In recent years, many persons with undiagnosed or diagnosed hypertension have begun checking their blood pressure in public places using devices that are provided as a free service by pharmacies or other retail stores. These devices offer persons who might not have the resources to own blood pressure monitors or to see their doctor frequently the ability to check their blood pressure. Home blood pressure monitoring has been established as an acceptable alternative and supplement to office-based measurement.^1,2 However, significant concerns about the blood pressure monitors available in public places have limited their value as a screening and monitoring tool (see accompanying table).

Although it is logical to think that public blood pressure measurement devices would improve the detection and treatment of hypertension, the accuracy of these measurements has not been established. Hamilton and colleagues³ have shown that public blood pressure monitoring devices increase self-measurement rates. They placed a validated home blood pressure monitor (Omron HEM-705 CP) in 13 public places in lower socioeconomic areas of Exeter in England. Over six months, 758 persons measured their blood pressure for the first time; 221 (29.2 percent) had blood pressure measurements above 135/85 mm Hg and were referred to their family physician for further evaluation.³ However, a community-based study by Lewis and colleagues⁴ in Canada found that neither the Omron HEM-705 CP nor the Vita-Stat 90550 provided accurate blood pressure measurement in a community pharmacy setting.

Current information suggests that public blood pressure measurement devices are poor screening tools for hypertension. One device overestimated the presence of systolic hypertension, misclassifying 23 percent of normotensive persons as hypertensive. More importantly, it misclassified as normotensive 16.4 percent of persons with previously confirmed hypertension.⁵ This false-negative rate is far too high for a screening tool.

The questionable accuracy of these devices suggests that they should not be relied on for ongoing blood pressure monitoring. Over the past 25 years, several studies^5–11 have evaluated the Vita-Stat devices, the most common blood pressure monitoring devices available in public places. Most of these studies have focused on the Vita-Stat 8000, which uses the auscultatory technique for blood pressure measurement. In all studies, there was much better agreement for diastolic blood pressure (DBP) than for systolic blood pressure (SBP). However, there was an unacceptably high variation in SBP and DBP measurements; systolic readings in individual patients could be as much as 60 mm Hg below to 58 mm Hg above the reference auscultatory mercury measurement.

Salaita⁸ found that patient age significantly affects the accuracy of the device. Compared with measurements using the random-zero auscultatory method, one automated device overestimated SBP by a mean of 7.4 mm Hg in persons 25 years of age and underestimated SBP by a mean of 6.3 mm Hg in persons 75 years of age (P < .001). Similar differences in DBP measurements were discovered. This study also found significant differences in measurements between the 10 machines that were tested.

Another concern is whether arm circumference affects the accuracy of measurements obtained with these devices. The Vita-Stat 90550 is clearly marked as being applicable to persons with arm circumferences of 9 to 13 inches (22.9 to 33 cm). Using arm-circumference data from the Third National Health and Nutrition Examination Survey, this cuff size would be applicable to only 63 percent of the general population and 50.3 percent of hypertensive persons in the United States.¹² The use of cuffs that are too small can lead to a significant overestimation of SBP and DBP and a misdiagnosis of hypertension.¹³ Thus, the size of the cuff used on some devices significantly limits their use, especially in the hypertensive population.

Even if public devices were accurate, no data are available to help interpret the measurements. To respond with appropriate treatment, more data are needed about the effect these devices have on blood pressure measurements. These devices often are located in noisy, busy areas of a store, so it is unclear whether readings correlate best with measurements taken in a physician's office (140/90 mm Hg or greater for a diagnosis of hypertension) or in out-of-office settings (135/85 mm Hg or greater). Conformational studies, such as comparison of public readings with daytime values for 24-hour ambulatory blood pressure monitoring, would be required to settle this issue. In addition, these freestanding devices do not connect patients with physicians if readings are significantly low (SBP less than 100 mm Hg) or high (SBP greater than 200 mm Hg or DBP greater than 120 mm Hg). Advances in wireless Internet technology could remedy this problem by allowing patients to enter their names and blood pressure measurements. At the very least, patients could obtain a list of local health care providers from a Web site run by local medical societies.

Blood pressure measurement outside of the physician's office may become a powerful tool to improve the diagnosis of hypertension, aid in drug titration, improve control of hypertension, and reduce long-term costs of hypertension management. However, current devices for public blood pressure measurement fail to meet accuracy criteria. Until manufacturers of these devices remedy the defects that cause inaccurate measurements, patients should be discouraged from using these devices, and physicians should not alter antihypertensive therapy on the basis of measurements derived from these sources.