Men’s Fitness

DUELING STUDIES: WHEN FINDINGS ARE CONTRADICTORY, WHICH SHOULD YOU BELIEVE?

By Michael Castleman

Life is a bitch. You know that. So, thank God for science, which makes a crazy world comprehensible. Except that quite often, science confuses things even more.
 
Just ask epidemiologist Gregory Burke, M.D., a professor of public health at Wake-Forest Medical Center in Winston-Salem, North Carolina. Every time some new health-related study hits the headlines, Burke knows he’ll spend much of the day on the phone and answering email. Recently, he’s fielded a steady stream of questions about the benefits and risks of low-carb diets. “Some experts proclaim that low-carb eating is the only way to lose weight. Others say it’s a terribly risky lifestyle. Friends and relatives have gotten confused. I tell them the truth lies in between the extreme pro and con positions.”
 
Ted Gansler, M.D., has the same problem--only worse. As medical editor for the American Cancer Society (ACS) in Atlanta, he’s responsible for all the information on the ACS Web site and what ACS operators tell people who call the organization. “Every new cancer study means new questions and new confusion. We always get a spike in calls. Fortunately, a week before publication, the major journals give us a sneak preview so we can train our staff and add extra phone lines.”
 
Frequently, new research findings are so contradictory, they could drive a man to drink--which is either good or bad for you, depending on which study you read. Even scientists have trouble distinguishing the real deal from findings that are trivial, biased, or just plain junk. “Science requires patience,” Burke explains.  “But when it comes to medical information, Americans are impatient. It takes time—often years—to sort out contradictory findings.”
 
As a result, research can be very confusing. “The public is more scientifically astute than it was 20 years ago,” says biostatistician Christopher Portier, Ph.D., director of the Environmental Toxicology Program of the National Institute of Environmental Health Sciences in Research Triangle Park, North Carolina. “But people want certainty. In science, the only constant is uncertainty.”
 
Yes, the world is uncertain. But if you know some basics about evaluating research, you can have a better idea what to believe--and what to toss in the garbage.
 
Does Coffee Cause Heart Disease?
 
Coffee gets us going--but not, we hope, into an early grave. The bad news is that coffee has been linked to three important risk factors for heart disease, the nation’s leading killer: high blood pressure, high cholesterol, and high homocysteine, an amino acid in blood. Researchers at the Oklahoma City VA Medical Center recently joined a chorus of previous studies showing that coffee raises blood pressure. Swedish researchers recently found that filtered coffee, the kind most Americans drink, raises cholesterol. And Danish researchers have discovered that drinking coffee raises homocysteine. Maybe we should stop drinking it.
 
Not so fast. During the past decade, when Starbucks outlets have proliferated like dandelions across an untended lawn, men’s risk of death from heart disease has FALLEN from 500,000 a year to around 440,000. In addition, several studies show that moderate coffee drinking (up to about three cups a day) does not significantly increase heart disease risk. In fact, two recent studies--one in Finland, the other in Greece--show that compared with those who drink no coffee, a cup or two a day actually REDUCES it.
 
What gives? A good deal of the confusion can be laid at the door of the news media. The media, Gansler explains, select for the unusual: “If a study confirms the prevailing wisdom, that’s not news. But if a study contradicts previous research, it’s more likely to get publicized--and confuse people.”
 
In addition, by shining a spotlight on the study du jour, new headlines imply that the latest research is more important than previous research--even if many previous studies found something else. “In the public’s mind,” Burke explains, “a bad new study often trumps many older good ones. News reports often lack perspective.”
 
Newspaper, television, and radio stories are short and simple. The subtleties often get lost.  In research, the subtleties are crucial. All the studies linking to coffee to increased blood pressure, cholesterol, and homocysteine show that the effect is “acute,” meaning sudden and temporary. However, sudden, temporary spikes in blood pressure, cholesterol, and homocysteine are NOT risk factors for heart disease. The risk factor is “chronic” elevation, high levels over time. If temporary spikes in blood pressure caused heart disease, doctors would urge Americans not to exercise, see action movies, ride roller coasters, or have orgasms. “To really understand a study, you have to read the fine print,” says Roseville, California clinical nutritionist Jeanie Moloo, Ph.D., R.D., a spokesperson for the American Dietetic Association. “Headline News reports don’t provide it.”
 
Finally, in human studies (clinical trials), the number of participants is critical. If a small study raises a red flag, but a much larger study does not, chances are the small study got it wrong. The studies showing that coffee raises blood pressure, cholesterol, and homocysteine involved a total of 3,000 people. But the studies showing that moderate coffee consumption does not raise heart disease risk involved more than 100,000. “Small studies may come up with important findings,” Moloo says,  “but in general, large studies are more likely to get at the truth.”
 
Bottom Line: Coffee isn’t the most healthful beverage on earth, but a daily cup or two does not increase risk of heart disease.
 

Does Coffee Cause Cancer?
 
In the past few years, scary studies have linked coffee to several cancers: bladder, brain, colon, ovarian, pancreatic, and rectal.
 
However, during the same period, just as many studies have shown that moderate coffee consumption has nothing to do with risk of colon, pancreatic, and rectal cancer, and little, if anything, to do with bladder cancer. In addition, the latest research shows that coffee actually REDUCES risk of esophageal, lung, mouth, and throat cancer.
 
What gives? First, you have to look at the dose. Most coffee drinkers down one or two cups a day. To get a statistically significant risk increase in brain cancer, you have to drink more than SEVEN cups a day for 20 YEARS. “Dose is crucial,” Gansler explains. “Many things that are harmful in large doses are no danger--or even beneficial--in small doses.”
 
Then there’s how the study was done. With the exception of the brain cancer report, all the studies linking coffee to increased cancer risk were “retrospective” or “case-control.” The researchers asked people with cancer, the “cases,” how much coffee they drank over the past many years, then they surveyed similar people who were cancer-free, the “controls.” The advantage of retrospective/case-control studies, Portier explains, is that they can be done relatively quickly and inexpensively. “The problem is that they depend on memory, and memory is notoriously unreliable.” Think about it: A researcher asks you how much coffee you drank from, say, 1996 to 1999?  Even a small lapse of memory can skew the results.
 
The more reliable type of study is “prospective.” The researchers survey people about their lives at that moment, then follow them for many years, resurveying them periodically. Prospective studies take longer and are more expensive, Portier explains, but they are less likely to involve faulty memories.  “Compared with retrospective/case-control research, prospective studies are more reliable,” he says. Two of the studies exonerating coffee, the ones dealing with colon, pancreatic, and rectal cancer, were prospective and very large. The others were retrospective, but generally larger--and therefore more credible--than the studies suggesting that coffee causes cancer.
 
Finally, two national trends provide some perspective on the coffee-cancer issue. First, America’s per capita coffee consumption has been one to two cups a day for almost 30 years, long enough to figure into cancer, which can take decades to develop. But while some cancer rates are rising, the overall cancer rate is FALLING. Both new cancer diagnoses and cancer deaths topped out in the early 1990s and have been declining slowly but steadily ever since. With coffee consumption steady for three decades and cancer on its way down, it’s hard to make a case that coffee plays a major role in cancer risk. “Before you get too upset that something in your life causes cancer,” Gansler explains, “it helps to keep the big picture in mind.”
 
Bottom Line: A cup or two a day doesn’t cause cancer.
 
Is Alcohol a Health Food?
 
Absolutely not, screams Mothers Against Drunk Driving (MADD) and a host of public health officials. Alcohol is involved in 40 percent percent of fatal motor vehicle accidents, some 17,500 funerals a year. It’s also a major risk factor for drowning, violence, and fires and burns. Drinking is a risk factor for several cancers, among them: mouth, throat, esophagus, breast, and colon. And as anyone familiar with Alcoholics Anonymous can attest, it has wreaked havoc with millions of lives. Seventy years after the repeal of the repeal of Prohibition, it’s not difficult to make the case that alcohol is the nation’s most dangerous drug.
 
But about 15 years ago, unexpected results began turning up in the medical literature concerning heart disease, findings that have cracked the once-solid wall of medical condemnation of booze. Alcohol abstainers have a low risk of heart disease, and alcoholics are at high risk. But in between them, those who have a drink or two a day--but no more--have the lowest risk. So risk drops with a little drinking, then rises with more than two drinks a day--medically known as a “J-shaped curve.”
 
When the heart benefits of a little alcohol were first uncovered, the alcohol-is-the-devil-incarnate crowd howled that under no circumstances should public health officials give Americans any excuse to drink. They’re still howling. But over the years, dozens of studies have confirmed the J-shaped curve, and others have parsed out the reason for it. Small amounts of alcohol boost HDL, the so-called good cholesterol that helps prevent heart attack. Red wine also helps prevent the internal blood clots that trigger heart attack. The latest studies show that light-to-moderate drinking also helps prevent stroke, diabetes, dementia, osteoporosis, and prostate cancer.
 
Bottom line: Never drink and drive. Don’t binge on booze. Don’t drink more than two cocktails, beers, or glasses of wine a day. And be extra-careful with alcohol if you have a family history of alcoholism. But for overall health, a little alcohol is actually better than none.
 
Does Sex Cause Prostate Cancer?
 
Prostate cancer probably isn’t high on your list of immediate worries. But if there’s a connection, do you want your honey to say, “Not tonight, dear, I don’t want to give you cancer....”
 
In the past few years, several studies have shown that the more you do the nasty, the greater your risk of this cancer. The largest report comes from researchers at the University of Iowa. Compared with men who had no sex, those who slipped it in three times a week were 20 percent more likely to develop prostate cancer.
 
Or maybe not. Just as many studies show that sex has nothing to do with prostate cancer. In fact, two recent studies show that a hot sex life during your twenties and thirties REDUCES risk of later developing this cancer. One of these was a huge prospective trial by National Cancer Institute (NCI) researchers involving 30,000 men who were surveyed every two years for eight years. Compared with men who ejaculated once or twice a week during their twenties, those who ejaculated five or more times cut their lifetime risk of prostate cancer by one-third.
 
So what explains the studies showing that sex increases prostate cancer risk? Two things: First, says Michael Leitzmann, M.D., lead author of the NCI report, the studies showing that sex boosts cancer risk are retrospective--and less reliable than his, which was prospective and involved a huge number of men.
 
Second, Moloo explains, just because two things are associated--sex and prostate cancer--doesn’t necessarily mean one causes the other. In the studies showing a link between sex and prostate cancer, sexual frequency was also linked to something else, sexually transmitted infections (STIs). Many studies show that STIs increase risk of prostate cancer. So it’s the STIs--and not the sex, per se--that increases prostate cancer risk.
 
Bottom Line:  Assuming that you take precautions against STIs, sex does not increase risk of prostate cancer. In fact, it’s a good bet that sex reduces it. Now, prostate cancer may not be high on your list of immediate worries, but it just might improve your luck: “Hey, babe, want to help me prevent prostate cancer?”
 
Should You Believe It?
 
Life is messy and so is science. It’s hard to know what to believe. But here’s a quick guide that can help you decide what to take to heart and what to toss into the trash:
 
* What’s your source? Newspapers, TV, and radio disseminate information very quickly--and in the process often make mistakes. “In general,” Moloo says, “magazines do a better job of covering science news. They take more time, so their information has more perspective and greater depth.”
 
* Who paid for it? When a study shows that half of Americans don’t sleep well because they have old, worn-out mattresses, that’s interesting. But if the study is funded by the mattress industry trade organization, forget it.
 
* How big was it? Usually, the more subjects, the more credible the findings.
 
* Prospective or retrospective/case-control? Studies that compare healthy people with those who have some health problem are less scientifically credible than studies that follow a large number of people over many years to see what happens to them.
 
* Does the study deal with a hot-button issue? Be extra-skeptical of studies concerning things like marijuana, abortion, gun violence, alcohol, and other hot-button social issues. The findings might be politically influenced. Read the fine print.
 
* Call the organization. If you’re confused about a study dealing with any major condition--heart disease, cancer, stroke, diabetes, whatever--there’s a national organization that deals with it: the American Heart Association, the American Cancer Society, the American Diabetes Association, etc. These organizations provide perspective on new studies. Call or email them. “That’s what we’re here for,” says the ACS’s Gansler, “to help people separate the wheat from the chaff.”
 
* Before the latest study has you transforming your life, embrace the time-honored fundamentals of good health: Don’t smoke. Don’t have more than two alcoholic drinks a day. Get 30 to 60 minutes of exercise a day. Always use a seat belt in motor vehicles. Eat at least five servings of fruits and vegetables a day. (French fries don’t count). Maintain the weight recommended for your height and build. Sleep at least seven hours a night. Cultivate close relationships with a lover, friends, and family.
 
* Finally, never change your life on the basis of just one study, especially if it contradicts tried-and-true health wisdom.
 
If you do all these, you’re likely to live a long, healthy life--even if you largely ignore the steady stream of contradictory studies.
 
 
Sidebar
Animal Planet: Can Animal Studies Be Believed?
 
Coal miners took canaries into the mines because the birds were highly sensitive to poison coal gas. If a canary died, the miners knew that gas was seeping into the shaft, and they evacuated. But today, when studies show that a chemical is hazardous to animals, the compound’s maker typically curses the canary, and insists that animal results can’t be believed.  They also say: Give mice enough of anything, and they get cancer.
 
Wrong on both counts. In the Headline News version of reality, it often seems like everything causes cancer. This is simply not true. In fact, the latest report by the National Toxicology Program of the National Institute of Environmental Health Sciences shows that of the thousands of compounds currently in wide use, only 228 cause cancer, mostly pesticides and industrial chemicals.
 
Furthermore, while mice are not men, the latest research shows that we share 99 percent of our furry friends’ genetic material. Without dismissing our unique 1 percent, the scientific consensus is that when it comes to detecting health hazards, animal results can almost always be extrapolated to humans. In fact, all human carcinogens have been shown to be carcinogenic in laboratory rodents, and many human carcinogens were first identified in animal studies.
 
“Animal studies are not infallible,” Portier explains, “but they are generally good indicators of things likely to prove hazardous to people. It’s impractical and unethical to test potentially toxic or carcinogenic substances on humans. Animal studies are our best compromise between science and ethics.”
 
 
San Francisco health writer Michael Castleman is the author of 12 books, most recently Great Sex: The Man’s Guide to the Secrets of Total-Body Sensuality (Rodale, 2004). Visit greatsexthebook.com.