THE "NEW" PROSTATE CANCER INFOLINK

According to a report on a presentation given yesterday at an ongoing meeting in Washington, DC, “Adequate sleep isn’t a luxury; it’s essential. And for men, it might even mean the difference between life and death”.

This report appeared on the HealthDay web site yesterday. A later quotation from the same report states that the data from this study are “intriguing” but that they are “not substantive enough to cause sleep-deprived males any alarm.” So why on Earth did the reporter first set out to alarm the reader?

The report is based on data presented by Susan Gapstur, vice president of epidemiology at the American Cancer Society, at the ongoing annual meeting of the American Association for Cancer Research (AACR).

What Gaspur and her colleagues actually report are data suggesting a possible association between sleep and risk for fatal prostate cancer among a cohort of 823,000 men in the USA:

• The men of < 65 years of age who said they slept for just 3 to 5 hours a night were 55 percent more likely to die of prostate cancer than the men who got 7 hours sleep a night.
• The men who said they slept 6 hours per night were 29 percent more likely to die of prostate cancer than the men who got 7 hours sleep a night.
• There was no such association among men of 65 years and over.

Your sitemaster should immediately declare a conflict of interest. He has “only” slept for about 4 or 5 hours a night for most of the past 50+ years. He doesn’t have prostate cancer. And he has a visceral dislike of this type of epidemiological study when it is (a) based to a high extent on people’s self-reporting about their own behaviors (which are well understood to be notoriously inaccurate); (b) picks out one life feature, while ignoring all of the others (e.g., what people ate over the same time frame); and (c) is then used to make statements like this to justify conformity to perceptions about “normal” human behavior patterns.

Now there is little doubt that a whole range of human behaviors certainly have some effects on risk for prostate cancer and for prostate cancer mortality.  But … parsing out exactly which human behaviors have how much impact on men’s risk of dying of prostate cancer is going to be a whole other issue.

Epidemiology is interesting. It allows us to put together associations like this one between male sleep patterns and prostate cancer mortality in specific populations, and use them to create hypotheses. However, not all hypotheses are necessarily good ones, and every epidemiological finding like this almost invariably comes with a whole raft of cautions about what we don’t know.

Note the words in bold italic type in the following direct quotes from Dr. Gaspur related to this study:

• “If confirmed in other studies, these findings would contribute to evidence suggesting the importance of obtaining adequate sleep for better health”.
• “… the findings contribute to evidence that the body’s natural sleep/wake cycle — circadian rhythms — might play a role in prostate cancer development”.
• “Lack of sleep may also contribute to the disruption of genes involved in tumor suppression”.

The fact of the matter is that the number of hours individual people sleep each night varies considerably over a “normal” distribution range. About 7 hours is an average value. Some people need more; others need less. But there are no data that your sitemaster is aware of that have ever confirmed that an increased risk for death from any truly biological illness (cancer, infection, sickle cell anemia, etc.) has ever been demonstrated to be caused by the number of hours that people sleep on its own. Might this be possible in extreme cases (e.g., men who slept only 3 or 4 hours a night when they knew they were seriously sleep deprived)? Sure, it’s possible. But those of us who have been sleeping only 3 or 4 hours a night because that’s all our bodies seem to need should not just be “lumped in” with that group of men. We aren’t sleep deprived!

Now sleep deprivation does increase risk for death — in car accidents, from other work-related accidents, and probably even in some psychological disorders that may lead to suicides. Sleep deprivation can be a very dangerous thing. We have no intention of suggesting that “a good night’s sleep” is unnecessary. However, it will take an enormous amount of data to persuade most good scientists that sleep deprivation (let alone not getting 7 hours of sleep a night) has any serious impact on risk for  any form of cancer-related mortality (unless that risk is being compounded by other significant behavioral abnormalities).

My suspicion is that Dr. Gaspur’s full presentation went into great detail about the limitations of this study — and there have to be a lot of those limitations. Whoever wrote and edited the HealthDay piece needs some further education about science writing. And one wonders to what extent the AACR deliberately “promoted” media coverage of this piece of research because they knew (or at least hoped) it would get the attention of the media. After all, the AACR now regularly runs television advertisements asking for donations for cancer research.

Filed under: Diagnosis, Risk | Tagged: deprivation, mortality, risk, sleep |