “magic sauce” around though and folks patent a lot of it. Back in the early days of the mid 90s, when all were working to create tools that actually helped us with business this was not really an issue, but the world has evolved today and that’s not always the case. We can look at Volkswagen as an example, which we all should know by now; they cheated with their computer code and built a “virtual world of cars that did not pollute”.
With software you can do something about anything, and the methodologies can be true and they can be false as we learned. I borrowed that line from the video documentary about the Quants of Wall Street, the Alchemists and if you have not seen it, scroll down to the bottom of this page and watch it, as we have tons of Quants in Healthcare today, just like we do in finance.
I read every day with the frustrations of the average consumer trying to make sense out of what’s going on and they can’t. This is a world controlled and ruled by individuals who are both responsible and irresponsible in what they design and create out there and both exist. All one has to do is look at Wall Street and know there’s a lot of folks up there who cheat with computer code, an issue that even those who are elected to make laws in this country can’t understand, and that’s scary as that’s their job. Former Head of the Department of Justice, Eric Holder, determined he didn’t want to explore this avenue of crime either and turned his head away from having to deal with crimes that are committed with cheating with computer code and math models. We saw that plain and clear in the PBS documentary called “The Untouchables”.
I used to write software and I look at what some of the journalists put out there today and truly I ask myself “where in the hell did they get that interpretation” with some of the news articles? Again just like the average consumers, they don’t understand or make an effort and we get “junk” news and interpretations. There are some good exceptions out there though, with real journalists, and not jounobots that do dig in and create some good articles, so this not an attack on the entire industry, just the ones who are blinded with Perception-Deception and produce a lot of just with quantitated madness statistics on things that are not true. We all know at this point that a lot of the news we read today is “rigged” and the complaints are numerous.
Media newspapers are running out of money to put out the news as Facebook and a few others have taken a lot of their ad revenue which is used to support them and wrapped it into their news feeds which also use algorithms to tell you “what you should read”. God help us with the dupes that are totally sucked in there as well. I call it the Dupes of Hazard society today. Take a look and see how easy it is to produce fake news today with a bot that does all the writing. One bot is so good that the creator used it to create 10,000 books and put them on sale at Amazon. These jounobots are not just restricted to writing content either; you’ll find them writing content to verbally justify a lot of quantitated madness we deal with today.
News Rigging Has Arrived! Astroturf and Manipulation of Media Messages-TED Video About “The Fake Grass Roots” Of Big Pharma and Other Campaigns That Fool and Fool Again…
So now we move to the next level of quantitated madness out there with the new CMS MACRA rules for grading doctors. We all know that the Meaningful Use model has bit the dust and logically, all models are not created to run forever, as some might think. If you talk to a good Quant though, they will pretty much tell you that you can’t take another model, load it on top of a model that’s already broken and expect a good “fix”, and that’s what I see happening here. There’s almost a 1000 pages of quantitated madness that is going to reward doctors for better care? We do need some reporting and that’s the point of this post, the point is that with putting a new model on top of one that is already broken, well guess what, this one will be broken too.
Broken models are not fixed by created yet another model on top and lead us into very complex areas of math to where nobody will understand what’s going on. With that in mind, health insurers have done this for years and I’m not the first to state that they make money based on complexities and keeping consumers “off balanced”. Insurers have a captive audience and they know it and have been modeling like this for years!
Others know this as well so I’m no genius with this conclusion at all. As a matter of fact way back in 2009 when Sebelius was nominated to run HHS, before she ever stepped foot in that office, I said exactly this would occur, that insurers would reduce her to mincemeat as she’s out of the same mold at Eric Holder was with not wanting to deal with reality and how those smarter than her would use their quantitated skills to model all over her and get what they want, the insurers.
So not much has really changed in essence from that day and we are still continuing on the same path. I call it STAT RAT FEVER as my own coin when I talk about the government’s activity in healthcare as that’s all they do. Currently we have two folks running HHS and CMS that were former McKinsey consultants. If you have not read some of the books out about the brain washing institute, do it and you’ll understand what goes on.
I call it the “Scientology effect” of the business world as you find their consultants everywhere and some have even gone to jail. Recently we had Mckinsey written all over Valeant, who again before he was the CEO of that company, was a long time McKinsey consultant. Almost anymore when I see any study from McKinsey using the “T” word, “trillion” that could be saved, you have to laugh as they use quantitated madness reports to get you upset and motivated to believe whatever kind of numbers they want to sell that day. Of course if you go back in time and look up some of their old studies and compare them to what has really occurred, you do find a lot of fictional marketing in some of what they have published.
We do have one doctor at the ONC who is no longer the chair person, DeSalvo who for about a year bit on the perception-deception tall tale that medical record developers were deviously writing code not to interoperate. That was a marketing ploy of a big EMR company that was planted out there as “marketing” to make people think that medical record vendors were “on purpose” writing code that would “block information”. Of course anyone whose’ been in the business long enough knows that’s a farce as it takes time and money to write code and no vendor is going to do that, and maybe they just didn’t have the time, money and resources to write enough code yet to match her perception of what interoperable code should be (grin). I thought it was great when Dr. Halamka addressed it for what it really is, and he’s a doctor and a computer scientist at Harvard Medical and if anyone knows what the correct perception is, he does.
So hold on to your hats with MACRA as both Burwell and Slavitt are both former McKinsey consultants, where some basic training was attained, if you will. Neither have any backgrounds in the clinical side of healthcare so at some point, one has to ask, why don’t we have some folks with clinical backgrounds at the top of the helm in government healthcare? I think it’s a good question. We can all go back and see the long extensive background of Andy Slavitt with his former position at CEO of Ingenix, which was renamed to Optum Insights and how when thing got really hot over there with all the lawsuits, United Healthcare folks just moved him over to run QSSI when United purchased that company to get in on some of the Healthcare.Gov action with more algorithms and software.
In addition the “do not disturb” sanctions given to Slavitt at CMS were unheard of. All those lawsuits would distract him from doing his job? Well why don’t we look at how and why all those lawsuits came to be? He’s a Quant and also spent some time at Goldman Sachs, and the banks really heavily on their Quants to model their financial models. If you happened to see “The Big Short”, they offer an explanation there as to “what a quant does” in a humorous way. You hear “listen to my quant”, and that indeed was a good portion of the movie that tells you right there what the quants are hired to do to ensure profits. It was funny in the way that the quant was presented as an Asian individual who didn’t speak English, but could write good code and then there was the insert of him blowing that perception into the dirt as he explained “what he really does”.
There’s a lot of models out there that don’t work. One of the most well known examples is the Bloomberg Big Gulp. The developer of that process actually ended up leaving Bloomberg and went to work at the CDC and created yet another model that didn’t work called Obesity for the Workplace. Again these are theories and some kind of numbers that may or may not work. It’s hard sometimes not to get caught up in the perceptions at times too as we are presented with a ton of mathematics that say it will work. As we all know by now, the Bloomberg Big Gulp mathematics failed as “people don’t work that way”. I see it time after time when quants and developers think they have “the answer” with a vision in their heads on how people will work with their code and models and it flops. There’s no easy answer to a lot of it as well and you need at least some proof of concept developed, but we don’t even get that anymore. I speak of this as I got fooled too in this game and saw it a long time ago. You think you have the answer only to find out you don’t.
When the financial stakes get high though, you do get marketing folks called in to write all kinds of quantitated justification stories as to why it has to be the cat’s meow and the peanut galleries on the web, toss out all their potential skeptic feelings and simply just buy in to the madness. I wrote about this a while back and called it “Operation Perception-Deception” in the US before everyone else started jumping on the same band wagon. You implement enough failed models and people eventually stop drinking the Kool-aid. In addition, some of the skewed quantitated madness has people just flat out doing strange things with attacking symbols like flags and so forth, as again they are wrapped up in a world of technology that they can’t see or understand, with a very small number of Quants and developers creating algorithms that are denying access and eventually taking away a lot of our freedoms in the US.
If I were a doctor today being subjected to this new grading system, which is basically a re-write of a lot of what United Healthcare has been developing with a lot of their models, I would be a skeptic indeed. When you look at the fact that CMS sanctioned Cigna health insurance for not providing quality care, and then took away their penalty (nullified the rules of the game) and then come back with this to grade doctors and think it’s going to create better value for care? We have some folks out there not looking at big picture. Admittedly the health insurance models for grading has failed so what makes you think this MACRA model is going to be an accurate representation? Think about that.
I’ve learned some of what I know from mathematicians that work for universities that have written and put out some extensive work in this area. Charlie Siefe at NYU in Mathematics has been very outspoken and if you watch video #1 in the footer, you can learn a lot from just one video. He also wrote the book called “Proofiness, the Dark Arts of Mathematical Deception”. I’ve also had some chats with Ed Frenkel, mathematician at Berkeley and we all seem to be on the same page. A quant is both a mathematician and one who writes code. Watch video #2 in the footer to understand how they function in the financial world and you can easily see how their work slides right over into healthcare, especially when there are profits to be made. This is the real deal out there today.
How do we know that MACRA is not going to turn out to be another Bloomberg Big Gulp type of model? There’s no proof of concept, no pilot, it’s just a matter of tossing a bunch of quantitated madness out there that they “think” will be able to create value in healthcare. At this point with all the complexities of this 962 page model of mathematical complexities I think it’s a good idea to read the post from Dr. Halamka, CIO of Harvard Medical, who for years sat on the committee to help guide the ONC and HHS on their models…basically it’s a monster that provokes one to think about leaving the practice of medicine. He’s smarter than anyone in government who’s responsible for creating this model of madness that has no proof of concept what so ever.
Again as I mentioned early, you can’t put a new model on top of a broken model that’s already failing and hope to have a good fix, it just won’t work and you need a model that doctors can work with and not a ton of quantitated madness for things that just maybe not be true. When it comes to models, I think we all know or should know by now that United Healthcare has been the mentor in place at HHS for a long time and there’s tons of documentation all over the web that substantiates this. I still think about the New York Times article from 2009 when Senator Rockefeller asked both Slavitt and the CEO, Hemsley of United “how do you folks sleep at night”, link to the post below.
Is This a Case for a New Law – Illegal Algorithms? How Do You Sleep at Night Rockefeller asked the CEO of United Health Care
Here’s another comment from a doctor on the entire MACRA process…and I’ll close with this. BD
“When Blumenthal was at ONC , he understood doctors needed to see patients, and the EHR was one of many tools to help doctors care for patients. Farzad got us off track with insane busy work in stage 2, on now we are completely off the rails.”