What is Normal - Article

I t seems that the definitions of normal, healthy levels of blood pressure, choles- terol, and other health indicators keep changing, so that increasing numbers of peo- ple are classified as being at risk and therefore in need of treatment. Yesterday’s normal be- comes today’s needs-to-be-treated. More than a few people have wondered why the bar keeps moving, and whether stan- dards change just so doctors, laboratories, and other health care providers can generate more business for themselves. Doctors and others in health care aren’t immune to the pursuit of self-interest, but there’s a bet- ter, fuller explanation for the migrations of normal. Normal ≠ healthy For any medical test, nor- mal is initially defined after taking measurements in thousands of apparently healthy people. The result is usually a bell-shaped curve, in which the most frequently found values are in the mid- dle of the curve and the lowest and highest values, which occur in just a small percentage of people, are in the tail ends. Somewhat arbitrarily, the 2.5% of people with the lowest and highest values are said to have abnormal results, whereas the 95% who fall in between them are said to have normal ones. Used this way, normal means typical— values that fall into the broad range of what’s common. What it does not mean, although it’s often understood this way, is that there’s nothing to worry about if you’re “normal.” Take blood sugar (glucose) levels, for exam- ple. In the late 1970s, a group of respected experts set the upper limit of normal for fast- ing (no eating six to eight hours before the test) blood sugar level at 140 milligrams per deciliter (mg/dL). That meant that someone with a level above 140 would be classified as diabetic. It’s important to recognize diabetes early, even when it’s not yet causing symptoms, because if left unrecognized and untreated, it increases the risk for many other diseases, including heart disease and stroke. About 10 years ago, another group of ex- perts changed that upper limit of 140 to 126 mg/dL and created a new gray-area category of prediabetes (sometimes called impaired fast- ing glucose) for values between 110 and 125. The American Diabetes As- sociation has now lowered normal fasting blood sugar even further, to values below 100 mg/dL, and set 100 to 125 mg/dL as the range for the prediabetes diagnosis. Each of these rollbacks was prompted by evidence from large epidemiologic studies that showed an increased risk of de- veloping diabetes—and the condi tions as- sociated with it — as fasting blood sugar rose above 100: at 125 the risk was greater than at 100, and at 140 it was greater than at 125. Currently, there’s no solid evidence that the risk at 99 is greater than the risk at, say, 90, which is why 99 is now defined as the upper limit of normal. The definition of normal blood pressure has been reconfigured in a similar fashion. The “old” normal used to be readings under 140/90 millimeters of mercury (mm Hg). But in 2003, revised guidelines carved out a new category of prehypertension for readings be- tween 120–139/80–89 mm Hg. That change effectively toughened the standard for normal, lowering it to readings below 120/80. What is normal? The shifting boundaries of normal and healthy cholesterol levels, blood pressure, and other health indicators. INSIDE Waxing enthusiastic Having some earwax is actually good for your ears. . . . . . . . . . . . . . 3 Do they help where it hurts? A look at the effectiveness of topical pain relief medications. . . . . . . . . . 4 – 5 Fish and tips Six things you should know about fish and omega-3s. . . . . . . . . . . 6 – 7 By the way, doctor How can someone who has passed a stress test still have a heart attack? How much sleep is necessary to stay healthy? . . . . . . . . . 8 Write to us at health _ letter@hms.harvard.edu Visit us online at www.health.harvard.edu For customer service, write us at harvardHL@strategicfulfillment.com VOLUME 34 • NUMBER 1 NOVEMBER 2008 Share the precious gift of health this holiday season Give a gift subscription to the Harvard Health Letter . It’s the gift that keeps on giving, every month of the year! To order, call 877-649-9457 (toll-free). -2.5 +2.5 Normal

2 | Harvard Health Letter November 2008 www.health.harvard.edu The creation of prehypertension and other pre- conditions has been criticized for taking people who are fundamentally well and categorizing them as now having an illness—with the not-unintended consequence of generating millions of new patients for doctors and hospitals and new sales for drug companies. The rebuttal: there’s good evidence from well-designed studies that people with readings in the prehypertension range are likely to develop blood pres- sure levels that keep on rising until they reach full-fledged hypertension levels, and that taking action early can prevent this from happening. Moreover, the pre- ferred action isn’t taking a pill, but losing weight, getting more exercise, and eating healthier food. If more people followed this advice, the makers of blood pressure drugs would end up with fewer custom- ers, not more. The same can be said for the other pre-conditions. Does changing levels help? Once it’s been established that, statisti- cally, normal and healthy aren’t one and the same, the next question is whether treatments that drive an unhealthy level into a presumably healthier range actu- ally improve health. Blood cholesterol is an interesting case in point. Decades ago, total cholesterol levels of below 280 mg/dL were consid- ered normal for Americans, using the 95% middle portion of the bell-shaped curve as the yardstick. The biological connection between cholesterol and heart disease hadn’t been established. Once it was, medical researchers then asked the obvious question: whether the risk of developing heart disease is higher in someone with a cholesterol level of 280 than someone with a cholesterol of 240, which also fell in the normal range of the curve. The answer was yes, and they also found that the risk was high- er at 240 than at 200. So while people with total cholesterol levels between 200 and 280 had typical values, those values weren’t healthy. After cholesterol-lowering drugs be- came available, the next step was to conduct studies testing whether lower- ing cholesterol actually improved health. Dozens of randomized clinical trials con- ducted around the world in many differ- ent sorts of people proved that lowering cholesterol, in a variety of ways, to about 200 mg/dL prevented heart attacks and other cardiovascular problems. Randomized trials of the powerful statin drugs have taken things a step fur- ther by showing that lower is better for people with heart disease or with risk factors for it. National cholesterol guide- lines were revised several years ago to say that aiming for an LDL cholesterol of less than 70 mg/dL should be a “therapeutic option” for certain high-risk patients. For most people, levels below 100 mg/dL are deemed optimal and 100 to 129 mg/dL, “near-optimal.” The push to lower cholesterol has had a pronounced effect on average levels in the United States. In the early 1960s, the age-adjusted average cholesterol level was over 222 mg/dL. By the late 1970s it had dropped to between 210 and 215. And in 2005–2006 it was down to 199. Replacing normal with desirable In our view, in many contexts, we should stop talking about what numbers are normal and start talking about the ones that are desirable, based on solid scien- tific evidence that achieving them will bring about health benefits. This has always been the language used by cho- lesterol guideline writers, and it would be helpful if other guideline writers were to follow suit. At one level, this is just a quibble about semantics. On the other hand, a lot of the confusion about nor- mal comes from its several meanings. “Desirable” does a better job of captur- ing what the guidelines, doctors, and patients really mean when they use the word normal. It might also reduce some of the angst about normal (aka desirable) if the nonpharmacological approaches to the various pre-conditions received more emphasis. Too often, diet and exercise are given lip service when they should be taken more seriously as treatment for an emerging health problem. What is normal? continued from page 1 Editor in Chief Anthony L. Komaroff, M.D. Editor Peter Wehrwein peter_wehrwein@hms.harvard.edu Art Director Heather Derocher Acting Art Director Mary Allen Production Coordinator Charmian Lessis Copy Editor Robin Netherton • Editorial Board Board members are associated with Harvard Medical School and affiliated institutions. They review all published articles. Cardiology Thomas H. Lee, M.D. Dental Medicine R. Bruce Donoff, D.M.D., M.D. Dermatology Kenneth Arndt, M.D. Emergency Medicine John Tobias Nagurney, M.D. Gastroenterology Stephen E. Goldfinger, M.D. Genetics Susan P. Pauker, M.D. Gerontology Kenneth L. Minaker, M.D. Internal Medicine Nancy Keating, M.D., M.P.H. Neurology Dennis Selkoe, M.D. Edward Wolpow, M.D. Nutrition Bruce Bistrian, M.D., Ph.D. Walter C. Willett, M.D., Dr.P.H. Oncology Robert J. Mayer, M.D. Ophthalmology B. Thomas Hutchinson, M.D. Orthopedics Donald T. Reilly, M.D., Ph.D. Otolaryngology Jo Shapiro, M.D. Preventive Medicine JoAnn E. Manson, M.D., Dr.P.H. Psychiatry Michael C. Miller, M.D. Surgery Richard Hodin, M.D. Urology Jerome P. Richie, M.D. Women’s Health Soheyla Gharib, M.D. • Customer Service Call 877-649-9457 (toll-free) E-mail harvardHL @ strategicfulfillment.com Online www.health.harvard.edu/subinfo Letters Harvard Health Letter P.O. Box 9308 Big Sandy, TX 75755-9308 Subscriptions $32 per year (U.S.) Bulk Subscriptions StayWell Consumer Health Publishing One Atlantic St. Stamford, CT 06901 203-653-6266 888-456-1222 x31106 (toll-free) ddewitt @staywell .com Corporate Sales/Licensing StayWell Consumer Health Publishing One Atlantic St. Stamford, CT 06901 jmitchell @staywell .com Editorial Correspondence E-mail health_letter @ hms.harvard.edu Letters Harvard Health Letter 10 Shattuck St., 2nd Floor Boston, MA 02115 Permissions Copyright Clearance Center, Inc. Online www.copyright.com Published monthly by Harvard Health Publications, a division of Harvard Medical School Editor in Chief Anthony L. Komaroff, M.D. Publishing Director Edward Coburn © 2008 Harvard University (ISSN 1052-1577) Proceeds support research efforts of Harvard Medical School. Harvard Health Publications 10 Shattuck St., 2nd Floor, Boston, MA 02115 The goal of the Harvard Health Letter is to interpret medical information for the general reader in a timely and accurate fash- ion. Its contents are not intended to provide personal medical advice, which should be obtained directly from a physician. PUBLICATIONS MAIL AGREEMENT NO. 40906010 RETURN UNDELIVERABLE CANADIAN ADDRESSES TO: CIRCULATION DEPT., 1415 JANETTE AVE., WINDSOR, ON N8X 1Z1 E-mail: ddewitt @ staywell.com