Correspondence
Control of Hypertension
N Engl J Med 2001; 345:1778-1780December 13, 2001
- Article
To the Editor:
Although the article by Hyman and Pavlik (Aug. 16 issue)1 provides important insights into the control of hypertension, the results of some of the analyses may be misleading. In the analysis of predictors of lack of awareness, people without hypertension were used as the reference group. People without hypertension are not at risk for being unaware that they have hypertension, so they are an inappropriate reference group.2 It appears that the reference group should have been persons with known hypertension, since only such persons are at risk for being unaware. The results mostly reflect the influence of risk factors for hypertension rather than risk factors for lack of awareness, which is why age is so important. The mean age was actually quite similar in the group of people who were unaware of their hypertension and in the groups of people who were aware of their hypertension.
The proportions of Mexican Americans and non-Hispanic whites with acknowledged but untreated hypertension were similar, as shown in Figure 2 of the article (19 percent and 17 percent, respectively). However, it appears that among all patients with known hypertension (59 percent of Mexican Americans and 70 percent of whites), Mexican Americans were more likely than whites not to be receiving any treatment (32 percent vs. 24 percent). This finding is consistent with that in a previous report3 and underscores the need to increase efforts to control hypertension among Hispanics.
The authors' methods also minimize the possible importance of health insurance. Any effect of lack of insurance is likely to be greatest among persons under the age of 65 years. This group was not analyzed separately. Moreover, because the effects of being uninsured are mediated by the lack of a regular source of care and by a reduced number of visits to physicians, it would be helpful to know the results without either of these variables in the model.4
4 ReferencesDavid W. Baker, M.D., M.P.H.
Case Western Reserve University, Cleveland, OH 44109-1998
dwb@po.cwru. edu 1
Hyman DJ ,Pavlik VN . Characteristics of patients with uncontrolled hypertension in the United States. N Engl J Med 2001;345:479-486
Full Text | Web of Science | Medline2
Kleinbaum DG, Kupper LL, Morgenstern H. Epidemiologic research: principles and quantitative methods. Belmont, Calif.: Lifetime Learning, 1982.
3
Sudano JJ Jr ,Baker DW . Antihypertensive medication use in Hispanic adults: a comparison with black adults and white adults. Med Care 2001;39:575-587
CrossRef | Web of Science | Medline4
Weinberg CR . Toward a clearer definition of confounding. Am J Epidemiol 1993;137:1-8
Web of Science | Medline
To the Editor:
As a geriatrician, I am disturbed that Dr. Chobanian, in his editorial on the control of hypertension,1 dismisses concern on the part of practicing physicians, particularly geriatricians, about the overzealous lowering of blood pressure in frail elderly patients. The results of two recent studies2,3 suggest a J-shaped curve for the incidence of stroke. When antihypertensive therapy for isolated systolic hypertension caused diastolic pressure to fall below 65 to 74 mm Hg, in one study, the curve for the incidence of stroke was reversed from a downward-sloping curve to an upward-sloping curve, particularly among patients with a diastolic pressure below 65 mm Hg.2 Also, reanalysis of the data from the Systolic Hypertension in the Elderly Program (SHEP) showed a 14 percent increase in the risk of stroke among those who had an average blood pressure of 177/77 mm Hg at base line and whose diastolic blood pressure was inadvertently reduced by 5 mm Hg or more with therapy.3 Thus, the fear of harm in elderly patients with isolated systolic hypertension who receive aggressive antihypertensive therapy cannot be completely alleviated by current data.
3 ReferencesValery A. Portnoi, M.D.
Beth Israel Medical Center, New York, NY 10003
vportnoi@bethisraelny.org 1
Chobanian AV . Control of hypertension -- an important national priority. N Engl J Med 2001;345:534-535
Full Text | Web of Science | Medline2
Voko Z ,Bots ML ,Hofman A ,Koudstaal PJ ,Witteman JC ,Breteler MM . J-shaped relation between blood pressure and stroke in treated hypertensives. Hypertension 1999;34:1181-1185
Web of Science | Medline3
Somes GW ,Pahor M ,Shorr RI ,Cushman WC ,Applegate WB . The role of diastolic blood pressure when treating isolated systolic hypertension. Arch Intern Med 1999;159:2004-2009
CrossRef | Web of Science | Medline
Author/Editor Response
The authors reply:
To the Editor: We are grateful to Dr. Baker for giving us an opportunity to comment on the use of persons without hypertension as a reference group for analyzing predictors of a lack of awareness of hypertension. We initially considered the approach proposed by Dr. Baker. However, since our interest was in determining whether being unaware of one's hypertension could be ascribed to infrequent contact with the health care system, we assumed that in a cross-sectional design, a portion of any increased use of health care services among persons who were aware of their hypertension could be a result of treatment they were receiving for a diagnosed condition. We therefore concluded that the appropriate reference group was the general population of people who were also “unaware” of hypertension.
Regarding the other two points, our report clearly underscores the problems of lack of awareness and less successful treatment of hypertension in Mexican Americans, and we fully agree that this is an important area for research and intervention. We disagree that we have in any way minimized the importance of health insurance. Since we did not find a significant interaction between age and health insurance status in the multiple logistic-regression models predicting the control of hypertension, a separate analysis of data for persons who were less than 65 years old was not warranted. We reported in our article that the odds ratios associated with health insurance status and the presence or absence of a usual source of primary care were not affected when the potentially mediating variable, frequency of visits to a physician, was omitted from the model. In response to Dr. Baker's inquiry, we tested models in which both the presence or absence of a usual source of care and the frequency of contact with a physician were omitted. In these models, health insurance status was not a significant predictor of lack of awareness of hypertension (odds ratio, 0.88; 95 percent confidence interval, 0.60 to 1.38) or lack of control of hypertension among persons who were aware that they had the condition (odds ratio, 1.21; 95 percent confidence interval, 0.76 to 1.92).
David J. Hyman, M.D., M.P.H.
Valory N. Pavlik, Ph.D.
Baylor College of Medicine, Houston, TX 77005
dhyman@bcm.tmc. edu Author/Editor Response
The editorialist replies:
To the Editor: I certainly agree with Dr. Portnoi's comment that antihypertensive therapy should be used with caution in frail elderly persons who have isolated systolic hypertension, and I made this point in my editorial. In this regard, particular care should be exercised in treating those who are confined to bed or who have restricted physical activity, since orthostatic hypotension and postprandial hypotension may complicate antihypertensive therapy in such persons.
Whether the reduction of diastolic blood pressure below some critical level increases the risk of stroke or other cardiovascular complications in patients with isolated systolic hypertension remains controversial. Although the recent analysis of the SHEP data suggested that the occurrence of stroke might have increased among those in whom diastolic blood pressure was lowered by 5 mm Hg or more,1 the findings are not supported by other studies. A meta-analysis of eight clinical trials involving elderly persons with isolated systolic hypertension2 showed no increase in the incidence of stroke among those in whom diastolic blood pressure was reduced to a level below the pretreatment value. On the other hand, the base-line diastolic blood pressure was inversely related to the risk of death at two years, suggesting that low pretreatment diastolic blood pressure may be a prognostic factor associated with adverse outcomes in patients with isolated systolic hypertension. In each of these trials, including the SHEP study, the overall reductions in the incidence of stroke that were associated with antihypertensive therapy were impressive. Although the new findings from SHEP should not be discounted completely and the problem should be examined further, it is clear that the treatment of isolated systolic hypertension in elderly persons results in more than a 30 percent reduction in the overall incidence of stroke. Aggressive treatment of the hypertension is therefore warranted in the vast majority of such persons.
2 ReferencesAram V. Chobanian, M.D.
Boston University School of Medicine, Boston, MA 02118
achob@bu.edu 1
Somes GW ,Pahor M ,Shorr RI ,Cushman WC ,Applegate WB . The role of diastolic blood pressure when treating isolated systolic hypertension. Arch Intern Med 1999;159:2004-2009
CrossRef | Web of Science | Medline2
Staessen JA ,Gasowski J ,Wang JG , et al. Risks of untreated and treated isolated systolic hypertension in the elderly: meta-analysis of outcome trials. Lancet 2000;355:865-872
CrossRef | Web of Science | Medline
