Today the American Cancer Society released figures showing a decrease in cancer deaths in the United States in 2004 as compared to 2003. (See full report here) Says Reuters in its coverage of the announcement:
…the big decrease shows that not only has the death rate from cancer been reversed — but it has been reversed so much that fewer people are dying, even though the population of elderly people, who are most susceptible to cancer, is growing.”Cancer death rates have been declining for a long time. The declines have now outpaced the growth and aging of the population,” Elizabeth Ward, director of surveillance research for the American Cancer Society, said in a telephone interview.
She said a small decline seen in the previous report had grown considerably, showing the trend was real.
Decreases in smoking may be a major factor, Ward said.
I would suspect that the huge decline in smoking is behind this falling off of cancer deaths. Since 1955 the number of males who smoked has decreased by almost 60% and the number of females smokers by 35%. Since lung cancer is one of the most common cancers and is indisputably related to smoking, it would seem reasonable that as people wean themselves from this noxious habit (or never start in the first place), deaths from cancer would be on the decline. And not only are deaths from cancer on a downtrend, so is the incidence of cancer. Looking at the rates of new cancer cases in 1997 and comparing them to those in 2006 (adjusted for population) we find that the incidence of lung cancer has fallen by about 18%. So, it is reasonable to conclude that the fall off in smoking would translate into fewer cases of lung cancer, and it has.
Now let’s look at heart disease. If we look at the American Heart Association (AHA) statistics, we are told of the tremendous decline in death rates due to heart disease over the past couple of decades. Based on what the AHA recommends to reduce the risk for heart disease, we learn that we should eat less saturated fat and cholesterol, reduce our serum cholesterol levels (with drugs, if necessary), reduce our blood pressure, quit smoking, and exercise. Do these things were told and deaths from heart disease will fall even further.
On page 8 (large pdf) of the Heart Disease and Stroke Statistic–2007 Update much is made of studies showing that lowered cholesterol, blood pressure, and smoking greatly reduce the risk of heart disease. The trends for these lifestyle changes are shown in the graph below.
So, it appears that the AHA is right. As a country we’re eating less saturated fat, less cholesterol, we’re all taking statins, and many of us have stopped smoking, and it’s paid off because deaths from heart disease have dropped dramatically. But not so fast. All these AHA recommendations are to reduce the risk of developing heart disease in the first place. Since it’s difficult to die from heart disease if you don’t have it, it makes sense that preventing the onset of the disease would reduce deaths from it. Which is precisely what the AHA wants you to believe. Follow our guidelines, they tell us, and you will reduce your risk for heart disease, and, consequently, your chances of dying from it. If you don’t believe us, look at the statistics.
Harkening back to the cancer statistics, remember that both the incidence of cancer (i.e. how many people were afflicted) and the deaths from cancer decreased. From these statistics we can say that fewer people died from cancer because fewer people got it. I’m sure medical treatment improved, but primarily fewer people had cancer.
What about the incidence of heart disease? We know that deaths from heart disease declined significantly, but what about the actual rates at which people are getting heart disease?
Well, that figure is a little tougher to find out. The AHA doesn’t particularly want us to know about the incidence of heart disease; they just want us to know that deaths from it are declining. To find the incidence you have to go to a table called Hospital Discharges with Cardiovascular Disease as the First Listed Diagnosis.
As you can see the rate of these discharges is increasing. When you correct for the increase in population over the years, the line doesn’t increase as rapidly, but still increases slightly. What does this mean? It means that despite all the decrease in cholesterol levels, all the statin drugs, all the blood pressure drugs, and all the people on low-fat diets, that the number of people developing heart disease hasn’t dropped at all. If anything it has increased.
How can we tell this from this chart? When doctors hospitalize patients the doctors list the disorder requiring the hospitalization first. If someone with heart disease has a car wreck and breaks his femur (the thigh bone), he would be admitted with thigh fracture as his first diagnosis. Ditto for someone with heart disease who develops pneumonia; he (or she) would be admitted with pneumonia as a first diagnosis. The people represented on this chart are people who have been hospitalized for heart disease. It’s probably not the most reliable source for a truly accurate incidence of heart disease statistic, but it will have to do because the AHA doesn’t report anything else. They even tell us not to believe their own incidence statistics from year to year.
Incidence is an estimate of how many new cases of a disease develop in a population in a one-year period. For some disease statistics, new and recurrent attacks or cases are combined.
The incidence of a cardiovascular disease (CVD) in the United States is estimated by multiplying the incidence rates reported in community- or hospital-based studies by the U.S. population. The rates are not computed annually; they change only when new data are available. The estimates were revised to reflect the 2000 U.S. Census. Neither the incidence nor the rates should be compared with those in past issues of the Heart Disease and Stroke Statistics Update.
If we’re still getting heart disease at the same rate, why are deaths from heart disease going down? Easy. Better, more rapid medical intervention. We’re treating it better once we’ve got it, but we’re still getting heart disease at the same rates as before.
We know that smoking, a known risk factor for both heart disease and lung cancer, is on the decline. It would make sense then that the incidence of these diseases should be on the decline as well. And indeed the incidence of lung cancer has fallen by 18%. But the incidence of heart disease is the same despite the drop off in smoking.
What does this tell us? It can tell us a number of things. It could tell us that smoking isn’t really a risk factor for heart disease. (Millions have quit smoking and heart disease incidence hasn’t changed) It could tell us that all the changes that the AHA has recommended–reducing cholesterol (with drugs, if necessary), reducing blood pressure, following a low-fat, low-cholesterol diet–are making us have more heart disease to replace that that has fallen due to smoking cessation. Or it could tell us that something else is increasing the incidence of heart disease to make up for that that has fallen due to fewer smokers.
If you look back at the chart above showing the trends in cardiovascular risk factors, you will notice diabetes on the far right. The chart is constructed poorly in that it doesn’t show the relative rates of increase or decrease of these factors, which would be much more meaningful. If you look at the far left, you can see that high cholesterol has fallen from 33.6 to 17, which represents a decrease of about 0.5. Look again at diabetes and you can see that it is up by a factor of almost 3. And that’s diagnosed diabetes. There are many more cases of undiagnosed diabetes out there than there are of diagnosed diabetes. And there are many more cases of glucose intolerance than there are even undiagnosed diabetes.
It would seem therefore that the reduction in the incidence of cardiovascular disease we should expect to see from the decline in the numbers of smokers is made up for by the increased in people with diabetes.
And remember, diabetes is a disease of too much sugar in the blood, not too much fat. We’ve got a situation in which the heart disease authorities, the AHA, have recommended that we cut the fat in our diets, keep our blood pressures low, and stop smoking to reduce our risk for heart disease. We’ve done all those things, yet the incidence of heart disease hasn’t changed. Why? Because we’ve increased the sugar in our blood. It’s more data showing the lipid hypothesis is flawed while putting another notch in the belt of the sugar hypothesis.
To paraphrase James Carville: It’s the sugar, stupid!