Using Less Treatment

Up to one-third of medical treatment does little or no good for the patient [see the end of my blog entry “Slowing the Growth of Medicare and Medicaid Cost”].  Several accounts have recently become available illustrating our healthcare system’s tendency to overtreat.  For example, see the Atul Gawande’s article at

or the radio program available at:


Why do doctors order treatments that don’t benefit their patients?  There are at least four reasons:

1.  Profit seeking – Doctors spend their long medical education learning how to treat patients.  Our fee-for-service payment system then gives doctors a strong incentive to use this training and provide additional services.  Medicine is slowly changing from a healing profession to a business.  Read the Gawande article.

2.  Ignorance – In many cases, doctors don’t have adequate knowledge about the cost-effectiveness of different treatment options.   Without clear guidelines, they often will, in good conscience, order the procedures they are trained to provide that earn them the most income.

3.  Patient demands – Patients often ask their doctors to treat ailments for which there is no cost-effective treatment.  They sometimes demand treatments that their doctor knows are not likely to be effective.  These requests are more likely when a patient has insurance and doesn’t have to bear the cost.  Doctors often have a difficult time saying no, especially when their income increases if they give in.

4.  Malpractice lawsuits – To protect themselves from malpractice lawsuits, doctors may order additional tests and treatments.

The best way to reduce healthcare costs and improve the quality of care in the U.S. is to change doctors’ incentives so they will stop ordering ineffective treatments.

I will devote the next blog entry to discussing the last two reasons listed above and will argue that, while they are important, they are not a primary cause of the high cost of healthcare.  Profit seeking and ignorance are the primary causes of over treatment.


Actions to reduce ineffective treatments

A. Improving Payment Methods and Encouraging Self-Monitoring by Groups of Doctors

Dr. Arnold Relman has written a book, A Second Opinion, and many articles, e.g.,

in which he argues that medical costs in the U.S. will be reduced and patient outcomes improved when more doctors start practicing as salaried members of privately-owned, multispecialty groups.  The Mayo Clinic is probably the best known such group.  The cost of care at the Mayo Clinic during the last six months of life is half the cost of such care at, for example, the UCLA medical school.  Several group practices similar to the Mayo Clinic exist around the U.S., including the Cleveland Clinic, Kaiser Permanente, the Billings Clinic and the Health Group in Seattle, and all have been successful in lowering costs.  For a video examining these groups, watch

As a salaried member of a group practice, a doctor’s income depends less on ordering unneeded treatments and he has more opportunity to discuss and coordinate care with his colleagues.

Making group medical practice the norm will take many years to accomplish.  In the interim, two intermediate steps should be taken now.


1. Accountable Care Organizations

Doctors in an area should organize into “Accountable Care Organizations” (ACO).  ACO’s would hold regular meetings at which the member doctors review how medicine is being practiced in the local area and provide feedback to doctors who are either overtreating or undertreating their patients.  The ability to review local practice standards would be facilitated by the widespread use of electronic medical records.  The formation of ACO’s could be encouraged if Medicare and private insurance companies shared part of any cost savings with hospitals and the doctor members of ACO’s.

2. Increase the use of “capitated” payments and “bundling” of payments

Medicare and private insurance companies should move away from fee-for-service payments and pay part of a patient’s care with a single yearly payment to the group of doctors providing his care (“capitation”).  With part of a patient’s care covered by the fixed payment, reimbursement for routine tests and procedures could be reduced—reducing the incentive for extra treatments.  “Bundling” of payments would provide hospitals and doctors with a lump sum in advance to provide care for a particular medical condition.  “Bundling” would encourage doctors and hospitals to improve quality since repeated treatments or readmissions to the hospital would not be reimbursed.


B. Providing Doctors with Better Information about Treatments

For a surprisingly large number of illnesses, doctors currently have inadequate guidelines about the most cost-effective treatments.  Prostate cancer is a good example. Many treatments for prostate cancer are available, including radical prostatectomy, several types of external radiation, brachytherapy (radioactive seed implants), and “watchful waiting.”  The cost of these treatments varies enormously—from very little for “watchful waiting” to over a hundred thousand dollars for newer types of external radiation treatment.  For early stage prostrate cancer, no consensus exists on whether one type of treatment is any better than another type.

Another example of our failure to provide doctors with adequate information is medical devices (e.g., defibrillators and artificial hip joints).

Healthcare costs in the U.S. could be significantly reduced by funding more research on the effectiveness of different treatments and by giving doctors clearer guidelines on which to base their recommendations.


Slowing the Growth of Medicare and Medicaid Costs

Without major changes, the federal deficit will explode and U.S. national debt will reach dangerous levels within the next twenty years. Any realistic plan to balance the federal budget must include two main elements.

1. A significant reduction in the growth rate of federal healthcare costs — primarily Medicare and Medicaid.

2. An increase in taxes.

The more we can slow the growth of Medicare and Medicaid costs, the less we will need to increase taxes.

Medicare and Medicaid cost growth can’t be slowed without similar changes in the private healthcare system

Attempts by Congress to slow the growth of federal healthcare costs have been unsuccessful primarily because Medicare and Medicaid are so closely linked to the rest of our healthcare system. All doctors and hospitals treating Medicare and Medicaid patients also have private patients. If Medicare and Medicaid payments are cut much below the standard charges to private patients, healthcare providers will stop seeing the poor and elderly. If the government restricts treatments available to Medicare and Medicaid patients, Congress will quickly receive complaints that it has created an inferior healthcare system for the poor and old. Consequently, slowing government healthcare costs in the long run is not politically possible without reforms that also apply throughout the U.S. healthcare system.

Slowing the growth of healthcare costs

Between 1975 and 2005, per capita healthcare costs grew much faster than costs in the rest of the U.S. economy — 2.4% faster for Medicare, 2.2% faster for Medicaid, and 2.1% faster for private patients. Rising costs are now threatening to bankrupt not only the federal government, but also many businesses and private patients. How can the growth of healthcare costs be slowed? That is now the $64,000,000,000,000 question.

There are three main approaches to reducing healthcare costs: reducing the need for treatment, paying less for treatment, and using less treatment.

Needing Less Treatment

The U.S. is currently suffering from an epidemic of obesity and diabetes. Getting Americans to eat better diets and exercise would reduce chronic illnesses now and in the future. However, we all will eventually get old and die, so the main effect of healthier lifestyles is to delay healthcare costs. The studies are inconclusive about long-run savings and, unfortunately, finding ways to motivate people to change their lifestyles is difficult.

Paying Less For Treatment

Congress has passed and rescinded many laws to reduce Medicare and Medicaid payments. Government efforts to pay less have so far been unsuccessful in reducing cost growth. Increasing competition among providers might reduce charges, but is very difficult to implement because of the special characteristics of the doctor-patient relationship (I’ll have more on this in a later blog entry.) More competition among insurance companies could reduce costs and this is one of the reasons the Task Force supported a voucher plan. However, unless government takes over the healthcare system, it is unlikely that the growth in payments per treatment can be slowed enough to solve our problem.

Using Less Treatment

This is where the hope (if there is any) lies. Evidence from two sources indicates that about one third of medical treatments have little if any benefit and, in some case, may harm the patient. For twenty years, researchers at Dartmouth have been examining Medicare payment records and have published the data in the now famous Dartmouth Atlas of Health Care.

Large variations exist in the cost of healthcare in different parts of the U.S. — with some areas spending twice as much as other areas. The surprise is that patients in high cost areas don’t do any better than those that receive less treatment in low cost areas.

The second source comes from international studies of healthcare. The U.S. spends about a third more of its GDP on healthcare than any other advanced country and other advanced countries usually have as good or better outcomes. See this report from the White House, particularly the discussion starting on page 9.

Our healthcare system seems to be surprisingly inefficient. If we can make better treatment decisions, we should be able to reduce costs without affecting the quality of care. How can this be accomplished? I’ll discuss this in the next blog entry.