Use of automated office blood pressure measurement to reduce the white coat response Martin G. Myers a,c, Miguel Valdivieso a and Alexander Kiss b

Transcription

1 28 Original article Use of automated office blood pressure measurement to reduce the white coat response Martin G. Myers a,c, Miguel Valdivieso a and Alexander Kiss b Objective To examine the possibility of reducing the white coat response using an automated sphygmomanometer designed for office use, the BpTRU. Consecutive patients referred from physicians in the community to an ambulatory blood pressure (ABP) monitoring unit in an academic hospital were included in the study. Participants and methods A total of 39 patients referred for diagnosis or management of hypertension were studied. Differences between mean awake ABP and BP readings taken by the patient s own physician using a manual sphygmomanometer or the automated BpTRU device with the patient resting alone in the ABP monitoring unit were compared. Results BP recorded in the examining room using an automated device (132 W 19/75 W 12) was similar to the mean awake ABP (134 W 12/77 W 1) with both values being lower (P <.1) than the BP recorded on a routine visit to the patient s own family physician (152 W 18/ 87 W 11). The coefficient of correlation between the systolic/ diastolic ABP and the automated office BP (r U.62/.72) was higher (P <.1) than with the family physician s manual BP (r U.32/.48). The prevalence of white coat hypertension in untreated patients (n U 146) was significantly (P <.1) lower with automated office BP (16%) compared with the routine family physician BP (55%). Conclusion The white coat response associated with office BP measurements can be virtually eliminated by recording BP with the automated BpTRU device with patients resting alone in a quiet examining room. J Hypertens 27: Q 29 Wolters Kluwer Health Lippincott Williams & Wilkins. Journal of Hypertension 29, 27: Keywords: automated sphygmomanometers, blood pressure measurement, white coat hypertension Abbreviations: ABP, Ambulatory blood pressure; ABPM, ABP monitoring; BP, Blood pressure a Division of Cardiology, Schulich Heart Centre, b Department of Research Design and Biostatistics, Sunnybrook Health Sciences Centre and c Department of Medicine, University of Toronto, Ontario, Canada Correspondence to Dr Martin G. Myers, Division of Cardiology, Sunnybrook Health Sciences Centre, 275 Bayview Avenue, Room A2 2, Toronto, ON, Canada M4N 3M5 Tel: ; fax: ; martin.myers@sunnybrook.ca Received 7 April 28 Revised 17 September 28 Accepted 1 October 28 Introduction During the past decade, there has been a trend towards greater reliance on out-of-office blood pressure (BP) measurement for determination of an individual s BP status. Numerous studies have reported that 24-h ambulatory BP monitoring (ABPM) and self-measurement of BP in the home are better predictors of the risk of future cardiovascular events than conventional BP readings taken in the office [1,2]. As a consequence of these findings, recording of BP outside of the office or clinic has now been incorporated into the algorithms of guidelines for diagnosing hypertension [3,4]. Despite these advances, most physicians in routine clinical practice still rely on century-old technology for recording BP such as the mercury sphygmomanometer and aneroid devices. ABPM and home BP may be superior to the office BP, but these proven techniques may not always be feasible and ABPM is generally not practical for long-term management of hypertensive patients. During the past 5 years, a new approach to measure BP in the office has become available. Two devices designed for professional use, the BpTRU [5] and the Omron HEM 97 [6], have been used by health professionals to obtain BP readings in the office with minimal patient observer interaction, thus reducing the white coat response. Recently, a third device, the Microlife WatchBP Office, has been evaluated for possible use in clinical practice [7]. To date, most investigators have used the automated BpTRU device, which provides a mean of five readings taken at 1 or 2 min intervals, with the patient resting quietly alone in the examining room. Preliminary studies [8,9] have reported significantly lower office BP readings when taken with the BpTRU compared with the casual BP recorded in the office or research setting. In the present study, we compared routine BP readings taken in the offices of family physicians in the community with measurements performed using the BpTRU device with the patients resting alone in an examining room in an ß 29 Wolters Kluwer Health Lippincott Williams & Wilkins DOI:1.197/HJH.b13e32831b9e6b

2 Reduction of white coat response Myers et al. 281 ABPM unit, relating each to the mean awake ambulatory BP (ABP). Measurements obtained by the technician in the ABPM unit with a mercury sphygmomanometer were also examined. Methods Patient population BP readings were obtained in 39 consecutive patients referred by their family physician in the community for 24-h ABPM. Referring physicians were routinely asked to include the last BP reading taken by the patient s family physician or assistant (designated the FP BP reading) in the office prior to requesting ABPM. Basic demographic data were documented for each patient including age, sex and treatment status. Patients were instructed to take any hypertensive medication at their usual times. Written informed consent was not deemed to be necessary after review by the Institutional Ethics Review Board. Procedures Upon arrival in the ABPM unit, each patient was placed in a quiet examining room for a set of automated readings using the BpTRU model 1 (BpTRU Medical Devices Ltd., Coquitlam, British Columbia, Canada). The BpTRU was set to take readings at 1-min intervals (from the start of one reading to the start of the next) in 134 patients and at 2-min intervals in the remaining 175 patients. Subsequent analysis did not show any clinically important difference between readings taken at intervals of 1 versus 2 min [1]. The BpTRU device records BP using an oscillometric technique. Different cuff sizes were available to match the circumference of the patient s upper arm. The standard BP protocol was followed with the technician using the first reading to verify that the cuff was positioned properly and a valid BP reading recorded. The patient was then left alone seated in the examining room for 5 or 1 min and five more readings were taken. The mean of these five readings was noted as the automated office BP value. Patients were then connected to a SpaceLabs model 927 (SpaceLabs Healthcare Ltd., Issaquah, Washington, USA) 24-h ABPM unit. Two readings were taken simultaneously with the ABP recorder and by the technician using a mercury sphygmomanometer and a T-tube connector to verify the accuracy of the ABP device in the individual patient. The mean of the two readings taken by the technician was noted for comparison with readings taken in the family physician office and automated measurements. Patients were instructed to engage in routine daily activities during ABPM over the next 24 h. The ABPM device was set to record BP at 15- min intervals between 6 and 22 h and at 3-min intervals during the night. Mean awake ABP was then calculated according to the actual awake period as reported in each patient s diary. Statistical analysis The primary outcome measure was the difference between the routine BP as recorded in the office of the family physician and mean awake ABP versus the difference between the automated office BP using the BpTRU device and the mean awake ABP. Other outcome measures included the difference between the awake ABP and the manual recordings taken by the technician in the ABPM unit compared with the other readings. Mean SD was calculated for each of the four sets of BP readings. Pearson coefficients of correlation were computed for the manual and automated office BP versus the mean awake ABP. Individual data for the conventional family physician readings and automated office BP were displayed using a Bland Altman type plot [11]. Regression analysis was performed to predict the mean awake ABP from readings taken by the patient s own family physician. The proportion of untreated patients with white coat hypertension was obtained using systolic BP less than 14 mmhg and diastolic BP less than 9 mmhg as the normal value for manual office BP and less than 135 and less than 85 mmhg as the cut-point for normal awake ABP and automated office BP. The selection of 135/ 85 mmhg as the cut-point for normal automated office BP was based on the data analysis of Beckett and Godwin [8] derived from hypertensive patients being treated by primary care physicians in the community. The proportion of treated patients with a white coat effect defined as an office BP more than the mean awake ABP by at least 2 mmhg systolic and/or 1 mmhg diastolic [12] was determined for patients receiving antihypertensive therapy. The proportion of patients exhibiting masked hypertension was obtained with masked hypertension defined as an automated office systolic BP less than 135 mmhg and diastolic BP less than 85 mmhg and a mean awake ABP of 135 mmhg or more and/or 85 mmhg or more. All analyses were performed using SAS version 9.1 (SAS Institute, Cary, North Carolina, USA). Results The study population consisted of 39 patients, 139 men and 17 women, mean age years with 146 patients being previously untreated. Mean BP readings taken using both manual and automated devices are shown in Table 1. The routine office BP taken during the last visit to the patient s own family physician and the manual BP taken by the technician were significantly higher than the mean awake ABP, whereas the automated office BP taken in the ABPM unit using the BpTRU device was slightly lower than the mean awake ABP. The mean difference between the mean BP in the

3 282 Journal of Hypertension 29, Vol 27 No 2 Table 1 Mean (WSD) blood pressure (mmhg) values for mean awake ambulatory blood pressure compared with blood pressure readings performed in the ambulatory blood pressure monitoring unit and by the patient s own family physician ( P values relate to the ambulatory blood pressure) Mean blood pressure Blood pressure measure Systolic BP P Diastolic BP P Automated BpTRU office BP <.1 Manual technician office BP < <.1 Family physician routine office BP < <.1 Mean awake ABP (reference) ABP, ambulatory blood pressure; BP, blood pressure. family physician office and the mean awake ABP (18 18/1 11) was significantly (P <.1) greater than the difference between the mean BP taken using the automated BpTRU device and the mean awake ABP ( 2 15/ 2 8). All differences between the office BP and mean awake ABP remained statistically significant after adjustment for the Bonferroni correction. Similar findings were noted in the 146 patients who were not receiving antihypertensive therapy. In these patients, the mean awake ABP was /8 1 compared with the mean automated office BP of / The mean heart rate during the awake ABP recordings (72 13 beats/min) was the same as the mean heart rate noted during the automated office BP measurement (72 14 beats/min). Heart rate values were not reported from the offices of the family physicians. Individual data were displayed using a Bland Altman type plot with the mean awake ABP of individual patients plotted against the difference between the automated office BP and the mean awake ABP (Fig. 1) and the difference between the manual office BP taken in the family physician office and the mean awake ABP (Fig. 2). Positive differences in systolic BP of less than 1, 1 2 and more than 2 mmhg were present in 68, 63 and 129 patients for the manual family physician s BP and 57, 37 and 23 patients, respectively, for the automated BpTRU BP. Similarly, positive differences in diastolic BP of less than 5, 5 1 and more than 1 mmhg between the mean ABP and the family physician s manual office BP were present in 48, 68 and 136 patients and for the automated BpTRU in 49, 48 and 2 patients, respectively. Digit preference was also seen with the individual family physician readings. Systolic and diastolic BP values ended with a zero 62 and 56% of the time, respectively. Fig. 1 (a) BpTRU minus mean awake ambulatory systolic BP (b) BpTRU minus mean awake ambulatory diastolic BP Regression analysis predicting the difference between family physician or BpTRU readings and the mean awake ABP showed a significant (P <.1) negative relationship (slope parameter ¼.56) for systolic BP with the regression equation being FP BP ABP difference ¼ systolic ABP (Fig. 2). There was also a significant (P <.1) negative relationship (slope parameter ¼.49) for diastolic BP with the regression equation being FP ABP difference ¼ diastolic ABP (Fig. 2). In contrast, there was no significant relationship for the awake ABP and the difference between the systolic readings taken in the examining room with the BpTRU and the mean awake ABP (Fig. 1). However, there was a weak but statistically significant relationship (P <.1) for diastolic BP with the slope parameter being.2 and the regression equation diastolic ABP. Coefficients of correlation between the mean awake ABP and the manual office BP readings or automated BpTRU Mean awake ambulatory systolic BP (mmhg) Mean awake ambulatory diastolic BP (mmhg) Individual data for mean awake ambulatory systolic (a) and diastolic (b) blood pressure are plotted against the difference between the BpTRU readings and the ambulatory blood pressure. BP, blood pressure

4 Reduction of white coat response Myers et al. 283 Fig. 2 (a) Physician minus mean awake ambulatory systolic BP (b) Physician minus mean awake ambulatory diastolic BP readings are shown in Table 2. Both the manual office BP and the automated BpTRU readings correlated significantly with the mean awake ABP. However, the automated BpTRU readings showed a significantly (P <.1) better correlation with the awake ABP compared with the manual family physician office readings. The proportion of the 163 treated patients with a white coat effect (office systolic BP 2 mmhg and/or diastolic BP 1 mmhg versus mean awake ABP) was greater (P <.1) for the BP recorded in the family physician s office (65%) versus the automated BpTRU readings (12%) and the manual BP taken by the ABPM technician (26%). White coat effect was also more common with the Mean awake ambulatory systolic BP (mmhg) Mean awake ambulatory diastolic BP (mmhg) Individual data for mean awake ambulatory systolic (a) and diastolic (b) blood pressure are plotted against the difference between the family physicians readings and the ambulatory blood pressure. BP, blood pressure Table 2 Pearson coefficients of correlation are shown for mean awake ambulatory blood pressure (reference) versus the office blood pressure recorded by the BpTRU, technician or family physician Coefficient of correlation (r) Blood pressure measure Systolic BP Diastolic BP Automated BpTRU office BP Manual technician office BP Family physician routine office BP.32 a.48 a P <.1 versus either manual technician or automated BpTRU office blood pressure (BP). technician s readings (P <.1) than with the BpTRU. This magnitude of a white coat effect was considered to be clinically relevant, in that a difference of 2 mmhg systolic and/or 1 mmhg diastolic would likely lead to a change in drug therapy in treated patients. White coat hypertension in the 146 untreated patients was present in 55% based on manual readings taken by the patient s family physician, 18% for manual technician s readings and 16% for the automated BpTRU. White coat effect was more common using manual family physician readings (P <.1) compared with values obtained with the BpTRU device, whereas the manual technician readings were similar. Masked hypertension was present in 59 of 39 patients (19%) using 135/85 mmhg as the cut-point for normal versus high BP for both the automated office BP and the mean awake ABP. Discussion The findings in this study support the use of automated office BP measurement as an alternative approach to the diagnosis and management of hypertension. Readings taken with the automated BpTRU device with the patient left alone in a quiet room virtually eliminated the white coat component of the office BP with the mean values being similar to the mean awake ABP. BP recordings by the patient s own family physician during routine office visits were 18/1 and 2/12 mmhg higher than the corresponding mean awake ABP and automated BpTRU measurements. These differences have potentially important implications for patient care, both for making a diagnosis of hypertension in the individual patient and for avoiding unnecessary drug therapy in patients who do not have persistent hypertension. Indeed, the white coat response may lead to overtreatment of hypertension. In a previous report from our unit, 25% patients followed for 1 year with ABPM after discontinuation of antihypertensive therapy remained off treatment, suggesting that drug therapy had initially been started because of their white coat response [13]. Individual patient data obtained in the present study were also evaluated to determine the precise relationship

5 284 Journal of Hypertension 29, Vol 27 No 2 between the ABP readings and the difference one might anticipate finding between the routine manual office BP and the awake ABP. Regression analysis showed a strong negative relationship between these parameters for both systolic and diastolic BP when the office readings were taken manually. In contrast, the slope of the regression line was virtually zero for systolic BP and marginally negative for diastolic BP when the awake ABP was compared with differences between the automated office BP and the ABP. The plot of BP readings taken by the family physician (Fig. 2) demonstrates the impact the white coat response can have on the diagnosis of hypertension. The BP readings taken in the offices of the family physicians tended to be higher at lower awake ABP readings; thus, systolic ABP in the range of mmhg was often associated with hypertensive manual family physician readings. When the ABP was high, the family physician readings were as likely to be lower than to be higher. The same was true for the diastolic BP. In contrast, the slope of the BpTRU data did not exhibit such a strong negative relationship with the ABP values, suggesting that the observed differences between the BpTRU and awake ABP were more due to random fluctuation than due to any systematic measurement bias such as the white coat response. Given the overwhelming evidence supporting ABPM as a better predictor of cardiovascular outcomes than the office BP [1], any improvement that brings the office BP closer to the awake ABP is almost certain to enhance the clinical utility of BP measurement in routine clinical practice. Not only was the mean automated office BP within 2 mmhg of the ABP, but also the correlation between the awake ABP and the automated office BP was significantly better than that for the routine family physician office readings. Beckett and Godwin [8] have also reported a significantly better relationship between the automated office BP and BpTRU readings taken in the offices of family doctors in the community. In their series of 481 treated hypertensive patients, the mean automated office BP taken using the BpTRU (14/8 mmhg) was similar to the mean awake ABP (142/8 mmhg), with the mean office BP recorded by family physicians being significantly higher (151/83 mmhg). Automated office systolic BP correlated significantly better with the ABP (r ¼.571) compared with the correlation between the mean manual family physician systolic BP and the ABP (r ¼.145). The findings in the present study extend these results to untreated patients not receiving antihypertensive therapy. The mean awake ABP in the 146 untreated patients was similar to the automated office BP, with both values being substantially lower than the readings taken in the offices of the patients family physicians. The difference in BP in untreated patients was similar to the differences observed in the entire study population. The unique aspect of our study and that of Beckett and Godwin [8] is the comparison between what is possible using automated office BP measurement with patients alone in the examining room and the type of readings that are actually being obtained in routine clinical practice. If we had attempted to educate the patients physicians on how to improve their BP measurement techniques, the main objective of studying the white coat response in a real-life setting would have been distorted. Had primary care physicians been told that their patients were in a research study, they almost certainly would have started taking more careful BP measurements. This phenomenon, known as the Hawthorne Effect [14], has been reported in numerous studies of physician behaviour, including a previous report from our centre comparing physicians routine BP readings with readings taken for research purposes [15]. One of the reasons we included the manual BP readings taken by the ABPM technician was to show that it is indeed possible to measure BP more accurately using a mercury sphygmomanometer if proper guidelines are followed, especially if taken outside of the treatment setting. Our study is certainly not the first to show that BP measurement in routine practice is suboptimal. For example, in a study of 42 unselected hypertensive patients in the community, Gustavesen et al. [16] noticed a substantial difference between the routine office BP taken by the patient s family physician (167/15 mmhg) and a careful reading in the clinic taken for research purposes using a mercury sphygmomanometer (128/ 83 mmhg). If the benefits of automated office BP readings seen in our study were to be applied to routine clinical practice, the use of automated devices by family doctors would almost certainly improve BP measurement in the real-world setting. Analysis of the data from the untreated patients in our study also showed a significantly greater prevalence of white coat hypertension based on the family physician s manual BP reading compared with the automated BpTRU measurements. White coat hypertension was more than three times more likely to occur if measurements were taken manually by the patient s own family physician compared with automated BP readings taken in a quiet examining room in an ABPM unit. The same findings were noted for treated patients, with more than five times as many patients exhibiting a white coat effect when BP status was based on the routine manual office BP compared with readings taken with the BpTRU. The relatively high prevalence of a white coat response in this population may, in part, be due to patient selection. It is likely that many of the patients were sent for 24-h

6 Reduction of white coat response Myers et al. 285 ABPM because office-induced hypertension was suspected by their primary care physicians. Regardless of the prevalence of the white coat response, automated office BP measurement seems to eliminate the office induced increase in BP in most patients. If used more widely, devices such as the BpTRU should reduce the need for performing 24-h ABPM, especially in the followup of patients with a documented white coat component to their office BP. A variety of factors are likely to contribute to the white coat response when BP is recorded in the office of a physician using a mercury or aneroid sphygmomanometer. Le Pailleur et al. [17] have reported that conversation during BP measurement results in a marked increase in the BP level. This observer patient interaction is eliminated when an automated device such as the BpTRU is used to take readings. Health professionals frequently round off BP to the nearest zero as seen in the present study, which reduces the precision of the measurements. The venue for recording the BP may also affect the reading. In a previous report [18] from our unit, BP recorded manually on a visit to a hypertension specialist (174/92) was substantially reduced (158/9) when taken by a technician in an ABPM unit a week or two later. Unlike manual BP measurement, the setting may not influence the BP reading if it is taken with the BpTRU device. In a series of 62 patients, we have examined possible differences between readings taken with the BpTRU in the physician s office versus the ABPM unit and found that the mean values were virtually identical [19] and also similar to the mean awake ABP. In the present study, the prevalence of masked hypertension (19%) in the 39 patients referred for 24-h ABPM was within the range reported [2] for other patient populations (13 21%). It is not known how frequently masked hypertension would occur if automated office BP were to be recorded in a general hypertension population in the community. Another interesting aspect of automated office BP is the rapid fall in BP shortly after the patient is left alone in the examining room. In a recent report by Sala et al. [21], serial BP readings taken with a mercury sphygmomanometer over a 16-min period resulted in a decrease of 11.6/4.3 mmhg with 75% of the fall occurring within 1 min. The main advantage of automated office BP measurement with devices such as the BpTRU is that the fall in BP occurs much more rapidly. In a previous study [19], we observed that 75% of the decrease in BP, when taken by the BpTRU, occurred within 2 min of the patient being left alone in the examining room. In a more recent study [22], mean office BP taken three times at 1- min intervals using the Omron HEM 97 device was similar to the mean awake ABP. Thus, automated office BP measurement with patients resting alone in a quite examining room achieves the same level of BP as the mean awake ABP within 1 3 min without requiring the patient to remain resting for a prolonged period as seen in the study by Sala et al. [21], in which multiple readings were taken using a manual device in a researchtype setting. The use of automated sphygmomanometers such as the BpTRU does not eliminate all of the potential sources of error associated with BP in the office setting. However, the automated device does have the advantage of not talking with the patient, avoiding digit preference and, perhaps most important of all, taking readings with the patient being alone in a quiet examining room without the observer being present. Automated measurement of BP in the office has recently been proposed for possible use in routine clinical practice. In a recent editorial, Pickering [23] suggested that devices such as the BpTRU and Omron HEM 97 could be used to reduce the impact of the white coat response. In the latest American Heart Association monograph on BP measurement [24], the benefits of using automated devices that automatically give mean values of multiple readings were noted. Furthermore, in a recent position paper on the use of ABPM in primary care, O Brien [25] recognized the potential benefits of automated office BP measurement in the diagnosis and management of patients whose BP is affected by the white coat response. In summary, automated BP readings obtained in the office/clinic with a device such as the BpTRU are superior to the manual BP recorded in routine clinical practice for determining the BP status of patients. Demonstration of a closer relationship between the automated BP and mean awake ABP suggests that automated office BP should also be a better determinant of cardiovascular risk for the individual patient. The most important aspect of automated office BP measurement would seem to be the absence of the observer during the actual BP recording. No matter how carefully BP is taken, whether manually or with an automated device, the presence of a health professional in the room appears to produce a white coat response in some individuals. Measurement of BP in the office can still provide clinically useful information, especially if existing improvements in automated BP measurement are incorporated into routine clinical practice. Acknowledgements The study was supported in part by grant ESA 5745 from the Heart and Stroke Foundation of Ontario. The authors would like to thank Jacqueline Headley for secretarial assistance and Marko Katic for performing the data analysis.

7 286 Journal of Hypertension 29, Vol 27 No 2 References 1 Myers MG. Ambulatory blood pressure monitoring for routine clinical practice. Hypertension 25; 45: Stergiou GS, Kalogeropoulos PG, Baibas NM. Prognostic value of home blood pressure measurement. Blood Press Monit 27; 12: Hemmelgarn BR, McAlister FA, Myers MG, et al., for the Canadian Hypertension Education Program. The 25 Canadian hypertension Education Program (CHEP) recommendations for the management of hypertension. Part 1: Blood pressure measurement, diagnosis and assessment of risk. Can J Cardiol 25; 21: Pickering TG. JNC 7.5. J Clin Hypertension 27; 9: Wright JM, Mattu GS, Perry TL Jr, Gelfer ME, Strange KD, Zorn A, Chen Y. Validation of new algorithm for the BPM-1 electronic oscillometric office blood pressure monitor. Blood Press Monit 21; 6: White WG, Anwar YA. Evaluation of the overall efficacy of the Omron office digital blood pressure HEM-97 monitor in adults. Blood Press Monit 21; 6: Stergiou GS, Tzamouranis D, Protogerou A, Nasothimiou E, Kapralos C. Validation of the Microlife WatchBP Office professional device for office blood pressure measurement according to the International Protocol. Blood Press Monit 28; 13: Beckett L, Godwin M. The BpTRU automatic blood pressure monitor compared to 24 h ambulatory blood pressure monitoring in the assessment of blood pressure in patients with hypertension. BMC Cardiovasc Disord 25; 5:18. 9 Myers MG. Automated blood pressure measurement in routine clinical practice. Blood Press Monit 26; 11: Myers MG, Valdivieso M, Kiss A. Optimum frequency of automated blood pressure measurements using an automated sphygmomanometer. Blood Press Monit 28; 13: Bland JM, Altman DG. Statistics notes: some examples of regression towards the mean. BMJ 1994; 39: Myers MG, Reeves RA. White coat phenomenon in patients receiving antihypertensive therapy. Am J Hypertens 1991; 4: Myers MG, Reeves RA, Oh PI, Joyner CD. Overtreatment of hypertension in the community. Am J Hypertens 1996; 9: Sonnenfeld JA. Shedding light on the Hawthorne studies. J Occup Behav 1985; 6: Myers MG, Oh PI, Reeves RA, Joyner CD. Prevalence of white coat effect in treated hypertensive patients in the community. Am J Hypertens 1995; 8: Gustavesen PH, Hoegholma A, Bang LE, Kristensen KS. White coat hypertension is a cardiovascular risk factor: a 1 year follow-up study. J Hum Hypertension 23; 17: Le Pailleur C, Helft G, Landais P, Montgermont P, Feder JM, Metzger JP, Vacheron A. The effects of talking, reading, and silence on the white coat phenomenon in hypertensive patients. Am J Hypertens 1998; 11: Myers MG, Valdivieso MA. Use of an automated blood pressure recording device, the BpTRU, to reduce the white coat effect in routine practice. Am J Hypertens 23; 16: Myers MG, Valdivieso M, Kiss A. Reproducibility of automated office blood pressure measurements with comparisons to ambulatory blood pressure. Can J Cardiol 28; 24 (Suppl E):183E. 2 Verberk WJ, Kessels AGH, de Leeuw PW. Prevalence, causes, and consequences of masked hypertension: a meta-analysis. Am J Hypertens 28; 21: Sala C, Santin A, Rescaldani M, Magrini F. How long shall the patient rest before clinic blood pressure measurement? Am J Hypertens 26; 19: Myers MG, Valdivieso M, Kiss A, Tobe SW. A comparison of two automated sphygmomanometers for use in the office setting. Blood Press Monit (in press). 23 Pickering TG. Should doctors still measure blood pressure? J Clin Hypertens 26; 8: Pickering TG, Hall JE, Appel LJ, et al. Recommendations for blood pressure measurement in humans and experimental animals. Hypertension 25; 45: O Brien E. Ambulatory blood pressure measurement: the case for implementation in primary care. Hypertension 28; 51: