In the old days, say about 20 or 30 years ago, people went to the doctor to get their questions answered and their problems solved. The doctor “owned” the knowledge. We paid to get it. Seemed to work…or did it? We have come to understand that, even today, the answers you get and the treatments that are recommended may be based more on your doctor’s opinion or belief than on hardcore evidence.
The cover story in this week’s Business Week Online, titled “Medical Guesswork,” gives a nice summary of the problem and helps to describe one possible solution. The article points out that only about 20-25% of medical care has been proven effective. We call that type of care “evidence-based medicine”. One or more randomized clinical trials usually have to be performed and published in reputable peer-reviewed medical journals before we consider a medical intervention evidence-based. Too often, clinical practice is based on studies that don’t meet that rigorous standard. Or, the practice may not have been studied at all (for example, the efficacy of hand-washing). That doesn’t mean a medical intervention doesn’t work (although it may not), rather it means that doctors and other health professionals are getting tests and recommending treatments based on something other than solid evidence.
The article in Business Week focuses on the work of Dr. David Eddy, the smartest doctor I have ever met. In fact, David is not just a doctor. He also has a Ph.D from Stanford’s Engineering-Economics Systems Department. David, his colleague Len Schlessinger, and others, with funding from Kaiser Permanente, applied their knowledge of medicine, engineering, mathematics, and economics to the development of a sophisticated computer modeling program, they call Archimedes. Using this program, they can simulate populations of virtual patients they use to test various medical interventions.
Taking on diabetes
One of the first clinical problems they addressed with Archimedes was diabetes. Diabetes is a complicated condition that is associated with high blood pressure, abnormal lipids, cardiovascular disease, stroke, kidney failure and blindness, and amputations, amongst other things. Many doctors and patients focus the bulk of their attention on lowering blood glucose levels. Now that is important, but it turns out that it isn’t the only thing or even the most important thing that doctors and patients with diabetes can do. Archimedes simulation was able to demonstrate that treatment with three medications—all available in inexpensive generic formulations—were just as important, and may be even more important than blood glucose control. Those medications are
- A cholesterol-lowering statin
- Drugs called ACE inhibitors (angiotensin-converting enzyme inhibitors, a common blood pressure lowering drug).
The benefits of these medications on the simulated outcomes of diabetes patients were greater than what can be achieved with aggressive glucose control alone.
This is a big deal. Archimedes provided “evidence” in a 30-minute simulation that would have taken years and millions of dollars to achieve using traditional clinical research approaches. Although we are still not 100% certain these simulations are going to mimic reality perfectly, Eddy and colleagues have published papers in peer-reviewed medical journals that demonstrate that Archimedes can reproduce clinical trial results.
David is a man with a mission. He wants to change the way we do medicine for the better. He says he hopes that in 10 years, “no one will make an important decision in health care without first asking: ‘What does Archimedes say?’“