Tuesday, October 6, 2009

A Review: Chaos and Organization in Health Care

There has been a great number of books analyzing health care over the past year and one suspects that this wave will continue. The book by Lee and Mongan entitled Chaos and Organization in Health Care (MIT Press, 2009) is one of the more recent. The premise of the book appears to be that the delivery of health care has problems because it is in an organizational state of chaos and if order is restored by the means proposed by the authors then all things will improve.

On page xi they specifically state that "the solution is organization" and it is from this assertion that they continue to build their argument. On page xiii they assert their proposal that a "tightly structured delivery organization" is the ideal and they proceed to use several examples throughout the book. Before continuing, I would introduce an interesting historical observation. When I spent time in and around Longwood Avenue, the Harvard Medical School area, in the 1960s, I could actually park my VW in the lot in front of one of the hospitals. By the late 1980s I had to use a multistory garage, for what I thought was a great fee of $8. Last week I used the gigantic subterranean parking edifice for $30. The authors seem to recommend that the patients come to them, where they are collected as a group, but the vignette on parking just is the tip of the exclusion iceberg. It is quite difficult to get patients to trek to a single location for intermittent or routine care, they are all too often difficult to get to, especially for a patient who would then have to take great time from an already pressured schedule. The answer has been the single or multi- practitioner practice.

On p xii the authors speak of team based solutions to treating Diabetes. If one looks at Type 2 Diabetes, then in the overwhelming majority of cases it is s self inflicted disorder due to obesity and diet. The disease can be cured by just losing weight, and there should be actions taken to make that happen, otherwise a cost should be applied to the patient's life style choices. Assuming that nothing can be done with such a patient is just wrong and places the costs on the rest of us who comply.

On pp 39-40 the authors begin to their analysis with the data from Medicare showing that patients see multiple physicians each year. They have a chart which shows that for Diabetes the Medicare patient sees 9 different physicians 3 of which are primary care and 6 of which are specialists. One will accept the data but one must question it. If the patient has Type 2 Diabetes, then we know that there are kidney, cardiac, neurological, ophthalmological, and possible endocrine issues, but if the average Medicare patient with type 2 Diabetes sees all of this every year then this is clearly an overload. The authors put this data out there without adequately explaining it and allow it to speak for itself. It does not do so well.

On p 47 the authors have a table which shows the tests that Internists no longer routinely perform. Let me address a few. First, the treadmill stress test is better performed by a cardiologists just in case a patient does suffer an MI while be tested. The general internist is no prepared to handle this and the insurance could skyrocket if it were taken on, as well as the insurance would not reimburse it. There are lots of reasons why not. Now for a liver biopsy. No Internist would take the risk of performing an inter peritoneal invasive surgical procedure in their office unless there was no alternative. Slicing an artery in the liver is a substantial risk.

Part II starting on p 55 is where the authors begin their proposal for organization. Their overall proposal is in Chapter 6 for a Tightly Structured Health Care Delivery Organization, on p 97 and they use the VA as an example. The essence of the proposal is that one can create a dense and tightly integrated delivery system and that one gets the patients to come to that system and because of the efficiencies in this delivery mechanism the units costs and thus total costs are reduced. This is the classic centralized architecture approach introduced into production in the 1800s. Yet one questions whether that is the sole paradigm for the delivery of health care.

In Chapter 10 on p 175 the authors detail many of the reasons for lack of change. They focus on the provider and carefully list the key barriers to any for of evolution. On p 184 they present an interesting chart, charts of this type one can find in almost any environment, that one physician in the ER ordered 40% more CAT scans then the second highest ordering physician. Rather than bemoan that statistic one should try to understand why, the devil is in the details.

On p 205 the authors appear to support the bundling of payments, a plan which has worked its way into the current health care bills (2009). Bundling is a natural outgrowth of institutionalizing health care delivery. However if one can argue for the permanence of a distributed health care delivery system, which is in proximate contact with the patients, then bundling would be just another word for institutional control and the reduction of physician autonomy and patient choice. It is a sticky issue and the authors do not do credit to both sides.

On p 229 in the final chapter the authors stress that organization of providers is essential for change. They also seem to promote the single payer system approach on p 237.

The problem is that there is no financial or business analysis of the proposals. There are many generalizations and many anecdotes but frankly not a single analysis and what would be obtained by defined actions. I return to the metric we used before, namely that for any specific disease we have the total costs being:

Total Costs=Population X Incidence X Procedures per Patient X Costs per Procedure

Thus we can look at costs as driven by the four elements above. We see a growing population and thus no influence there. We can reduce incidence. Smoking has been reduced by taxing cigarettes and thus we have seen a reduction in male lung cancer incidence. Yet on the other hand we see a massive explosion in obesity and the resulting Type 2 Diabetes. A great deal can be done on incidence. On procedures; for acute coronary syndrome and the like we now have many procedures we can do today that we could not do 40 years ago and we have reduced mortality 50% as a result. Is that good, should we perform those procedures.

Then finally the costs per procedure. The last element is what the authors seem to be speaking to. What are they and how do their proposals reduce the costs and by how much. The devil is in the details, and more importantly in the numbers. It is with the numbers that the authors come us very short. The book is much too anecdotal and way too lacking in details.