Saturday, November 28, 2015

Climate Change, The Nuclear Option, And Organizations

With the Paris conference on climate change beginning in two days, and with many thinking that this is the world's last realistic chance to plot a path toward survival, a venture capitalist named Peter Thiel makes a case for nuclear energy in the NYT.

He is not wrong. The overwhelming issue of our time is, will mankind survive? In my early lifetime the threat was the Nazis – yes, mankind would have survived, but the history would be troubled. Then the threat was nuclear weapons, and truth be told, it still is. On The Beach was not unreasonable fiction, except how beautiful Ava Gardner was, and I guess Gregory Peck. But the threats are accelerating. nClimate change, né global warming, is arguably the most serious of all. James Lovelock sees disaster but not total annihilation. There will still be “mating pairs in the Arctic,” he says, which is less than totally reassuring.

What is behind it all is the exponential expansion of the anthropocene age. ( I have used the image of the beehive before, and it comes to mind again. When humans crawl all over the world and all over each other, they stay alive by an interconnectedness that works for a limited time. But nothing stays the same for long, and if you are precarious you eventually fall, and that's inevitable with our exponential-ness. The only alternative is planning, and restraint, and a sense of all being into it together. Good luck with that.

Nonetheless, we keep trying, and nuclear is an important option, as Thiel says. Years ago, when it became apparent that we had a problem and my Dad had not yet died, he and I discussed it in a couple of words, which was all we needed since we tended to think alike. All of his professional life as a neurosurgeon, my Dad had looked at dangerous operations and unpalatable choices, delayed decisions and followed cases assiduously until a choice became clear and sometimes inevitable. And when we talked, even in his decline, he faced the problems of the earth with the same coldly reasonable calculations that he used in judging the advisability to operate. So when I asked him, “Dad, what do you think about climate change and energy?” he responded, “I think we have to go nuclear.” Which was just what I thought, too. I figured, you just have to be careful, find the right way to do it, and put the radioactive debris in the right spot – I figured, send it to outer space, but then, I tend to be a risk taker.

The problem was, however, that we were both thinking like doctors, like Lone Rangers, which is what old fashioned doctors tend to do. We were thinking, get your best doc and go on in and be careful to do it right, concentrate, do it right.

But that way of thinking is a mistake. Nuclear power can't be operated by a Lone Ranger. Nuclear power has to be run by a large organization; as a matter of fact, since the world is a big place, it has to be run by many large organizations. So it's not an individual challenge, it's an organizational challenge, which is a horse of a different color.

Earlier this year we visited Japan and, as is my custom when traveling, I read the local English language press. There was an article about the Fukushima accident of 2011. I remembered that it was clear early on that Tepco, the utility company that ran it, had disingenuously reported about and dealt with the accident, not to mention having made mistakes with the initial design. It seemed to be a poorly run organization with strong political support charged with what could be a dangerous mission. Now the Japan Times reported that no significant changes in administration have been made. I read that a citizens' committee – a citizens committee! What a sign of entrenched bureaucracy! -- had twice (twice!) recommended that the top officials of Tepco should be tried for ineptitude and negligence. They reasoned that punishment would have a deterrent effect in the future, and thus perhaps change organizational behavior. But the article reported that twice had this recommendation has been dismissed by the establishment prosecutors.

It reminded me of the investigation of Chernobyl, which found that political appointments had so infested the organization that it was non-functional. To the Russians' credit, they made the investigation, just as the Japanese did. But the consequences for personnel were few; after all, these were favored apparatchiks who had been placed there in the first place.

Here's the point: you can build organizations for a purpose and get impressive success. You can win World War II, go to the moon, build the 50's Yankees. But then, over time, organizations decline. They just do. Guardians of nuclear missiles in North Dakota get lazy and drunk and no checks are made on the missiles. NASA leaders dismiss warnings from their intimidated scientists that the O-rings might not work at low temperatures. CDC forgets to safeguard the smallpox vials. Achievement organizations become blame organizations, as incentives shift from high accomplishment to avoiding blame, where personnel shift from dream-seekers to get-a-job-and-get-promoted seekers. It's just the natural history of organizations.

So as the VC Peter Thiel touts nuclear as the answer, he thinks technically, not organizationally. VC's think about starting a company and selling it, usually, not operating it into perpetuity. Could you build nuclear plants all over the world that would never fail, or if they failed, would have damage contained? Maybe. Computers have become very powerful. But there are still operators all over to oversee the plants, and if they are not unqualified apparatchiks, or Tepco bureaucrats, they can still be Homer Simpsons. 

Thiel thinks that the downsides of accidents have been exaggerated, and that side effects can be contained. Maybe he's right. Maybe my father and I were right. Or maybe we need nuclear not as a permanent solution, but as a bridge to a future technology that would truly work automatically. And maybe running the risk is better than simply continuing on as we are on the road to mating pairs in the Arctic. And maybe Space X can take our nuclear debris on a quest to find dark matter.

But in the meantime, if nuclear turns out to be the necessary risk we take, let's not just concentrate on the technology. Let's pay as much attention to advance organizational theory as to engineering theory. That could be the biggest challenge.

Budd Shenkin

Friday, November 20, 2015

United Health Care Might Exit The Exchanges

There is some consternation today over United Health Care's considering exiting the field of the Health Insurance Exchanges – the Affordable Care Act's policies for individuals. Does this mean that the ACA is not working?
Well, it's a complicated situation. I am indeed afraid that the ACA is not working in a very significant way: the ACA has incorporated High Deductible Health Plans into its midst, and they suck, as both the NYT and the Boston Globe (below – quoting me!) have pointed out graphically this week:
The worse the plans are for people the more enrollment will fall, and I have to think that the high deductibles make people feel they are not getting much for their money, so some don't enroll or reenroll.
But I think the UHC issue is different, and doesn't indicate ACA failure as such. UHC is very big and the exchange programs are a relatively small part to their business. They can probably make a profit on them, even though they say they can't, but the profit might not be big enough for them to want to concentrate on it. GE's strategy under Jack Welch was to make sure that every GE unit was #1 or #2 in the industry, and if they weren't, they either had to be built up or sold. From a corporation's point of view, you need to have some focus. And as we appreciate that point of view, I guess it's not such a bad thing to see them exit from this business segment. It's just a decision that there are more profits to be had elsewhere, and a corporation can't do everything. Is it such a bad thing for us that UHC can't find that much profit in Exchange programs? I think it's fine.
Looking back, the idea of the ACA was to make insurance companies make money a different way from the way they did it before, which was largely by excluding subscribers by underwriting, and writing tricky policies. Instead, insurers would have to make money by efficiency and service. That was a big gamble for both sides -- can insurance companies reform themselves? Can corporate cultures change? If not, then better to exit the field, concentrate on traditional markets, and other health-related fields like Optum.
Here is the way I put it six (!!!) years ago:
There will have to be a period of time where old companies try to adapt, and where new companies could even come into being to meet the new standard. The new companies might emerge from health care delivery models themselves, I would think. We'll just have to see.

Budd Shenkin

Saturday, November 14, 2015

EMRs, Government, and Capitalism

I wrote in my last post how the government screwed the pooch on Electronic Medical Records by not requiring interoperability as it invested $30 billion in the enterprise. With that kind of money going to by EMRs and equipment, surely the proviso that they be interoperable could be included. Was it corruption or incompetence that omitted that proviso? I couldn't be sure. Maybe a mixture

So here's a follow up from JAMA, September 22/29, pages 1213-1215, a medical news article: “On the Road to Interoperability, Public and Private Organizations Work to Connect Health Care Data.” The reporter, Julie Jacob, interviews the Office of the National Coordinator for Health Information Technology (ONC) spokesperson, Erica Galvez MA. The article states:

Although the ONC road map presents the movement toward interoperability as a smooth, straight road, in reality it's more like fitting together the pieces of a giant jigsaw puzzle. Several public and private organizations are working on different pieces, including organizations creating technical standards, establishing health information exchange (HIE) networks, connecting those HIEs, and developing principles of governance and usage. The hope (emphasis added), according to health care information technology experts, is that all the pieces eventually will interlock.

Because the Feds have basically opted out of direct action and laws requiring interoperability, several organizations have formed to try to fill the breach, one from the AMA called Healtheway, another called CommonWell Health Alliance which was spawned by the EMR industry sans Epic, and which requires a fee from participating institutions. Another older one is the Indiana HIE. Epic has their own system called Care Everywhere, available for a fee when you buy Epic and which was charging for each chart searched when Epic was presented to me, which links together Epic systems and some others who choose to join (for a fee which has been waived for five years, I guess under some pressure.)

Why all the flailing around? The JAMA reporter:

The highest barriers are financial and policy ones, not technical ones, noted those with expertise in health care technology.

Technology is almost never the obstacle,“ noted Indiana HIE's (John) Kansky. “It is really the governance of the data and getting organizations with different agendas to agree on what you are going to do.”

In other words, it is precisely the role of government that is not being well filled here. Government has taken a hike. Amazingly, this is the ONC goal:

...electronic exchange of health information among clinicians, health facilities,and patients should be widely available within 10 years.

Holy moley – 10 YEARS!! Let's see: Kennedy's speech, 1961; moon landing, 1969. ONC is not what we would call, let's say, overambitious.

I was interested that the spokesperson from ONC JAMA interviewed had only an MA degree, so I googled her, and got a hit from four days ago. From Healthcare IT News:

Former ONC Interoperability Manager Erica Galvez joins Aledade.”

What is Aledade?

Aledade was founded in 2014 by Mostashari with a focus on supporting independent, primary care physicians creating and or running ACOs. They provide healthcare technology, business transformation services and upfront capital.

More from Forbes:

Farzad Mostashari, who was formerly national coordinator for health information technology, raised $30 million in series B funding for Aledade. ARCH Ventures Partners led the round, with participation from return investor Venrock, bringing total funding to $35 million.
Mostashari, an internist, oversaw the nationwide implementation of electronic health records in the initial stages between 2011 and 2013. Last June, he founded Aledade to help doctor practices, as small as one, band together in so-called accountable care organizations (ACOs).
...By the end of the year, Aledade expects to have more than 100 physician practices in its ACOs, managing 75,000 Medicare patients. Practices pay a membership fee of $500 per general practitioner per month. They keep 60% of savings, with 40% going to Aledade.
In other words, the former head of ONC left to raise money for a private venture and hired a former staffer for him at ONC to come to the private venture. Let me note, however, that Aledade actually is interesting in that it focuses on private practices and has software that is designed to keep them independent of large hospital networks. Still, as they used to say, is this any way to run a railroad? But I have to say that I myself was in government early in my career, and looking around at what took place there, I decided that it was not a good place for me in the long run, much as I learned and loved some of my years there. I can't criticize others for making the same decision.

So, here is the picture. There are so many loose ends in the current non-system, it's not at all clear that this puzzle can be put together. To my mind, the Federal government has never looked so bad. They fail to look forward to see where private enterprise – here, EMR development – needs some help, some regulation if you please, to pave a road ahead so that companies can innovate and compete, but still develop a system that works together for the public's benefit. Companies themselves put a low priority on working together, as would be expected; that's why government is needed, as a convenor. Even when the Feds have a $30 billion windfall from ARRA, which would ensure they had the clout to enforce standards, they fail to exercise it. ONC continues to be meek and silly, setting a 10 year horizon for something that has governance and political impediments rather than technical ones. Government continues to fail to see, or at least talk about, the implications for the system as a whole, where some large capitalistic systems will benefit from lack of interoperability to the detriment of true competition and progress. Government accepts the role of bystander and beggar. Private companies seek to fill the government void of data coordination and interoperability. The problem of rotation of personnel from government to private business continues in health care as with other fields. Given the toothlessness and fecklessness of government, one can hardly blame them, although I'm sure there is much connivance from the private world. In fact, perhaps I am blaming the victim as I blame government. Who knows?

What I do know is, this really sucks.

Budd Shenkin

Sunday, November 8, 2015

How the Government Screwed the Health Care Pooch

When I get angry, I've learned to think, who do I think should be doing something that they're not doing? Then I can think, is that expectation valid? Sometimes it is, sometimes it isn't, and even when it is valid, thinking about it still kind of tempers my anger. At least, after a little bit. Personal anger management.

And so it is with our Federal government and Electronic Medical Records (EMRs). I complained to my brother Bob, who said, “What do you expect? Who do you think is in the government? The best? What kind of person wants to become a government employee?” He did not reference Donald Trump, but he could have, because that's one thing Trump is right about.

It's a well recognized phenomenon. When the ACA website imploded, veteran official Leon Panetta was thunderstruck: “Obama left that to the bureaucracy to do???” In other words, if it's important, make sure the government doesn't do it. Or, as I read in the WSJ yesterday about how Medicare doesn't check on the possible validity of claims before it pays them, as a modern credit card company would: “The government – yesterday's technology tomorrow.”

“The government” is a collective noun. It is composed of lots of people and agencies, and also the “advisers,” outside of government, often in academia, often in industry, who are supposed to be supplying the intellectual power. Maybe they're the ones I should be complaining about.

Anyway, here's the thing. When Obama came into office he needed to stimulate the economy with ARRA, the American Recovery and Reinvestment Act of 2009. Good, it was needed; in fact, more money than what they put into ARRA was needed, and he could have gotten more if he hadn't negotiated with himself, which I notice he has stopped doing, his current philosophy being, “I don't give a fuck.” Good; nice philosophy. But I digress.

So the health care “experts” thought that a program that put Electronic Medical Records into hospitals and medical offices would be just the ticket – marry the needs of the economy with the needs of health care. They claimed too much for what EMRs would do, they didn't really know much about them, and they didn't think they needed to know much about them. They had their theories, they had their analogies to other industries, they didn't need to get their hands dirty actually seeing what these programs were like. Details, details. The fact that EMRs generally suck didn't have to be addressed. The fact that it would slow down rather than speed up health care didn't have to be addressed. The fact that EMRs turned doctors into data input clerks didn't have to be addressed. Those were mere facts. In theory, it was brilliant.

My San Diego colleague Stu Cohen pointed out that the now-billionaire owner of Epic, Judy Faulkner, the major EMR system adopted in America, was a major Democratic donor and on the committee that inaugurated HITECH, the ARRA EMR program. A little self-dealing, seems like. What else is new?

When the Feds shelled out the money to get these EMRs into hospitals and offices, they unforgivably stupidly or cravenly or somethingly, (I don't know really how it was done so I don't know who to hate) didn't mandate that all the computer programs be standardized enough to be able to talk to each other., which is called interoperability. So here they all are all over the place, and they can't cooperate with one another.

In the words of North Carolina colleague Graham Barden:

What is even more WTF is that the Rand Report that was often quoted as saying how EMR’s were going to save vast amounts of money for the country, stated in the paragraph just above the often quoted figure something to the effect, “Once interoperability is achieved, ….” Unfortunately our 20 something leaders either did not read that paragraph or did not understand the big word…”

And more from Pasadena colleague Glenn Schlundt:

Over the years, there certainly appears to be a recurring and persistent, not improving theme about EMRs. Whereas a few people on this list sing their praises, and can't appear to imagine life without them, a significant fraction of posts speak to legitimate, persistent, and meaningful limitations that many EMRs presently appear to impose.

Here's my "off-the-top-of-my-head" list:

1. They don't talk to each other, so communication between them is essentially impossible.
2. They are absurdly expensive, and the return on investment on balance, while debatable in some instances, is pretty negligible.
3. They are expensive to maintain. The initial capital costs lead to monthly maintenance costs.
4. They are inefficient in that it takes doctors longer to do the same tasks, usually with little if any apparent benefit to the quality of clinical care provided.
5. Liability costs resulting from security risks, including HIPAA violations, may be considerable.
6. Some young physicians are declining to take employment positions where they will have to use hospital EMRs because they are unwieldy, inefficient, and time consuming (this from a
    previous post to this listserv).
7. They purport to allow doctors to collect data that will improve both patient care and cash flow, but other than auto-correcting doctors' previous undercoding, or allowing other doctors
    to purportedly "game the system" by clicking a few additional boxes,  well-publicized data or evidence for clinical or economic improvement appears lacking. Little convincing data suggests
    the financial benefits of EMR outweigh its costs, when measured directly, indirectly, or both.
8. They can weaken the doctor-patient relationship because some patients feel the doctor spends so much time looking at their computer screen, the doctor does not even know what their
    patient looks like. This has been used as a selling point for telemedicine, where, in a bizarre and peculiar twist, patient have reported they prefer telemedicine because "at least the doctor
    knows what I (the patient) looks like."
9. They can alter the way doctors are audited to include only data that supports downcoding or an insurer's interpretation of a clinical encounter based on data that is displayed. Several
    members of this listserv have noted that the medical decision making portion of their documentation, as measured by the software they use, drives payments, even when this does not
    appear to be the only or best criteria for medical work using CMS criteria.
10. The companies that sold or maintain them can decide to go out of business or change the terms of their contracts, creating additional access and liability issues

When one takes a Corporate Management of Risks in law school, one is taught that one frequent source of liability is over-commitment to a project or plan based on the time, effort, and resources that have already been invested. In short, one is taught that people and organizations find themselves in hot water because they are unwilling to objectively re-evaluate projects as they progress. At some point it is time to cut bait.

I'm not suggesting there is no role for EMRs; clearly in some cases they work, and work well. Nor do I mean to be either indiscreet or provocative here. What I don't understand is why, if having adopted a system that clearly has so many flaws and that it impedes efficiency, tarnishes one's balance sheet, increases liability and diminishes quality of life, why, at some point, the whole system doesn't find its way into the trash, and doctors don't just go back to the good ol' days (which, based on the animus transparent in many of the posts on this list, really are perceived as the good ol' days), until something genuinely better really comes along?

Why not just go back to paper for the time being?”

OK, that's bad, there was self-dealing, the programs are not ready for prime-time, and they don't talk to each other. Lots and lots of waste, and worse.

But, here's the deeper point of this post (sorry it took so long to get here.) There is an under-appreciated implication of the horrible omission of not requiring interoperability, which is brought out in a recent post from Boston health policy colleague Paul Levy:

Short look back: before the computer revolution, one of the foremost thinkers in health care organization was economist and ex-McNamara Whiz Kid Alain Enthoven. He proposed the model he called Managed Competition (MC). He envisioned many ICNs competing for business. He was a corporatist, and in those pre-computer days, the transaction costs of communication and information gathering were intimidating. Managed Competition was an attractive model.

The current incarnation of MC is called the Integrated Clinical Network (ICN) model, examples of which would be Kaiser, or the VA. In an ICN, when you are in the network, that's where you stay. Need a referral to neurosurgery? Off you go to the Redwood City Kaiser, where all their neurosurgery gets done. Need a neuro MRI? Off to Richmond, or wherever. It's all in-network. In theory, such specialization is good, more efficient, expert. But, if you are a clinician in a network, you don't have overt competition. The plan's members are a captive audience. You have to do well, maybe, but you really don't have to be the best. Each individual unit is shielded by the collective – the patient might choose Kaiser knowing there are some substandard units she hopes she won't need.

But now, with computers, another patient-centered model is possible. This is called the Centers of Excellence (COE) model. In a COE, you come to me, your primary care doctor. You need a referral. My job as your doctor is to get you to the very best place possible, and I have not just my in-network choice, but a full choice of the whole Bay Area, or even beyond. Several centers are competing for my business, and that's good for the patient. If they start to screw up one way or another, they will lose my business.

This is the real significance of interoperability of EMRs. The Mother Jones article cited above by Stu, written by an author with complex medical problems, points this out in great detail. To deliver the best care most efficiently, the primary care doctor and the referral unit need to have the same medical record in front of them. If the EMRs are interoperable, problem solved, COE feasible. If not, we have the same old difficulties of coordination, cooperation and efficiency.

If the EMR is not interoperable, referral within the system is easy, but referral to a possibly superior or more convenient center for x or y specialty is hard. But if patients might suffer, the same is not true of the ICN institutions at all. Although our economic system is based on competition, the role of individual corporations is to try to escape that discipline as much as possible, by combining with others to form oligopolies, or by fencing customers in to your network (as Apple does, for instance). So Epic won't cooperate with other systems – they sure don't want a small computer company that is great for a primary care doc networking into their system – God forbid! And for the integrated networks like Kaiser as well, interoperability is a threat rather than an opportunity.

The consequences of the Feds' decision not to require interoperability when they allocated $30 billion for EMRs are clear, then? This decision skewed the fight in favor if the ICN model, when the COE model has much to recommend it from the patient's point of view.

Maybe this situation is an example of LUC – the Law of Unintended Consequences. But “unintended” doesn't mean “unforeseeable.” The Feds, and if not them then their advisers, should have foreseen this consequence. Is this an example of inside dealing? Is this an example of selling out to corporate interests? Were the Feds convinced by others that “it would be too difficult” for them to insist on interoperability, and they settled for something less so they could “get the money out there?”

I don't know how it came about. They say when you need to choose between ignorance and malice, choose ignorance. Maybe. But at this time in our history, there is a very strong trend of increasing concentration within every industry, of corporate domination of government, of weaker and weaker government ability, and more and more violation of individual rights and welfare.

Sometimes it makes me mad, and sometimes it makes me sad. But either way, it sucks.

Budd Shenkin