The Pebble and the Avalanche

Moshe Thumbnail
Current Revolutions in Business and Technology

by Dr. Moshe Yudkowsky,

author of The Pebble and The Avalanche: How Taking Things Apart Creates Revolutions


Thu, 2007-Jan-25, 14:45

Story Marker
Medical Conflict of Interest

"Public interest" groups — I prefer the more accurate British term, "pressure" groups — continue to press for more disclosure of conflicts of interest in the medical industry. The quetsion becomes, what constitutes a conflict of interest? I've found many recent claims of "conflict of interest" to be quite thin, grasping at straws in order to generate publicity for whoever publishes the claim; but public anger over medical costs makes anyone in the medical profession fair game, and doctors are the least powerful and most plentiful targets. In any industry, there will always be a connection between people who have businesses and those who evaluate the business. The degee of separation is the issue, not whether or not there will be a connection.

But here's a new twist. The NAACP has complained that Medicare does not pay for BiDil, a medication created to reduce heart disease in African-Americans. Insurers state that BiDil is simply a combination of two drugs that are available generically at one-tenth the cost of BiDil, and therefore they prefer to pay for the generic medication instead. The NAACP disagrees, and ascribes failure to prescribe the drug to discrimination.

But here's the twist. Today's Wall Street Journal mentions but carefully avoids drawing conclusions about a grant to the NAACP:

The NAACP has received a $1.5 million grant from the drug's maker, NitroMed Inc., to improve health care for blacks.
If a doctor's group were promoting a drug under these circumstances, they would find themselves under intense scrutiny. The NAACP, as a pressure group, escapes scrutiny. Who watches the watchers?

Fri, 2007-Jan-19, 08:05

Story Marker
Insurance Numbers

One thing to remember, when confronted with a problem, is to carefully examine whether or not there's actually a problem and what that problem might be.

I'm reminded of this as I read the recent spate of stories about "fixing" health care. Although I won't discuss how to improve health care just at the moment, I do want to point out that the numbers as presented are indigestible lumps. After reading newspaper reports I can deduce that people in the US spend a lot of money on health care and some of them pay for it out of pocket rather than through insurance. But any finer measurements of health care costs are lost in the news stories, and it's impossible to determine if the plans presented by various political, business, and (as the British call them) "pressure" groups will actually improve access to health care or if the plans will simply add another layer of unwieldy bureaucracy to the health-care industry.

I believe disaggregation is a key to a better health care industry in the US — and I also believe that disaggregation of health care statistics is key to a reasonable debate over what improvements to make.

Thu, 2006-Dec-14, 08:44

Story Marker
Students "Unprotected"

A book review [registration required] in today's Wall Street Journal discusses Unprotected, a book about university students' health care.

While doctors at the university health care systems freely offer advice about general life issues, one category of problems is treated differently than the rest: sexual issues. If your lifestyle is too sedentary, your physician will have something to say; if your sexual activities put you at risk, either psychological or physical, the doctors are silent.

I diagnose the problem as disaggregation, but with lack of care to make certain that proper interfaces continue. In other words, while the universities may choose to treat some forms of risky behavior as different from other risky behaviors, this doesn't mean simply cutting the behavior loose from medical supervision entirely. If you take something apart to make it work better, you must make certain that the parts still work together afterwards.

Wed, 2006-Nov-22, 13:56

Story Marker
Quiz: Business, Medicine, and Government

This morning I read an article about health care and the migration of hospitals away from city centers into the suburbs. As I was reading it, I put aside my usual highliner and pulled out a blue one to start underlining every phrase in the document that discussed the role of government. When I was finished the text was awash in blue ink. Just about every major business decision a hospital makes is second-guessed by local, state, and Federal officials, and then subject to protests and lawsuits from anyone who is unhappy with the decision.

The aggregation of all business decisions into the hands of government produces completely predictable results.

Fri, 2006-Oct-13, 09:27

Story Marker
Medicine: Safety and Efficacy of Drugs

United States Senator Grassley suggests that the U.S. Food and Drug Administration (FDA) should disaggregate the function of monitoring post-approval drug safety from the process approving drugs. With an independent drug safety office, the people who approved the drug in the first place would not be tempted to overlook safety problems that might reflect poorly on the process.

While it's tempting to accept this disaggregation, not everyone agrees. The New York Times notes that even though the FDA needs reform (and what government agency in the world does not?), the disaggregation of safety considerations from the approval process would be a bad idea.

This proposed change to the FDA management structure poses an fascinating problem. On the one hand, disaggregation of authority — of approval process from safety monitoring — would likely engender trust in the FDA. On the other hand, here's an interesting lesson from high-school science classes: certain things can't be disaggregated. "For every action, there is an equal and opposite reaction," otherwise known as Newton's Third Law, and that's a connection that simply can't be broken. As much as we'd like to, scientists and engineers can't disaggregate their way out of basic tradeoffs imposed by nature.

In a similar way, drug approvals must trade off efficacy versus safety; every known drug has side effects, and the approval process balances the harm of side effects with the benefits of medications. If the office that determines efficacy disaggregates from the office that monitors safety, it may become extremely difficult to approve any drug or keep it on the market.

Mon, 2006-Jul-31, 08:16

Story Marker
Health Care Insurers Run Amok, Abandon Disaggregation

A weird, almost incomprehensible article in the Wall Street Journal explains how health insurers now consider each and every business account as a separate entity and demand that each of them provide a healthy profit:

His [30-person company's] medical costs were roughly half of the premiums he paid between 2002 and 2004, according to his insurance data. But last year, after an employee was severely injured in a highway accident, WellPoint's Blue Cross Blue Shield of Georgia boosted premiums by 30%. It had asked for a 41% increase but came down after Mr. Perkinson agreed to make employees pay even more of their bills out of pocket.
Insurance allows you to disaggregate risk from a small group and spread it to a larger group; the idea is that if one person in a small company has an accident, the experience doesn't bankrupt the individual or the company. Yet here we see a health insurance company, in pursuit of higher profits, abandon this idea and punish the small company for a statistical fluke. What's the point of insurance is risk isn't disaggregated?

Although that's bad, it gets worse. Health insurance companies are now abandoning their role as arbiters of health care finances and quality, and pushing decisions to the consumer:

[E]xecutives believe the recent trend toward giving consumers more data on price and quality, with the goal of steering them to cost-effective care, will help bring health-care expenditures under control.
Allow me a moment of cynicism to translate this: by imposing harsher terms on consumers, they will be forced to scrounge for lower-cost care on their own.

The current insurance-driven health care system imposes an incredible price regime; I've seen a thirty-thousand dollar hospital bill for the uninsured magically transformed into only a couple of thousand dollars when an insurance company picks up the tab. The combination of cost squeezes by insurers and government-run programs have pushed the highest costs to the people who can afford it the least. While consumers, businesses, and medical programs go begging, the insurance business continues to hum along and generate profits. I have to wonder how long this state of affairs can last.

Thu, 2006-Jun-15, 18:30

Story Marker
Civil Servants to Control Ambulances?

If you're in an ambulance and your life is on the line, do you want civil servants in a "regional control center" to decide where to route your ambulance?

According to numerous news reports, the alternative — letting the ambulance crew decide where to take you — isn't any better: ambulances find they must divert their patients because of overcrowding and staffing issues at emergency rooms.

The proposed solution is to disaggregate the authority to decide where to go from the ambulance crews and give it to a regional authority that has an overall picture of the state of the emergency rooms:

The panel proposed regional ER systems, which would manage the flow of patients much like airports direct flight traffic. Patients would be quickly transported to emergency departments that are best equipped to handle their conditions. Stroke victims would go to stroke centers, for example.

I personally find the proposal makes me nervous. In my analysis, government intervention is the source of many of the problems the US health care system faces today, including the sad state of the emergency rooms. Further government intervention isn't likely to cure the problem.

Thu, 2006-May-04, 06:56

Story Marker
Mad Cows and Angry Ranchers

Ranchers and farmers continue to reject government proposals to track all farm animals in the United States. The government wants a comprehensive system to track and identify the movements of each and every farm animal to help isolate instances of mad-cow disease.

Farmers, ranchers, and others oppose the initiative for a variety of reasons. For example, rodeo competitors would have to inform the government each time they take a horse from one location to another — a rule they view as nightmarish to implement and an intrusion of privacy. But perhaps the biggest worry is lawsuits: if the government traces the source of infection back to a single farm, lawsuits would surely follow. The owners of a destroyed herd would be sorely tempted to file lawsuits, and if some human caught mad-cow disease, everyone who ever came into contact with an infected cow would find themselves in court.

One answer to the problem of lawsuits is strict disaggregation of identity from liability: grant legal protections to anyone who participates in the identification program or cooperates with post-infection investigations. Laws restrict access to hospital and medical records; maybe these veterinary records deserve the same protection. But restriction information would be hard to do, and since the destruction of a herd of cattle is hard to conceal, participants in the program should be immune from lawsuits. (And that immunity should have some teeth, because otherwise lawyers will file court briefs arguing that the law itself is invalid, or that a particular participant wasn't really in compliance and therefore shouldn't be protected — and defending against those allegations can also be hideously expensive.)

The current plan is a government-imposed mandate that will itself be litigated in court and grudgingly implemented — not the optimal response to a major risk to the health of people and the health of an industry.

Tue, 2006-Feb-28, 09:18

Story Marker
The Mystery of Medicine

The same terrific series of disaggregations that brought us amazingly cheap DVD players, computers, and automobiles has been progressing for some time in the health care industry. Instead of a single physician interacting with a single patient, health care is now an industry with middlemen, specialists in providing ancillary services, large specialized hospitals, and similar ecological niches. Instead of prices falling, however, health care costs continue to rise (as measured in constant dollars). Why is that one benefit of disaggregation missing?

I deliberately left discussions of health care out of the book; I'm working on an essay or two on the subject. But I can't ignore a terrific quote in today's Wall Street Journal that explains a great deal about what's wrong with health care in the US today:

"Republican delusions that health care can work like any other market apparently know no bounds," says Robert Berenson, a senior fellow at the Urban Institute and a top Medicare administrator under President Clinton. "They now even extend their notions of an ownership society to people in their last months of life."
And it's not that Berenson is right — it's that he's so completely wrong. Berenson and his ilk attempt to aggregate all authority about health care decisions, costs, and resources into Federal and state agencies. The more they succeed in aggregating authority, the higher prices rise. Fighting disaggregation means that you lose the benefits it would bring, and the cost reductions we'd ordinarily expect to see are missing in action.

Thu, 2005-Dec-29, 08:15

Story Marker
UK Car Insurance Company Has Its Own Clinic

The Norwich Union car insurance company in the UK opened its own clinic to treat whiplash patients. They offer physical therapy focused on rapid recovery. As it's their own clinic they can offer treatment without requiring any payment from the injured party, which cuts down on paperwork and increases customer satisfaction.

Of course the usual problem occurs: they've nibbled away at disaggregation of authority. The people who authorize the course of treatment, who determine what therapies are effective, and make other medical decisions all work for the people who foot the bill and have an enormous incentive to decrease costs. By removing the usual separation they've incurred some distrust.

It's an interesting development to watch. If this innovation comes to the US, I would expect to see an outside organization certify that treatment is correct and appropriate.