Nonadherence to Recommended Guidelines for Blood Pressure Measurement

Abstract

Accuracy of blood pressure readings, both in the physician's office and at home, is crucial in properly managing hypertension. Few studies have investigated adherence to measurement guidelines. This study focused on two important aspects of blood pressure measurement: waiting time before measurement and number of readings taken. A total of 103 patients completed self‐report questionnaires about office and home blood pressure measurements, with 77% reporting that physician measurements were obtained without waiting, and 56% reporting that only one reading was obtained. The proportions were even higher when measured by a nurse/technician, 96% and 81%, respectively. Home readings were taken without waiting by 60%, and 40% reported taking only a single reading. Most patients received no measurement instructions. Nonadherence to measurement guidelines is common, and may be affecting the validity of readings obtained both in physicians' offices and at home, with significant and potentially harmful effects on treatment decisions.

Hypertension is a well‐known modifiable cardiovascular risk factor. Correct diagnosis and treatment rely on accurate measurement of blood pressure (BP). In 2005, the American Heart Association (AHA) released guidelines for measurement of BP in the physician's office.1 These guidelines and others like it address the position and posture of the patient, proper position and placement of the BP cuff, rate of deflation of the mercury column when using the gold standard device (the mercury sphygmomanometer), and the recommendation that the patient be seated comfortably and unstimulated for 5 minutes before the first reading is taken.1, 2, 3 The AHA guidelines recommend collecting at least two readings, with a minimum 1‐minute interval between readings. Additional measurements should be obtained if more than a 5 mm Hg difference in systolic pressure exists between the two readings.1

In recent decades, self‐monitoring of BP at home has become widely practiced. Home readings provide more measurements upon which to base treatment decisions. They correlate more strongly with ambulatory readings than do office readings, and correlate at least as well as office readings with cardiovascular outcome.4, 5, 6 Home monitoring is also associated with improved adherence to medication and with better BP control.7, 8, 9 Home readings are particularly valuable in assessing and managing patients with suspected white‐coat hypertension.

The AHA recommends that patients who self‐monitor BP at home rest comfortably in a chair with a properly placed cuff for 3 to 5 minutes prior to taking any measurements.10 Patients should obtain three readings, separated by at least 1 minute.10 The European Society of Hypertension guidelines recommend two readings after a 5‐minute wait.3

Two recommendations of particular importance that could considerably affect treatment decisions pertain to the number of readings obtained and whether or not the patient is seated quietly prior to readings. The AHA guidelines, for office measurement, recommend a minimum of two readings.1 Since BP constantly varies, a single reading is less likely than two or three readings to accurately reflect a patient's usual BP. Further, the third reading is often lower than the first, often resulting in reclassification of the patient's hypertensive status.11 Guidelines almost universally recommend that patients are seated quietly for 5 minutes, consistent with the protocol of many large randomized trials and epidemiologic surveys.1, 3, 11, 12, 13, 14, 15

The degree of adherence in the real world to these measurement guidelines, whether in the physician's office or at home, has not been adequately studied. The purpose of this study was to evaluate adherence to the published recommendations for BP measurement obtained both in the clinic and at home, focusing specifically on the number of measurements taken and waiting time prior to obtaining the measurements.

Methods

Consecutive new patients with known or suspected hypertension were asked to complete a three‐part questionnaire at their initial visit to the Weill Cornell Hypertension Center, prior to being seen by a Hypertension Center physician. Patients younger than 18 years were excluded. Data were collected between May 2013 and December 2014, with 103 patients participating in the study. The protocol and consent form were approved by the institutional review committee at Weill Cornell Medical College.

Participants were given a questionnaire that explored how their BP was measured when taken at home and at the office of their primary or referring physician. With regard to home readings, the questionnaire asked how long participants sat after applying the cuff before taking the first reading, how many readings they usually took, and whether or not a physician had advised them as to how to perform the measurements (including how frequently to check BP, how long to wait before taking the readings, and how many readings to take each time the BP was measured). With regard to readings taken in their primary physician's office, the questionnaire asked about measurements taken by their primary physician and by a nurse or technician, if applicable. Questions included the amount of time they sat quietly before the first reading, the number of readings taken, and the type of BP device used (manual or automatic). Demographic information was obtained and the antihypertensive medications they were taking were recorded.

After the questionnaire was completed, office BP was measured by the participant's physician in usual clinical fashion, and is reported only for descriptive purposes. BP levels that had been obtained at home or at physician offices were not examined in this study.

Data Analysis

We summarized demographic and clinical characteristics using frequencies and proportions for categorical variables and means and ranges for continuous variables. The patient self‐report home, physician, and nurse BP measurements were examined using frequencies and response rates. Only patients who answered all questions of each part of the questionnaire were included in the section‐specific analysis. All analyses were conducted using SAS version 9.3 (SAS Institute Inc, Cary, NC).

Results

Demographic and clinical characteristics of the study participants are presented in Table 1. At their initial visit, participants were taking, on average, 2.1 antihypertensive medications, and their mean BP was 141/86 mm Hg.

Table 1.

Demographic and Clinical Characteristics of the Patients Enrolled in the Study (N=103)

Characteristics	No (%)
Age, mean (range) , y	55.2 (22–92)
Race
White	85 (82.5)
Black	8 (7.8)
Other	10 (9.7)
Male sex	58 (56.3)
First‐visit blood pressure
Systolic, mean (range) , mm Hg	141.3 (107–210)
Diastolic, mean (range) , mm Hg	85.5 (60–140)
Antihypertensive medications taken
None	15 (14.6)
1	19 (18.4)
2	34 (33.0)
3	18 (17.5)
4	14 (13.6)
≥5	3 (2.9)
Mean±standard deviation	2.1±1.4

Home BP Readings

Of the study's 103 participants, 82 (80%) reported that they had monitored their BP at home (Table 2). Among them, 29% measured their BP at least twice a day, 22% once a day, 22% one to three times a week, and 27% less than once a week. Sixty percent reported that they took the first measurements without waiting, 33% reported waiting 1 to 2 minutes, and only 7% reported waiting the recommended 5 minutes or more. Forty percent took only a single BP measurement, 35% two readings, and 24% three readings or more.

Table 2.

Patient Self‐Reports Regarding Home Blood Pressure Measurements (n=82)

Self‐Report On	No. (%)
Frequency of readings
≥2 Times a day	24 (29.3)
About once a day	18 (22.0)
1–3 times a week	18 (22.0)
Less than once a week	22 (26.8)
Waiting time before taking first measurement
None	49 (59.8)
1–2 minutes	27 (32.9)
≥5 minutes	6 (7.3)
Number of readings
Single reading, one arm	33 (40.2)
Single reading, each arm	14 (17.1)
2 readings, one arm	15 (18.3)
≥3 readings, one arm	19 (23.2)
≥2 readings, each arm	1 (1.2)

Eighty‐three patients of the study's participants (81%) responded to questions about the instructions they received from their primary or referring physician (Table 3). Among these, 64% reported that they had not received instructions on how frequently to measure their BP, 80% reported that they had not received instructions on how long to wait prior to taking the first measurement, and 83% reported having received no instructions as to how many readings to obtain when measuring their BP. One of the study's participants responded to questions about instructions from the physician but did not report details regarding compliance.

Table 3.

Patient Self‐Reports About Whether or Not They Received Instructions From the Primary or Referring Physician Regarding Home Blood Pressure Measurement (n=83)

Instructions Regarding	No (%)
How often to measure?
Yes	30 (36.1)
No	53 (63.9)
How long to wait before taking first measurement?
Yes	17 (20.5)
No	66 (79.5)
Number of readings?
Yes	14 (16.9)
No	69 (83.1)

Office BP Readings

As shown in Table 4, 64 (62%) of the 103 patients reported that their BP was routinely measured by the physician. Among these 64 patients, 77% reported that their physician did not wait at all before taking the first measurement. Thirty‐six (56%) reported that the doctor took only a single measurement.

Table 4.

Patient Self‐Reports Regarding Physician and Nurse Blood Pressure Measurements

	Physician (n=64),	Nurse or Technician (n=54),
Self‐Report On	No (%)	No (%)
Waiting time
None	49 (76.6)	52 (96.3)
A few minutes	15 (23.4)	2 (3.7)
Number of readings taken
One reading	36 (56.3)	44 (81.5)
Two or three readings	27 (42.2)	10 (18.5)
Four or more readings	1 (1.6)	0 (0)
Blood pressure monitor
Manual	50 (78.1)	31 (57.4)
Automated	10 (15.6)	21 (38.9)
Both	4 (6.3)	1 (1.9)

Fifty‐four patients (52% of the study's participants) reported that their BP was measured by a nurse or technician. Among them, 52 (96%) reported that there was no waiting from the time the cuff was placed to the measurement, and 44 (81%) reported that the nurse or technician took only a single measurement. A manual device was used by 57% of nurses and 78% of physicians.

Discussion

The degree of adherence in the clinical setting to the guidelines for measuring BP both in the physician's office and at home is largely unknown. The purpose of this study was to examine, based on patient recall, adherence to those guidelines, focusing on two clinically essential components of the guidelines: waiting time prior to measurement and number of readings taken. It is reasonable to suspect that a single reading taken without any waiting time could significantly overestimate or distort resting BP values. Nonadherence to these recommendations, as indicated by our results and reported elsewhere, could be widely affecting the diagnosis and management of hypertension.

Seventy‐seven percent of patients reported that BP measurement by the physician was performed without waiting and only a single reading was obtained in 56% of respondents. Similarly, only 21% of patients reported being instructed to wait before taking measurements at home, and only 17% were instructed how many readings to take at each session. Only 7% of patients reported that they waited the recommended 5 minutes, and 40% reported that they took only a single reading. Thus, there is substantial reason to believe that among many patients, both office and home readings differ from, and are likely higher than, readings taken according to guidelines.

Ironically, there are surprisingly few data underlying the widely cited recommendation to wait 5 minutes. Two recent studies address the drop in BP over time. Sala and colleagues16 reported that systolic pressure fell 11.6 mm Hg over a 16‐minute period, with half of that fall occurring in the first 5.8 minutes, and 75% of the fall over 10 minutes, after which further decrements in BP during 2‐minute intervals were not statistically significant. The largest fall in BP occurred in patients with the highest baseline BP.16 van Loo and colleagues17 reported a similar fall in systolic pressure (10 mm Hg) during the impractical waiting period of 25 minutes. In this context, the waiting period of 5 minutes has to be viewed as an arbitrary one, but one historically used in most large trials and consistently employed in studies for assessing hypertension status and cardiovascular risk.

The number of readings taken when measuring BP has also been documented to affect BP values. Among hypertensive participants in the 1999–2008 National Health and Nutrition Examination Survey (NHANES), in which BP measurement consisted of three readings, the third systolic BP reading averaged 2 mm Hg to 4 mm Hg lower than the first reading.11 In NHANES, 18% to 33% of participants classified as hypertensive based on the first reading were reclassified to a lower BP category after averaging in the second reading.11 In another report that also suggested overreading of BP due to nonadherence to measurement guidelines, office measurements taken by study personnel according to guidelines yielded readings that were 12/6 mm Hg lower than readings obtained during usual care.18 Further, regardless of the direction of change, because of the fluctuation of BP from reading to reading, a single reading is less likely to reflect a patient's usual BP than is the average of two or more readings.

In the current study, 66% of study participants were on multiple‐drug therapy for BP control. The high proportion of readings taken without waiting and the use of single readings suggest that resting BP levels are being overestimated, and possibly overtreated, in a considerable proportion of the millions of patients being managed for suspected hypertension. Finally, this study's sample was liited to patients with known or suspected hypertension. It did not include normotensive patients. A 5‐minute wait and repeat measurements are usually not necessary in a regularly normotensive patient whose first reading is well within the normal range; if reasings are elevated, repeat measurements afte rthe recommended waiting time would then be appropriate. In patients with controlled hypertension, repeat readings are also recommended because of the inherent variability of BP measurements.

Limitations

Clearly, better ways of assessing adherence to measurement guidelines are needed. It is difficult to observe waiting time and number of recordings in the physician's office without the physician being aware of it. Being observed could greatly affect how physicians obtain readings. Chart reviews are not an option as waiting time and number of readings are not generally recorded. In this study, such data were obtained based on patient recollection. This method has limitations, but nevertheless raises concern about how BP is measured in medical practices.

Another limitation is that even if the participant reported that the nurse or physician measured the BP immediately, he/she was likely sitting quietly in the waiting room prior to being seen. Thus, the immediate reading might have been at least partly mitigated by sitting time in the waiting room. Nevertheless, the physical exertion of arising and walking to the examination room and the sympathetic stimulation associated with that exertion likely require a new waiting period.

In addition, since a significant proportion of patients reported that both the nurse/technician and a physician obtained BP measurement, the physician's measurement in these patients may be the second measurement taken only after the nurse or medical assistant took the first reading. In these patients, the question of whether the readings were taken by the nurse and the physician at the same visit or in separate visits was not addressed.

Another possible limitation is that the study was performed with a sample of patients who were seeking care at a hypertension center at a major academic institution. This sample might not be representative of patients in primary care practices. It is likely that the proportion of patients monitoring BP at home was considerably higher among such patients than among primary care patients.

In this study, the effect of nonadherence to measurement guidelines on the BP readings obtained was not assessed. However, other studies have documented such effects.10, 15 Future studies will be challenged to examine the effect of nonadherence on the BP readings being obtained.

Conclusions

In this study, most participants reported that their BP measurements, both in the physician's office and at home, were obtained without adherence to guidelines regarding waiting time and number of measurements. This nonadherence could be expected to result in overdiagnosis and overtreatment of hypertension, as well as overuse of multiple‐drug therapies. The results suggest that nonadherence to measurement guidelines is common and may be considerably affecting the validity of BP measurements obtained both in physicians' offices and at home, with important clinical implications. Further study, with attention to methods to better evaluate physician and nurse adherence to BP measurement guidelines, is needed.

J Clin Hypertens (Greenwich). 2016;18:1157–1161. DOI: 10.1111/jch.12846. © 2016 Wiley Periodicals, Inc.