Search This Blog

Follow by Email

Thursday, August 25, 2011

Preventing health catastrophes--what works?

Today in the Journal of the American Medical Association David Jenkins MD and colleagues from Toronto, Ontario reported that certain dietary interventions really do help reduce cholesterol levels. Most patients believe, because we have told them, that improving their diets by eating less fat will significantly lower their cholesterol levels. According to a really well done study published in 1998, that is not true. Combining such a diet with exercise is helpful in reducing cholesterol, and probably also reduces many other bad health outcomes, such as diabetes, obesity and death. In Dr. Jenkins' study, however, patients were not told just to eat less fat, they were also instructed in what foods to add to their diets to reduce cholesterol. The study participants were instructed to eat a vegetarian diet and to add plant sterol containing margarine, soluble fiber (such as psyllium, oats or barley), soy protein products such as tofu and soy milk and peanuts or tree nuts. Simply being nagged to eat this way (2 sessions, total of 1 3/4 hours) lowered the cholesterol over 20 points (13%). The low fat diet, which was the control group, made almost no difference. Patients who actually followed the study diet had the best results, which is at it should be.

But having high cholesterol is hardly a health catastrophe. Heart attacks are catastrophes, and every year over a quarter of a million people die of heart attacks. We know that elevated cholesterol levels are a risk factor for heart attacks, but there has been no study that I can find that proves that lowering cholesterol levels by diet reduces heart attack. Perhaps it doesn't hurt, but even that is unclear. Cholesterol lowering drugs, such as the pharmacological blockbuster lipitor, definitely do lower cholesterol and lower risk of dying of a heart attack, but that may be due to any of a number of mechanisms, including lowering inflammation.

One of the most devastating catastrophes of aging is a bone fracture. As people age, bones become weaker and with relatively minimal trauma, such as with a low impact fall, can break. The most significant of these fractures is of the hip. Most hip fractures are fatal if untreated, and surgery to stabilize them is not a small thing. In the year following a hip fracture 15-20% of patients die, and many more require long nursing home stays.

People with osteoporosis are most likely to sustain hip fractures, and osteoporosis is increasingly common as our population ages. It is common to get a bone density study done after menopause to identify osteoporosis and physicians are often asked to prescribe medicine to strengthen bones when the bone density study shows that the risk of fracture is increased. Medicines such as fosamax (alendronate, now generic) can increase the bone density and reduce the risk of fractures, but all medicines for osteoporosis have side effects,  from esophageal ulcers to unintended bone fractures and jaw bone death. The side effects are infrequent, but that is no consolation to the occasional patient who gets the side effect. Because the medicines for osteoporosis are hard to love, we physicians often recommend to patients with waning bone density that they start calcium and vitamin D supplementation. These are inexpensive and pretty much natural, but there is no evidence that I can find that they actually work to prevent fractures. A 1998 study in the New England Journal showed that supplementation with calcium 1000-1500 mg  and vitamin D 400 IU daily slightly increased bone density, but did not reduce hip fractures. On the strength of that study, we physicians temporarily stopped recommending calcium and vitamin D, but we are back at it again, based only on some vaguely applicable studies of vitamin D alone in nursing home populations.

Of course, getting scientific proof of what works and what doesn't is closely attached to both funding and practicality. Drug trials can be performed in a double blind, placebo controlled manner, the most scientifically trustworthy design, whereas lifestyle and diet changes cannot be tested that way at all. It is very expensive to carry out a well designed trial of any sort, but big pharma has that kind of money, and can afford to spend it on science since the return on a positive study is very significant (Lipitor (atorvastatin) earned Pfizer 9 billion dollars in 2009).

What is a person to do? Should we as physicians stop giving health advice that is of dubious accuracy? Since people sometimes actually do listen to us and do what we recommend, should we work harder to produce some kind of guidelines for good health behaviors that will be easy to understand and of minimal complexity?

No comments: