Do studies of comparative effectiveness have a negative impact on the relations between doctors and patients – do they get in the way of individualized treatment?

Put more positively, how does a physician take data from studies whose results are expressed in statistical terms for a group of patients and use it in the treatment of individual patients? Does doing so take away from consideration of the individual nature of the person who is the patient?

This is an issue that has been raised for decades – probably since the 1930s and certainly since the 1950s.

It is a little bit strange, since all of the knowledge base of medicine is general knowledge accumulated from the sum of either a doctor’s individual experience or the cumulative experience of the profession. Medical science in clinical (as opposed to experimental) medicine is a way of taking that general knowledge and making it more valid – more true to the circumstances. Making it more reliable – knowledge you can count on. Making it more replicable – knowledge that will be the same even if you garner it from a different group of similar patients.

If you want to apply this knowledge, or any general knowledge, to individual patient then judgment is required. Judgment enters anytime the general is applied to the specific. Clinical medicine is a judgment profession.

It is true that we do not teach doctors how to make judgments, but it is something that they must learn because they must do it all the time. They must decide what to do for this particular patient based on what they know about similar patients with similar diseases.

One of the most important medical advances of the 20th century has been the quality of the evidence, the knowledge base on which medical action depends. This is largely the result of rigor in the conduct of clinical studies – of the effects of treatment, for example. The double-blind clinical trial that is now the gold standard is an example.

Good clinical research requires rigorous statistical methods. Statistics is not about my patient, the individual; it is about a group of similar patients. Here again, to apply clinical research about a treatment requires judgment – is my patient sufficiently similar to the patients and their difficulties in the clinical trial so that I can use the results to guide the treatment of my patient?

Over these recent decades, good doctors have learned how to make those judgments. They have been helped by work in the field of clinical epidemiology to make more precise judgments, but ultimately their judgments are about individual patients – and individuals, as we all know are all different.

In recent years there has been much more emphasis on “evidence based medicine.” This, despite the special name, is what we hoped doctors were doing all along – basing their decisions on the evidence.

Some have gone another step and assembled the evidence so as to make guidelines, rules for treatment of patients with this or that condition. As always however, good doctors who care about their individual patients do not apply these rules unless they are sure that the rules apply to their particular patient. It is rarely an all or none phenomenon – some part of the guidelines apply and others may not.

Some rules are so general that they do apply to all patients with a particular disease. For example, it is always important to know how well patients with diabetes are controlling their blood sugar because recent evidence has made it clear that persons with diabetes whose blood sugar is well controlled do better than those with poorly controlled blood sugar. Even here, however, there are particular diabetic patients who, for one reason or another, cannot achieve good control of blood sugar. The doctors of these patients – with the help of the patients themselves — have to find an alternative solution to the problem.

The proposal for studies of comparative effectiveness have raised the fear that doctors will be forced to follow the dictates of the studies whether they apply to an individual patient or not. I don’t understand those fears. The history of medicine, recent and past, is littered with treatments or diagnostic methods that seemed good at first but when studied didn’t pass muster.

There was a time when, for bleeding ulcers, we pumped ice cold saline into the stomach. There were even machines to do this. That treatment failed a trial of comparative effectiveness – the machines were all thrown out.

Comparative effectiveness studies don’t get in the way of the doctor-patient relationship; they provide a reason for making it better. The key to good judgment in medicine is knowing the particular patient as an individual. The better the doctor knows the patient, the better able the doctor is to judge how to apply the evidence that comes from the studies.

Good, solid evidence can only be a good thing, even if it makes us scratch our heads from time to time and change our judgments. Comparative effectiveness studies also emphasize a point that is sometimes forgotten: all good medicine flows through the relationship between patient and doctor. It is within that relationship – it cannot be done in abstract – that judgment about the importance of evidence to a particular patient should be made by an individual physician.

Eric Cassell, M.D., is an internist who practiced medicine for almost 40 years. He is an Emeritus Professor at Weill Medical of Cornell University and an Adjunct Professor of Medicine of McGill University’s Faculty of Medicine. He is a Master of the American College of Medicine and a member of the Institute of Medicine. He has written extensively about doctoring, the care of the dying, suffering, and moral problems in medicine. eric@​ericcassell.​com, 917 365 5639.