More epidemiological bunkum

This past week, two startling admissions came out of Harvard.

  • First, Harvard researchers admitted that relative risks of 21%, 31%, 42%, 66% and 102% in a recent epidemiology study — linking drinking a single diet soda each day to lymphoma and leukemia (but only in men) — were no greater than could be found by random chance and not to be taken as proof of anything. [Despite the null findings, Harvard released a press release, with a sensationalized headline, hyping the risks of the artificial sweetener aspartame. As you’ll remember, after the scientific community, including the FDA, called them on the scares, which the overwhelming body of scientific evidence had long ago dispelled, Harvard withdrew the press release.]
  • Second, they admitted the unreliability of the self-reported data and dietary intake estimates in the data used in their research. Their massive Nurses Health Study and Health Professionals Follow-up Study databases have been the source of hundreds upon hundreds of articles, all reported as finding an increased health risk of some food, drink, body weight, lifestyle habit or chemical exposure… And nearly all finding equally untenable correlations that are no greater than might have been found by random chance.

So, will media be more careful before sensationalizing epidemiological studies and reporting ill-founded health cares? Or will they continue to uncritically parrot press releases? The answer came this week as men were targets of another health scare reported  nearly verbatim from another press release.

Heart Disease Risk Factors Also Tied to Death from Prostate Cancer

Men with metabolic syndrome — a group of symptoms linked to heart disease and diabetes risk — may also face a higher risk of dying from prostate cancer if diagnosed with the disease, according to a large new study. Metabolic syndrome includes high blood pressure, high blood sugar and high blood fat levels, as well as greater than normal body-mass index (BMI), a measurement of body fat based on height and weight.

The study authors noted that by following health recommendations on diet and exercise to prevent heart disease and diabetes, men can also lower their risk of death from this form of cancer. Researchers from Umea University in Sweden, led by Dr. Par Stattin, a visiting scientist at Memorial Sloan-Kettering Cancer Center in New York City, examined data on more than 290,000 men enrolled in a long-term study on metabolic syndrome and cancer… 

Men with the highest body-mass index had a 36 percent higher risk of dying from prostate cancer. Those with high blood pressure had a 62 percent greater risk of death from the disease.

Had reporters taken a moment to even read the abstract, they could easily have seen another example of epidemiological junkscience and null findings.

  • It was another computerized dredge though a similar database: a compilation of Swedish and Norwegian cohorts called the Metabolic Syndrome and Cancer Project (Me-Can). It included self-reported data on lifestyle habits from questionnaires filled out on enrollment, with an average of 40% of enrollees dropping out and not followed up.
  • That 289,866 figure was a trojan number, giving the impression it was a huge study. The actual number of men diagnosed with prostate cancer over a 12 year period and examined in this study was 6,673, with only 0.3% (961) deaths.
  • The results were similarly inconsequential. Cox proportional hazard computer models were used to calculate relative risks… and no tenable correlations were found. After adjusting for age, body mass index had a relative risk of 1.36 and blood pressure 1.62  (“36% and 62% greater risk”) — nowhere near the two to three-fold increase needed to be more than random chance and statistical error.
  • Inconvenient findings were omitted from the media coverage, even though they were untenable, too. For example,  as other metabolic syndrome risk factors — blood sugar and triglyceride levels — increased, prostate cancers decreased. In other words, men with the highest of those “bad numbers” compared with the lowest numbers had 18% and 12% lower risks for prostate cancer, respectively.
  • Nor did the findings make sense and consistent with what is documented in the medical literature.

“Metabolic syndrome” is little more than an indication of how our “numbers” (weight, blood sugars, blood pressure, cholesterol, etc.) naturally change (rise) as we age. It’s not the numbers associated with diseases of aging that are important, it’s the aging. And, there’s not much we can do to stop aging.  

Even the American Cancer Society states that the greatest risks for prostate cancer is getting old. Cancer is rare in younger men. Prostate cancer also appears to have a genetic component and to run in some families, and be higher among certain races and nationalities. Most men diagnosed with prostate cancer do not die from it, according to the ACS.

Nor have most studies even been able to find a link between being fat and getting prostate cancer, according to the ACS. Such a link is likely just as coincidental as the reported link with tallness. Additionally, CDC scientists report that US health statistics continue to show little or no association of all-cancer mortality with any weight range.

If obesity and metabolic syndrome has been such a “growing” problem, purportedly affecting more than 75% of the adult male population, then it makes no sense that obesity would have much to do with developing prostate cancers. Why? Both the incidence and death rates of prostate cancers, including invasive cancers, have been dropping, according to the government’s own statistics. The American Cancer Society, the Centers for Disease Control and Prevention (CDC), the National Cancer Institute, and the North American Association of Central Cancer Registries collaborate annually to provide updated information on cancer occurrence, deaths and trends in the United States. The most recent data shows: The incidence of prostate cancers declined 2.1% between 2000 and 2009 alone. And deaths from prostate cancer dropped 30.6% between 1950 and 2009.  

Most prostate cancers (93%) are localized or regional with a 100% survival rate, with more aggressive types more rare. Research from Fred Hutchinson Cancer Research Center has found that obesity is linked to an 18% lower risk of developing the most common types of prostate cancer that account for most all cases. Studies have also failed to consistently show an association between obesity and incidences of more advanced forms of the cancer, according to the ACS.

Nevertheless, although researchers haven’t been able to show why, some studies have suggested that obesity may be associated with a slightly lower survival from more aggressive forms of prostate cancer. People are quick to assume it’s a problem with the men and their weight. Instead, it may be disparities in current treatment guidelines being used. For example, a review of three national health surveys, found that obese men are less likely than thinner men to receive PSA-result driven biopsies, which could affect the stage and lethality of their diagnosis.

2 responses to “More epidemiological bunkum

  1. “… some studies have suggested that obesity may be associated with a slightly lower survival from more aggressive forms of prostate cancer.”

    I presume each instance of survival implies the tumour has been surgically removed. Ask any surgeon whether it is equally easy to operate on lean patients and on the obese ones, and how the difference affects survival. You’ll get an earful.

  2. The US health stats have not found any mortality differences for cancers among all body weight ranges. In fact, the medical literature is filled with findings of a survival advantage for many long-term illnesses among patients of higher weights than lower.

Leave a Reply

Fill in your details below or click an icon to log in: Logo

You are commenting using your account. Log Out / Change )

Twitter picture

You are commenting using your Twitter account. Log Out / Change )

Facebook photo

You are commenting using your Facebook account. Log Out / Change )

Google+ photo

You are commenting using your Google+ account. Log Out / Change )

Connecting to %s