What's the Risk?

            A new study of 5700 consecutive COVID-19 patients hospitalized in the New York area is making waves because it reports a high rate of underlying chronic disease, seemingly amplifying findings from Wuhan and elsewhere. But what is striking about this group of severely ill COVID-19 patients is not so much their associated chronic conditions as how similar they are to much of the general population.  

            The study, published in JAMA, reported obesity in 47.7 percent of patients, very much like the rate among adults generally: 44.8 percent of 40-59-year-olds and 42.8 percent of those over age 60 are obese. For high blood pressure and diabetes, the rates of disease in the COVID-19 patients closely resembled the rates in the older population in general. The study found high blood pressure in 56.6 percent of the COVID-19 patients; that’s awfully close to the rate of 60 percent found in the general population among people over age 65—and considerably higher than the rate of 33.2 percent found in the general population among people aged 40-59. And the study noted that diabetes was present in 33.8 percent of the very ill COVID-19 patients; that is fairly similar to the rate of 27 percent found among the elderly in general—and markedly higher than the 17.5 percent typically found in the general population of 45-64-year-olds.

            To better understand the significance of the observations about chronic conditions in the COVID-19 patients, the authors of the JAMA article need to examine age-specific rates of those disorders. Without this information, we can’t say very much about risk factors—except that obesity doesn’t seem to be much of a risk factor at all since its rate in the hospitalized COVID-19 is very similar to that in the general adult population. What about hypertension and diabetes?

            Since the median age of the patients in the JAMA study is 63, that means that about half the patients are elderly and about half are not. If all we know is that the rate of high blood pressure in the patients is 57 percent, then there are three possibilities: 1) that 57 percent figure applies across the board, regardless of age; 2) the rate among the half of the study population that’s over 65 is greater than 57 percent (in which case the rate among the half that are under 65 is less than 57 percent); or 3) the rate among the half of the study population that’s over 65 is less than 57 percent (in which case the rate among the younger patients is more than 57 percent). 

            Let’s put sample numbers on these 3 situations, comparing them to what we know about rates in younger and older people in general. In the first case, where the 57 percent applies to everyone, regardless of age, this would mean that the risk of high blood pressure in the older population is the same (or a little lower) than in older people without COVID-19, where it’s 60 percent; and much, much higher than in the younger population, where it’s 33 percent. In the second case, let’s suppose the actual rate of hypertension in the older COVID-19 patients is more like 70 percent (higher than the 60 percent in the well elderly); that would imply the actual rate in the younger COVID-19 patients must be around 44 percent (higher than the comparable rate in healthy younger patients of 33 percent). In the third case, let’s suppose that an average hypertension rate of 57 percent means the actual rate of hypertension in the older COVID-19 patients is 44 percent (much lower than among healthy elderly) and the actual rate among younger COVID-19 patients is 70 percent (much, much higher than among healthy younger adults). What’s noteworthy among these three possibilities is that only in one of them is hypertension a risk factor in the elderly (case 2); in the other scenarios it's either not a risk factor or is actually protective. Moreover, if it is a risk factor, it may well confer only modestly increased risk.

            Whatever the relationship between chronic disease and the severity of COVID-19, what is clear is that Americans as a whole have high rates of chronic disease. A recent international comparison of health found that the US has a rate of chronic disease and obesity that is twice that of other developed countries. Among fee-for-service Medicare beneficiaries, the latest statistics reveal that 20 percent have between 2 and 3 chronic conditions; another 23 percent have 4-5 chronic conditions, and fully 17 percent have 6 or more chronic diseases. 

            Before we make older people with diabetes or high blood pressure unnecessarily anxious about contracting COVID-19—or falsely reassure those older people who don’t have diabetes or high blood pressure that they are at low risk—we need a more careful analysis. Perhaps the real take-away message from the JAMA study is that the U.S. needs to do a better job preventing chronic disease.

Down, Down, Down

         The most important recent development affecting older patients, without a doubt, is the issuance of new guidelines for the diagnosis and treatment of high blood pressure. The Wall Street Journal proclaimed “Nearly Half of US Adults Have High Blood Pressure Under New Guidelines,” and venerable health columnist Gina Kolata of the NY Times wrote: “The nation’s leading heart experts on Monday issued new guidelines for high blood pressure that mean tens of millions more Americans will meet the criteria for the condition, and will need to change their lifestyles or take medicines to treat it.” The report, which took me a while to track down and read since most of the references are to summaries of the report or commentaries on the report but not to the actual document, is 175 pages. It was issued or endorsed by eleven organizations, whose initials actually form part of the name of the report: its full name is the “2017 ACC/AHA/AAPA/ABC/ACPM/AGS/APhA/ASH/ASPCNMA/PCNA Guidelines for the Prevention, Detection, Evaluation, and Management of High Blood Pressure in Adults.” The subtitle is: “A Report of the American College of Cardiology/American Heart Association Task Force on Clinical Practice Guidelines.” Now that I’ve had a chance to digest this—what should we make of it?

         I remember the day in 1991 when JAMA published the results of the “SHEP” trial, a study of the treatment of isolated systolic hypertension in the elderly that upended what I’d been taught in medical school. Gone was the belief that high blood pressure was good for older people because they needed greater force to push blood through their stiff arteries; suddenly, there was compelling evidence that high blood pressure was dangerous—it caused strokes and heart attacks. Keeping the systolic blood pressure below 160 reduced the risk of stroke by 36 percent and the risk of a cardiac event by 27 percent. Overnight, the practice of medicine changed.

         To be sure, there were skeptics. Wouldn’t lowering the blood pressure cause older people to become dizzy and to faint? Just how far were we supposed to lower blood pressure anyway? But to me and many others, the choice was clear. Stroke was one of the worst fates that could befall older patients: sometimes it killed them, but more often, it left them impaired, often profoundly, significantly diminishing the quality of life.

         But I also remember a book by physician and historian of science, Jeremy Greene, published in 2008 and called Prescribing by Numbers: Drugs and the Definition of Disease. Greene argued persuasively that the pharmaceutical industry has a vested interest in lowering the cut-off for treatment of a variety of chronic conditions, such as hypertension, high cholesterol, and diabetes. Defining “pre-diabetes” as a real condition warranting treatment or repeatedly dropping the threshold for treating cholesterol with statins had clear implications for the drug companies: they would sell more pills. More volume, more revenue, more profits. So are the new guidelines just another instance of “diagnosis creep,” a way for Pharma to make more money?

         It’s important to realize that what tipped the scales for the guideline writers, what’s different now from the last time they wrote a guideline, three years ago, is the 2015 “SPRINT” study (Systolic Blood Pressure Intervention Trial), a randomized trial of high risk patients over 50 to either a systolic blood pressure target of 140 or 120. The study, which was reported in the New England Journal of Medicine, found a 25 percent reduction in cardiovascular events (such as stroke or heart attacks) when physicians tried to reduce the blood pressure to the lower value.

         As Gilbert Welch of the Dartmouth Institute, co-author of Overdiagnosed: Making People Sick in the Pursuit of Health (2012) and Less Medicine, More Health (2016), commented, that 25 percent sounds pretty good, but it’s the relative risk reduction. In fact, the rate of cardiac events fell from 8 percent in one group to 6 percent in the other—an absolute difference of only 2 percent. Or, looked at differently, 92 percent of people in one group and 94 percent of people in the other group did just fine. Suddenly, the impressive results don’t look quite so impressive. Moreover, all these patients were already at increased risk of cardiovascular disease because of other factors such as cigarette smoking or diabetes. Presumably, had the same study been carried out in people of average or lower risk, the benefits would have been correspondingly smaller.

         So if the new guidelines are based predominantly on the findings from one study, and the one study isn’t quite as compelling as it seemed at first glance, what should we make of the new recommendations? Should we really treat everyone with a systolic blood pressure over 120 who has at least a 10 percent risk of cardiac disease in the next decade? 

         The answer, I think, is yes, but with caveats. First, the guideline writers are at great pains to insist that blood pressure be measured the way it was measured in the SPRINT trial—in a way that it’s almost never measured in the doctor’s office, namely after sitting quietly for five minutes and averaged over three readings. They also advocate use of home monitoring to confirm (or refute) the diagnosis of high blood pressure, as well as to guide medication adjustment once treatment is started. Next, physicians need to be prudent about what medications to use if they’re going to embark on pharmacologic treatment. We have ample numbers of cheap, effective medicines that have been around for years, such as diuretics and beta blockers, and these should be tried before pricier medicines. Finally, both physicians and patients need to be vigilant about medication side effects, which means doctors need to tell their patients what to look for, patients need to report their symptoms, and doctors need to change course if symptoms develop. In addition, non-pharmacologic treatment should be attempted, including exercise, weight loss, and a low salt diet. If we follow all these steps, we can be confident we will be changing how we deal with blood pressure because it’s what’s best for patients, not because it’s what’s best for the pharmaceutical industry.