Stat counter


View My Stats

Sunday, July 3, 2011

Little Bets vs.Intelligent Design

I finished reading "Little Bets", a book by Peter Sims. It is sort of a conceptual cross between "Fooled by Randomness" by Taleb Nassim, "Why Things Fail" by Paul Omerod, and "Evolution for Everyone" by David Sloan Wilson. The lessons are both simple and profound.

The world is complex, so complex that no one person or entity can consistently predict and successfully plan for the future. Furthermore, complex systems do not develop as a consequence top down master plans but instead they develop as the product of many little bets. Most of these bets are losers but if enough bets are placed, some will pay off in a major way. This is how complex biological systems evolve over time. The information systems in biology are genetic based and the imperfections in DNA replication result is a consistent generation of little bets based upon mutations. Most result in either nothing or at least nothing good. However, some of the mutations result in some new characteristic which provides the organism involved with an enhanced ability to survive and pass on their newly acquired characteristics. The scenario is accepted by virtually the entirety of the scientific community. Intelligent design proponents are intellectually marginalized within the scientific community. However, they are not so marginalized when it comes to non-biological complex systems.

The thinking is not so widely accepted when applied to non-biological complex systems, although it is the focus of David Sloan Wilson's work. I have to recommend his book "Evolution for Everyone" since in my opinion it is so effective at taking this these concepts and showing how intuitive they are when moved to non-biological contexts. Unfortunately, the importance of these principles as they relate to complex social and economic systems has not be widely embraced.

This conceptual blind spot has real and potentially terrible implications as they relate to health care delivery and reform. If one believes that complex systems can and are the product of intelligent design, then it makes sense to take the approach that humans can and should design such entities. This assumption will find little support from the study of virtually any complex system. Perhaps the strongest empirical evidence that these assumptions are flawed comes from the study of economics in the 20th century. At the start of the century, there may be some disagreement regarding the extent of economic planning the state should undertake, there was consensus that states should seize "The Commanding Heights" of the economy. By the end of the century, the consensus had changed.

Sims brings an interesting perspective on why centralized, planned economies failed. He believes it was because they could not take on little bets and because they could not take on little wagers, these economies were unable to innovate.

Everywhere I look in health care I see the same problem. Innovation in the health care system is stymied by the inability to make small bets. Just look at the most recent proposed ACO regulations. While there is lip service payed to the pursuit of innovation, there is nothing in these regulations which can be viewed as fostering little bets.

Medicare is the quintessential innovation stifling entity. At the most basic level a physician is either all in or all out. The DME and IME monies Medicare supplies to institutions for training residents is predicated on the same assumption. Take any money to support any trainee under a given tax ID number and all trainees under that same tax ID number are required to comply with any and all Medicare rules, not matter how onerous or ambiguous.

The payment, regulatory, and liability systems in general obstructs taking little bets. What is defined as financially valuable in medicine is defined by a small cabal via a political process. We are about to substitute one new cabal (IPAB) for a previous one (RUC), but the basic architecture is the same. New or innovative approaches to delivering value to patients are remarkably constrained by this evil triad. For example, I could do follow ups on many of my patients using simple electronic tools (telephone and emails) but HIPPA rules make this legally unwise, payments rules make this financially unrewarding, and the liability climate makes this legally untenable.  I could experiment but only by making a big bet...withdrawing from the mainstream of health care delivery.

The present culture and tools used in the practice of medicine is also inherently hostile to the little bets approach.  Most little bets fail and failure, even low stakes failure is a problem within medicine. To make matters worse, how we practice in the ambulatory world is within a realm of rudimentary information tools. We can place the bets but we may never know if they pay off or not. Thus we fall back on what we can measure...money. We end up doing what the payments system rewards us to do and since no amount of innovative practice appears to change what is profitable, we become locked into doing things because they are what is required to maintain financially viable practices.

Using politics to change this substitutes a few huge bets for millions a little bets. It substitutes regulatory complexity for market simplicity. Based upon the experience of the planned economies of the 20th century, which tried similar approaches, there seems to be no reason to believe that our attempts at intelligent design will be any more successful, unless one takes the tack that now things are different and we are just smarter than our predecessors.  That sounds like arrogance and hubris to me.

 

No comments:

Post a Comment