Treatment of patients with mild hypertension.

OVER the past 10 years a steadily increasing number of asymptomatic people with diastolic blood pressures between 90 and 100 mm Hg have been treated with antihypertensive drugs, more in the United States than anywhere else in the world. The expanding treatment of these 20 to 25 million people unquestionably represents the most extensive use of drugs ever attempted for the prevention of disease. As a result, in the United States the treatment of mild hypertension is now the leading indication for visits to physicians' offices' and the use of prescription drugs. Many physiciansargue that this steadily expanding use of drugs to treat milder and milder degrees of hypertension is necessary and has already proved to be valuable in saving lives and preventing disability. They point to data showing a progressive and substantial decline in cardiovascular mortality since 1968 as a beneficial consequence of this practice. A smaller number of physicians" have questioned the wisdom of this push toward drug therapy. In the words of the English epidemiologist Geoffrey Rose: "We may be unable to identify that small level of harm to individuals from longterm intervention that would be sufficient to make that line of prevention unprofitable or even harmful. Consequently we cannot accept long-term mass preventive medication." In his analysis, Dr. Rose used the clofibrate experience as an example of the unforeseen hazards of long-term preventive medication. His admonition has been strengthened by the recently published final report of this trial, which documents an l l% higher mortality in the clofibrate-treated group attributed to a wide variety of causes other than ischemic heart disease. We believe that similar though less obvious hazards have surfaced in the clinical trials of the therapy of mild hypertension. We also believe that the ethical justifications for the use of drugs to treat mild hypertension have not been completely fulfilled. As delineated by Brett," three links of evidence are needed to justify risk factor intervention: "(1) the epidemiologic evidence that a factor confers risk, (2) the availability of a therapeutic modality that changes the numerical expression of the factor, and (3) the evidence that such a numerical change leads to improved outcome." For mild hypertension the first two of these links have been provided. In regard to the first link, however, we should remember that the degree of risk for most persons with mild hypertension is relatively small and that those who are at higher risk can be identified reasonably well. It is the third link — the evidence for an improved outcome — that remains in doubt. We would also add a fourth necessary link to justify preventive intervention: the absence of substantial side effects that interfere with the quality of life. We will come back to this issue after we examine the evidence for an