A Country Doctor Writes:

New medicines are like new fashions in clothing. They are introduced with great fanfare. Most turn out to seem fairly ordinary after a few years. Some are quickly forgotten or discarded and make us say: “What was I thinking?”

Evaluating a new drug is difficult, for the pharmaceutical and scientific communities as well as for us clinicians. It often takes years of general use before a drug can really prove its safety or usefulness.

As physicians with responsibility for our patients’ lives and well-being, we need to balance our desire to provide the best treatment with our obligation to avoid unnecessary risk. Unfortunately, many new drugs turn out to be less safe than we are told when they are first introduced. The increasingly common sources of drug information, advertisements and pharmaceutical representatives, also don’t tell you what the serious journals say about new medications. It is our duty as practicing physicians to keep up with the leading medical journals.

I may sound old-fashioned at times when I question new treatments or hold off on using them for a while, but I have seen enough new drugs hit the market and soon be withdrawn because of safety issues that were not known or understood when the medicines were first approved.

Most people still remember Vioxx, the arthritis medication that wouldn’t cause ulcers. Early on we heard about high blood pressures and fluid retention. The heart attack risk was apparently kept secret for a while before the drug was withdrawn.

I prescribed a fair amount of Vioxx, because all the other arthritis medications could also cause fluid retention and Vioxx seemed to work quite well. I had also almost lost a patient to a sudden intestinal hemorrhage from indomethacin once, so the stomach safety seemed like a valid selling point.

Before Vioxx, there was Duract, an anti-inflammatory pain medication for short-term use. It was eliminated through the liver instead of through the kidneys, like other anti-inflammatory medications. I never had time to prescribe it. I held off, because it was a new type of drug, and it was soon withdrawn amid reports of liver failure.

Some medications for Type 2 Diabetes bother me. Rezulin was the first drug in a new class, which makes the body more sensitive to its own insulin. Before I had warmed up to prescribing it, the drug was withdrawn. It was linked to liver toxicity. The two newer drugs in the same class, Avandia and Actos, seem safer on the liver, but from early on, there were concerns over fluid retention and heart failure risk. In 2007 Avandia was shown to increase heart attack risk by over 40%. Actos has so far looked safer, but I am still very nervous about it.

When the Scandinavian Simvastatin Survival Study (4-S) showed that deaths from heart attacks could be reduced by 30% with Zocor, I felt fairly comfortable prescribing it for patients with high cholesterol.  I had not really used the statin drugs that came before it, because there was no proof they reduced heart disease risk. I was still cautious with Zocor, because the cholesterol-lowering drug that came before the statins, Atromid-S, was associated with a surprising and unexplained increase in death rates by over 40%.

With every statin drug that came after Zocor, I stubbornly waited for “outcomes data” of some sort. One by one, the newer drugs proved themselves to fight atherosclerosis and heart disease, and I have ended up using all of them.

Baycol came along and made claims of having less risk for muscle damage, rhabdomyolysis, than the other members of the class. I tried it, but as it didn’t have outcomes data yet, I reserved my use of it to patients who couldn’t tolerate the other statins. Ironically, Baycol was taken off the market because it had a much higher risk of rhabdomyolysis than the other statin drugs.

I have seen many new drugs come and go, from diet pills like Redux to antibiotics like Omniflox, Tequin and Ketek during my thirty years in medicine. I also vividly remember the Thalidomide-induced limb deformities of my classmate in Junior High School in Sweden. Thalidomide was marketed there as a safe drug for morning sickness.

One of my professors in medical school said about medications for high blood pressure that they needed to have no serious or annoying side effects and be less dangerous than the condition for which they are intended. I have held on to his wisdom all these years.

A patient with an acute, life-threatening condition may be very willing to accept a certain risk if the treatment is effective. Even an unproven treatment may be the best option when older treatments are known to be risky or not very effective.

I question the value of being an early adopter of new drugs under normal circumstances, when slightly older drugs are still useful. Why use our patients for guinea pigs?

In medical school, as in any other educational endeavor, being good at test taking isn’t always the same as mastering one’s subject. Tests are easy to administer and their scores, particularly when multiple-choice questions are used, are indisputable and ideal for statistical analysis. Most people tend to agree that there is more to being a good doctor than scoring well on multiple-choice exams. Cultural competency, bedside manner, empathy and clinical problem solving often require other kinds of skills that don’t lend themselves as easily to numerical assessments.

As practicing physicians, we are constantly evaluated, and most of the time this involves the kinds of things that are easy to measure. Nobody is measuring how many years our patients with high cholesterol, hypertension, diabetes and obesity live before their first stroke or heart attack. Instead, we are often evaluated by how many of our patients reach treatment targets – certain levels of cholesterol, blood pressure, blood sugar and body mass index, as these numbers are thought to be approximations that indicate the same thing.

Human nature makes most people pay more attention to those things we are being scrutinized for. Numbers are easy to focus on. But there are problems when the numbers are viewed and pursued uncritically.

For example, two cholesterol medications lower LDL cholesterol to a similar degree, but one is proven to offer better heart attack protection than the other (Lipitor® versus Vytorin®). When physicians “treat to target”, they sometimes don’t help their patients get healthier at all, which was the topic of my very first blogpost.

The same applies to blood pressure medications; some of them prevent heart disease while some may actually increase the risk of cardiac disease and death.

A fundamental problem with treatment targets is how they are chosen. One example is the blood pressure target of 130/80 or less for diabetics. We have been held to this since 1992 as if it were handed down on the stone tablets along with the Ten Commandments. The UKPDS study in 1998 showed that lowering blood pressure for diabetics to a mean value of 144/82 reduced their cardiovascular risk. No study has actually proven that a blood pressure lower than 130/80 is ideal for diabetics with heart disease, and some have shown that pressures below 130/80 are linked not only to higher rates of serious medication side effects, but to an increased risk of death.

Yet I doubt the guidelines will change any time soon just because there are serious questions about their validity. Physicians who balance their professional judgment against the simplistic guidelines will continue to do so at their own peril.

An example of things that work, but cannot easily be measured, and therefore won’t be used to judge physicians’ performance is what diet our patients eat. Patients who eat olive oil have a 25% lower risk of heart disease than others, and patients aged 70-90 who follow a Mediterranean diet have a 50-60% lower risk of dying from heart disease and cancer than patients who eat a “regular” diet.

The Annals of Internal Medicine published an article from the Mayo Clinic last year, titled “Glycemic Control in Type 2 Diabetes: Time for an Evidence-Based About-Face?”. The authors summarize their viewpoints:

“Some diabetes guidelines set low glycemic control goals for patients with type 2 diabetes mellitus (such as a hemoglobin A_1c level as low as 6.5% to 7.0%) to avoid or delay complications. Our review and critique of recent large randomized trials in patients with type 2 diabetes suggest that tight glycemic control burdens patients with complex treatment programs, hypoglycemia, weight gain, and costs and offers uncertain benefits in return. We believe clinicians should prioritize supporting well-being and healthy lifestyles, preventive care, and cardiovascular risk reduction in these patients. Glycemic control efforts should individualize hemoglobin A_1c targets so that those targets and the actions necessary to achieve them reflect patients’ personal and clinical context and their informed values and preferences.”

The more pressure the pharmaceutical industry, insurance companies and healthcare administrators are under to prove the value patients get for their healthcare dollar, the more pressure we physicians will be under to adhere to numeric targets that others have chosen for us. And the more we concentrate on the numbers we are measured by, the greater the danger we won’t devote enough time and energy to doing the equally or more important things that nobody has figured out how to measure yet. We are at risk of acting like immature students, acing the multiple-choice questions but failing the hands-on clinicals. And this time our patients are not actors or volunteers, but sick people who come to us for help and advice in fighting their diseases.