Prevalence and accuracy of home sphygmomanometers in an urban ...

Prevalence and Accuracy of Home Sphygmomanometers in an Urban Population LORRAINE P. HAHN, MPH, AARON R. FOLSOM, MD, MPH, J. MICHAEL SPRAFKA, PHD, MPH, AND RONALD J. PRINEAS, MB, PHD multi-center blood pressure trial.15 Individuals who still had a working sphygmomanometer were asked a series of questions Abstract: The prevalence of ownership (7.5 per cent) and the on reason for purchase, use, number of hypertensives in the accuracy of home sphygmomanometers were determined in a pophousehold, and if they desired to have their sphygmomanomulation-based survey in the Minneapolis-St. Paul metropolitan area. eter calibrated. During August 1984, a trained technician visited Sixty-four per cent of home sphygmomanometers were accurate the participating households to calibrate the sphygmomanomwithin ±2 mm Hg of a calibrating sphygmomanometer; another 26 eter and to ask the participant some additional questions on per cent were within ±3-6 mm Hg. These results suggest that sphygmomanometer type, its purchase date, household knowlalthough many home sphygmomanometers are accurate, some are edge of how to use it, and the number of household members very inaccurate. Health care providers should advise regular caliwho were health professionals. bration when home sphygmomanometers are used for therapeutic Calibration was performed using a Baumanometer #300 1987; Public Health J (Am self-management of hypertension. (W.A. Baum Company, Inc., Copiague, New York) which 77:1459-1461.) was itself carefully calibrated prior to study onset. Using a plastic Y-connector the Baumanometer was connected in series to the device to be calibrated. The sphygmomanometer Introduction cuff was placed on a solid cylinder and inflated to 220 mm Hg on the Baumanometer scale. The value displayed by the High blood pressure is a major risk factor for cardiovassubject's device was recorded, and the pressure lowered to cular disease,' and reduction of high blood pressure has been the next level. In all, seven different calibration points shown to reduce cardiovascular morbidity and mortality.>J starting at 220 mm Hg and decreasing in decrements of 30 Many clinicians treating high blood pressure now recommend millimeters were used to assess sphygmomanometer accurathat hypertensives self-monitor their blood pressure using cy. For the purpose of this study, the differences between home sphygmomanometers. Such monitoring appears to be calibration and home sphygmomanometer were read to the convenient, cost-effective, and may improve treatment comnearest millimeter. No attempt was made to assess how pliance.5" Additionally, home blood pressure readings may accurately the participant measured blood pressure using be more representative than those obtained in the clinicians's their own device. office.9-13 Accuracy was computed in two ways: 1) the average The prevalence of home sphygmomanometers in the difference between the calibrating and home sphygmomanomgeneral population is currently undocumented. Furthermore, eters, and 2) the per cent of home sphygmomanometers within the extent to which home sphygmomanometers measure 0-2, 3-6, and -7 mm Hg of the calibrating device for each blood pressures accurately has not been adequately adcalibration point and for the average. Pearson correlation dressed. Consequently, we determined the prevalence of coefficients were used to assess possible associations among ownership and the accuracy of sphygmomanometers among device errors at the seven individual calibration points. participants of a population-based survey in the MinneapolisSt. Paul metropolitan area. Results Methods The prevalence of ownership of home sphygmomanomSubjects were participants in the Minnesota Heart Sureters in the MHS sample was 7.5 per cent. In comparison to vey (MHS), a cardiovascular risk factor survey of a probathe 1,386 MHS participants who did not own sphygmomability sample of 25-74-year-old men and women from the nometers, sphygmomanometer owners were almost twice as seven-county Minneapolis-St. Paul metropolitan population, likely to have had more than a high school education (Odds conducted in 1981-82. Survey methods have been described Ratio = 1.8; 95% CI = 1.2, 2.8) and be hypertensive (Odds previously.'4 In the survey, subjects were asked "Do you Ratio = 1.9; 95% CI = 1.3, 2.9). Age, sex, blood pressure, have a device in your home for measuring blood pressure?", and household income were not related to sphygmomanomfrom which prevalence of home sphygmomanometer ownereter ownership. Most participants purchased their sphygmoship was derived. manometer for either of two reasons: 1) they were a health Two years later, 85 per cent of MHS respondents with a professional who had a sphygmomanometer for home use, or home sphygmomanometer were recontacted by telephone in2) they bought it for a health concern on their own or on the terview for verification of ownership status. Also contacted advice of their physician. Although less than half of the were 23 individuals with home sphygmomanometers who were survey participants (41 per cent) had used their sphygmomanometers in the last six months, over 70 per cent of the participants in the Hypertension Prevention Trial (HPT), a hypertensive households had used theirs within that time. From the Division of Epidemiology, University of Minnesota School of A total of 1 14 individuals were interviewed for participation Public Health. Address reprint requests to Aaron R. Folsom, MD, MPH, in the calibration study (91 MHS, 23 HPT) and, of these, 76 Division of Epidemiology, School of Public Health, University of Minnesota, Stadium Gate 27, 611 Beacon Street SE, Minneapolis, MN 55455. This paper, agreed to have their home sphygmomanometer calibrated. The submitted to the Journal March 20. 1987, was revised and accepted for major reasons for not calibrating the remaining sphygmomapublication June 5, 1987. nometers were that participants no longer owned one, were no longer using it, or because it was non-operational. Only 16 per C 1987 American Journal of Public Health 0090-0036/87$1.50 AJPH November 1987, Vol. 77, No. 11

1459

PUBLIC HEALTH BRIEFS TABLE 1-Accuracy* of Home Sphygmomanometers at Various Calibration Levels Level of Accuracy (mmHg)

±7+

±3-6

±0-2 Calibration Level

(mmHg)

Total N

(N)

(%)t

(N)

(%)

(N)

(%)

220 190 160 130 100 70 40 All calibration levels (40-220)

75 75 76 76 76 75 75

53 50 53 51

14

14 21 17

(19) (25) (20) (21) (18) (28) (23)

8

46 53

(71) (67) (70) (67) (70) (61) (71)

9 8 5

(11) (8) (11) (12) (12) (1 1) (7)

74t

47

(64)

19

(26)

8

(1 1)

53

19 15 16

6 8 9

*Accuracy defined as the difference (mmHg) between the home and calibrating sphygmomanometers. tPer cents may not add to 100% due to rounding errors. t74 sphygmomanometers had complete data at each calibration level.

calibration point would be correlated with the differences at other points. Differences were highly correlated with one another at adjacent calibration points, but tended to be poorly correlated at distant points (data available upon request). Discussion Our findings indicate that a substantial proportion (7.5 per cent) of adults in the Minneapolis-St. Paul metropolitan area own home sphygmomanometers. In comparison to the rest of the population, those who owned sphygmomanometers tended to have more than a high school education and were more likely to be hypertensive. From a clinical point of view, although average accuracy of home sphygmomanometers was generally high-90 per cent were within ±6 mm Hg of the calibrating sphygmomanometer-some were inaccurate and a few were highly inaccurate. Home sphygmomanometers tended to underestimate blood pressure. Use of these devices could give misleading results if used for monitoring hypertensive therapy. For example, 11 per cent of hypertensive patients might measure their blood pressures in the "normal" range when

25

20

-~~~-

E 1

XE (A

0

*

-

-14 -13

-10 -9

-7 -6 -5 -4 -3 -2 -1

0

1

2 3

6

12

FIGURE 1-Comparison of Calibration and Home Sphygmomanometer Average Readings (N = 74)

cent of participants ever had their machines previously calibrated. The distribution of the sphygmomanometer types was: Standard mercury = five (7 per cent), Standard aneroid = 57 (75 per cent), Audio aneroid = 11 (15 per cent), and Automatic digital = three (4 per cent). Excluding the four non-operating sphygmomanometers, 64 per cent of home sphygmomanometers were accurate on average to within ±2 mm Hg (Table 1), and another 26 per cent were accurate within ±3-6 mm Hg. Conversely, 11 per cent of the home sphygmomanometers were .7 mm Hg out of calibration. For the individual calibration points, accuracy (±6 mm Hg) ranged from 88 per cent to 94 per cent. Figure 1 depicts the average difference between home and calibrating sphygmomanometers. The home sphygmomanometers tended to underestimate more often than overestimate blood pressure (mean difference = -1.5 mm Hg). Neither sphygmomanometer type, use in the last six months, nor having a household member as a health professional was associated with degree of accuracy. Within ±2 mm Hg, the Automatic digital sphygmomanometers were the most accurate on average (two of two machines), followed by the Standard mercury (four of five machines), the Audio aneroid (six of 10 machines), and the Standard aneroid (27 of 56 machines). We tested the hypothesis that the difference between the home and calibrating sphygmomanometer readings at one 1 460

actually their true blood pressures are out of control by more than 6 mm Hg. No sphygmomanometer type was more inaccurate than any other, although small numbers preclude meaningful interpretation. Furthermore, very few home sphygmomanometer owners (16 per cent) had ever had their devices calibrated after purchase. Our purpose was to examine the accuracy of home sphygmomanometers, not the accuracy with which measurements are made by patients at home. Previous reports concluded that patients could be trained to monitor blood pressures accurately.616 However, those studies used calibrated, standard sphygmomanometers provided by the investigators. Those who advocate self-monitoring of blood pressure should be aware of the potential inaccuracy and underestimation of self-purchased home sphygmomanometers. It can be recommended on the basis of this study that health care providers inquire as to the availability of home sphygmomanometers and that such devices be calibrated on a regular basis to avoid the inaccuracies noted in this study. ACKNOWLEDGMENTS

The authors acknowledge the following for their contributions to this study: Matthew Prineas, John Roesler, Gregory Silvis, and the interviewers and supervisors of the Minnesota Heart Survey. Supported by National Heart Lung and Blood Institute Grant-RO1-HL23727.

AJPH November 1987, Vol. 77, No. 11

PUBLIC HEALTH BRIEFS

REFERENCES

1. Inter-Society Commission for Heart Disease Resources-Atherosclerosis Study Group: Optimal Resources for Primary Prevention of Atherosclerotic Disease. Circulation 1984; 70:157-205. 2. Veterans Administration Cooperative Study Group on Antihypertensive Agents: Effects of treatment on morbidity in hypertension: Results in patients with diastolic blood pressures averaging 115 through 129 mm Hg. JAMA 1967; 202:1028-1034. 3. Veterans Administration Cooperative Study Group on Antihypertensive Agents: Effects of treatment on morbidity in hypertension: II. Results in patients with diastolic blood pressures averaging 90 through 114 mm Hg. JAMA 1970; 213:1143-1 152. 4. Hypertension Detection and Follow-up Program Cooperative Group: Five-year findings of the Hypertension Detection and Follow-up Program: I. Reduction in mortality of persons with high blood pressure, including mild hypertension. JAMA 1979; 242:2562-2571. 5. Burns-Cox CJ, Rees JR, Wilson RSE: Pilot study of home measurement of blood pressure by hypertensive patients. Br Med J 1975; 3:80. 6. Haynes RB, Sacket DL, Gibson ES, Taylor DW, Hackett BC, Roberts RS, Johnson AL: Improvement of medication compliance in uncontrolled hypertension. Lancet 1976; 1:1265-1268. 7. Gillum RF, Etemadi N, Boen JR, Kebede J, Anderson P, Prineas RJ: Home versus clinic blood pressure measurements. J Natl Med Assoc 1982; 74:545-549. 8. Edmonds D, Foerster E, Groth H, Greminger P, Siegenthaler W, Vetter W: Does self-measurement of blood pressure improve patient compliance

in hypertension? J Hypertens 1985; 3(suppl 1):31-34. 9. Julius S, Ellis CN, Pascual AV, et al: Home blood pressure determination. Value in borderline ("labile") hypertension. JAMA 1974; 229:663-666. 10. Julius S, McGinn NF, Harburg E, Hoobler SW: Comparison of various clinical measurements of blood pressure with the self-determination technique in normotensive college males. J Chronic Dis 1964; 17:391-396. 11. O'Brien E, O'Malley K, Fitzgerald D: The role of home and ambulatory blood pressure recording in the management of hypertension. J Hypertens 1985; 3(suppl 1):35-39. 12. Welin L, Svardsudd K, Tibblin G: Home blood pressure measurementsfeasibility and results compared to office measurements. Acta Med Scand 1982; 211:275-279. 13. Gould BA, Kieso HA, Hornung R, Altman DG, Cashman PMM, Raftery EB: Assessment of the accuracy and role of self-recorded blood pressures in the management of hypertension. Br Med J 1982; 285:1691-1694. 14. Luepker RV, Jacobs DR, Gillum RF, Folsom AF, Prineas PJ, Blackburn H: Population risk of cardiovascular disease: The Minnesota Heart Survey. J Chronic Dis 1985; 38:671-82. 15. Hypertension Prevention Trial Research Group (Writing Committee: Oberman A, Borhani NO, Cutler J, Detre K, Jeffrey R, Langford H, Meinert CL, Prineas RJ): Hypertension prevention trial-First year dietary changes. In: Mild Hypertension: From Drug Trials to Practice. Strasser T, Ganten D (Eds). New York: Raven Press, 1987, 187-202. 16. Steiner R, Luscher T, Boerlin H, Siegenthaler W, Vetter W: Clinical evaluation of semiautomatic blood pressure devices for self-recording. J Hypertens 1985; 3(suppl 1):23-25.

Clinical Trials to Assess Benefits of Experimental Drug on Alzheimer Patients HHS Assistant Secretary for Health Robert E. Windom, MD, recently announced a special two-year clinical trial of tetrahydroaminoacridine (THA), an experimental drug which may help control memory loss in some patients with Alzheimer's disease. The special study will be funded by the National Institute on Aging in cooperation with the Alzheimer's Disease and Related Disorders Association and the Warner-Lambert Company. NIA will provide $1.9 million for the two years; the ADRDA will provide $250,000; and Warner-Lambert will provide $3 million. In the clinical trial, a group of independent investigators from research facilities across the country will measure the safety and efficacy of THA by testing the drug in approximately 300 Alzheimer patients. This study follows a report by Dr. William K. Summers (New England Journal of Medicine, Nov. 13, 1986) of favorable results in 16 of 17 patients treated with THA, 12 of them for an average of one year. In all, the trial will involve more than two dozen investigators at 17 different sites: Mount Sinai School of Medicine (New York, NY); Johns Hopkins Medical Institutions (Baltimore, MD); Washington University School of Medicine (St. Louix, MO); University of Southern California (Los Angeles); University of Washington (Seattle); Massachusetts General Hospital (Boston); University of Pittsburgh (PA); Duke University (Durham, NC); University of California, San Diego; University of California, Los Angeles; Southern Illionis University (Springfield); Baylor College of Medicine (Houston, TX); Neuro/Medical Research Associates (Miami, FL); University of Minnesota (Minneapolis); University of Cleveland (OH); Burke Rehabilitation Center (White Plains, NY); and Rush-Presbyterian-St. Luke's Medical Center (Chicago, IL).

AJPH November 1987, Vol. 77, No. 11

1461