Unattended automated office blood pressure measurement: Time efficiency and barriers to implementation/utilization

Abstract

Unattended automated office blood pressure (BP) measurement (u‐AOBP) improves office BP measurement accuracy and reduces white‐coat BP elevation, but there are reservations about its time efficiency in primary care. We used time‐stamp methodology to measure u‐AOBP procedure times performed without a rest period in 130 patients during routine clinic visits to three primary care clinics with 2.5‐4.9 years u‐AOBP experience. We documented the clinical activities of 30 medical assistants during the u‐AOBP procedures. We also assessed MA and clinician satisfaction and knowledge about u‐AOBP performance and interpretation. Median u‐AOBP procedure time was <5 minutes, and MAs engaged in productive clinical activities during 83% of the procedures. Ninety‐three percent of MAs and 100% of clinicians in the clinics agreed that u‐AOBP is an efficient method to improve hypertension management. Barriers to effective u‐AOBP implementation and ongoing utilization included initial difficulty incorporating u‐AOBP into clinic workflow and medical staff knowledge deficiencies concerning correct u‐AOBP performance and interpretation despite prior training and experience with the procedure. Intensive u‐AOBP education and training programs are needed to facilitate effective u‐AOBP implementation into primary care. The time required to perform u‐AOBP can be utilized productively by staff.

1. INTRODUCTION

Because of the 15%‐30% prevalences of white‐coat blood pressure (BP) elevation1 and masked hypertension,2 out‐of‐office BP measurement is recommended to confirm the diagnosis of hypertension for most patients using either 24‐hour ambulatory or home BP monitoring.3, 4, 5, 6, 7, 8 Still, accurate office BP measurement (OBPM) remains essential to limit the need for labor‐intensive and expensive out‐of‐office BP monitoring. Unfortunately, OBPM in routine clinical practice is often performed inaccurately due to incorrect patient preparation and manual auscultatory measurement technique.9, 10 To improve OBPM accuracy, current guidelines recommend validated, automatic oscillometric devices to replace manual auscultatory BP measurement.3, 4, 5, 6, 7 But guideline‐recommended OBPM with these devices does not eliminate white‐coat BP elevation.10, 11 In contrast, validated, automated oscillometric devices that automatically perform and average three to five BP measurements over 3‐5 minutes while the patient rests alone in the exam room or another quiet place have been demonstrated to substantially reduce white‐coat BP elevation.12, 13 This approach to OBPM, termed unattended automated office BP measurement (u‐AOBP), is recommended as the preferred OBPM technique for patients with an initial elevated observed BP measurement by the 2019 American Heart Association Scientific Statement3 and by Hypertension Canada.6

Arguments against using u‐AOBP have included a lack of data about the cardiovascular prognostic significance of u‐AOBP and a perception that u‐AOBP may not be time‐efficient and feasible in clinical practice.14, 15, 16 While recent studies demonstrate strong associations of u‐AOBP with cardiovascular events and subclinical organ damage,17, 18 few studies have investigated the practicality of u‐AOBP in primary care.19, 20 We conducted a time efficiency study in three primary care clinics to measure u‐AOBP procedure time, medical assistant (MA) clinical activities during u‐AOBP performance, and MA and clinician satisfaction with and knowledge about u‐AOBP performance and interpretation.

2. METHODS

2.1. Study setting

We studied u‐AOBP measurements performed by MAs during routine patient clinic visits to two separate academic primary care practices (University Clinics 1 and 2) and one university‐affiliated community primary care practice (Community Clinic), all at different locations, from May, 2018 through March, 2019. Characteristics of the three clinics are listed in Table 1. In Table 1, note that mean MA experience with u‐AOBP was 2.5‐4.9 years in the three clinics with a range from 0.5 to 10.5 years for individual MAs. Therefore, AOBP protocols were well in place before this study began. The study was deemed exempt by the Institutional Review Board of the University of Utah Medical Center.

Table 1.

Characteristics of participating clinics

	University clinic 1	University clinic 2	Community clinic
Patient visits/year	7100	14 200	22 000
Clinician FTE	2.75	7.0	8.0
Formal u‐AOBP Roll‐out	February, 2016	February, 2016	November, 2016
u‐AOBP device(s) used	Omron HEM‐907 Welch Allyn Connex Spot BP	Omron HEM‐907 Welch Allyn Connex Spot BP	Welch Allyn Connex Spot BP
Number of MAs	5	10	20
Number of MAs monitored for u‐AOBP	5	9	16
Mean (range) years measuring u‐AOBP	2.8 (1.7‐4.5)	4.9 (0.5‐10.5)	2.5 (0.8‐4.5)
Number of patients monitored for u‐AOBP	26	47	67
Observation period	May‐June 2018	July‐August 2018	August 2018‐March 2019

2.2. Study design

MAs were notified at a staff meeting that an observer would follow them to evaluate general patient flow at multiple time points during entire routine clinic visits. MAs were not made aware that u‐AOBP time efficiency was the study focus. One of five trained observers not previously known to clinic staff recorded time‐flow data on an IPad tablet utilizing Excel Time stamp at the sequential time points listed in Table 2 and the Figure S1. Consecutive patients were followed until completion of at least four u‐AOBP procedures by each MA on patients with initial elevated observed BP. u‐AOBP procedure time was defined as the length of time from the MA leaving the room after AOBP device placement until re‐entering the room to document the u‐AOBP measurement.

Table 2.

Time‐stamp measures collected during patient visits

Weight: Patient steps on scale or refuses weight Patient enters room Vital sign measurement and completion: single observed BP, heart rate, temperature, and O2 saturation If u‐AOBP indicated due to elevated BP, then u‐AOBP explanation and initiation of u‐AOBP u‐AOBP procedure time: MA exits room MA re‐enters room MA records u‐AOBP value: EMR, sticky note for clinician, or verbally related to clinician Clinician enters room Patient exits room

Observers documented the percentage of u‐AOBP procedures during which MAs spent a majority of their time performing “clinical activities” (EMR charting, clinic phone calls, clinical interactions with other patients or with other members of the care team) versus “non‐clinical activities” (non‐clinic‐related conversation or computer activities).

A questionnaire (Table S1) was designed by the study investigators to determine MA and clinician knowledge about u‐AOBP performance, perceived u‐AOBP utility and efficiency, interpretation of AOBP results, and barriers to u‐AOBP implementation. Questionnaire items were determined by face validity. After completion of the entire study, the questionnaires were distributed to the MAs and clinicians and returned anonymously.

2.3. Blood pressure measurement

Office BP measurements and u‐AOBP were performed by trained clinic MAs using one of two validated oscillometric upper arm automated devices, the Omron HEM‐907 XL device,21 or the Welch Allyn Connex Spot BP22 in University Clinics 1 and 2; the Community Clinic used only the Welch Allyn device. An initial single observed BP measurement was performed without a defined rest period using appropriate cuff size for mid‐arm circumference, seated in a chair with back supported, cuff at midsternal level, arm supported on a flat surface, feet flat on the floor, with no conversation during the measurement. If the single observed BP measurement was ≥140/90 mm Hg or ≥130/80 mm Hg, according to individual clinician preference,4, 5 u‐AOBP was performed with the same device with the technique noted above but with the patient entirely alone in the exam room resting quietly (including no cell phone activity). The Omron HEM‐907 XL device used in University Clinic 1 was programmed to perform three BP measurements at 30‐second intervals with a 0‐second time delay and in University Clinic 2 to perform three BP measurements at 1 minute intervals with a 0‐second time delay.6, 23, 24, 25, 26 In all three clinics, the Welch Allyn Connex Spot BP was programmed to perform three BP measurements at 1 minute intervals with a 0‐second time delay.

2.4. Statistical analysis

Time‐stamp measurements (minutes, seconds) were converted to a continuous variable (minutes). Descriptive statistics were examined and due to non‐normal distribution data were presented as median (interquartile range (IQR)).

3. RESULTS

3.1. u‐AOBP procedure times

Across the three clinics, 30 MAs performed 140 u‐AOBP procedures on patients with an initially elevated observed BP. Individual time‐stamp u‐AOBP procedure times from device placement until the MA re‐entered the room are recorded in Figure 1. MAs prematurely re‐entered the clinic room prior to completion of the u‐AOBP procedure on 10 occasions in University Clinic 2 and the Community clinic, responsible for the very short procedure times in Figure 2. These procedure times were excluded from subsequent calculations. Figure 2 shows median (IQR) u‐AOBP procedure times in minutes in the remaining 130 patients in the three clinics as performed with a 0‐second time delay: 3.94 (1.63) Community Clinic, 3.79 (1.09) University Clinic 2, and 3.58 (0.81) University Clinic 1. Projected procedure times incorporating a 5‐minute delay would be 8.94, 8.79, and 8.58 minutes. The narrower range for AOBP procedure time noted in University Clinic 1 may result from their prior participation in a formal implementation study of u‐AOBP20 as well as workflow issues related to the larger number of MAs and clinicians in the other two clinics.

Figure 1.

AOBP procedure times in three primary care clinics

Figure 2.

Median u‐AOBP time‐stamp procedure times with 0‐second time delay

3.2. MA productivity during AOBP procedures

MAs engaged in productive clinical activities (EMR charting, clinic phone calls, paperwork, and clinical interactions with patients or other staff members while patients were alone in the exam room) during 83% of u‐AOBP procedures including 62% in University Clinic 1, 74% in University Clinic 2, and 97% in the Community Clinic.

3.3. Questionnaire results

Twenty‐eight of the 30 MAs and 15 of the 17 clinicians completed the anonymous questionnaire on u‐AOBP (Table S1). u‐AOBP procedures were almost always completed in the clinic room with only one MA also using a treatment room as an alternate performance site. Seventy‐one percent of MAs and 66% of clinicians indicated that u‐AOBP required minimal or no effort, and 93% of MAs and all clinicians agreed that u‐AOBP efficiently improved BP control for their patients. With respect to implementation barriers, one‐third of MAs and clinicians noted some initial difficulties incorporating u‐AOBP into clinic workflow, including orientation of new MAs and clinicians to the u‐AOBP procedure, patients arriving late to clinic with u‐AOBP deferred to the end of the visit, MAs forgetting to perform indicated u‐AOBP procedures, time lost having to clear spouses and relatives from the clinic room, and “occasionally” u‐AOBP causing clinicians to run late. Clinician, MA, or nurse resistance to using u‐AOBP was noted infrequently (14%) as a barrier to implementation.

Questionnaires revealed deficiencies in staff understanding of u‐AOBP performance and interpretation. Twenty‐one percent of MAs and 7% of clinicians reported performing u‐AOBP on all, most, or some patients rather than limiting u‐AOBP to patients with initially elevated observed BP.27, 28 Eleven percent of MAs and 33% of clinicians indicated that the patient was entirely alone in the exam room on just 30%‐69% of u‐AOBP procedures.12, 29 Only 36% of MAs and 27% of clinicians were aware that our clinic protocol and u‐AOBP device settings utilized a 0‐second time delay, the rest indicating that a 1‐5 minute rest period was in place.

Many clinicians and MAs did not recognize the u‐AOBP levels that correspond to guideline‐quality observed office BP and to mean daytime ambulatory BP on a 24‐hour study. Only 60% of clinicians and 21% of MAs correctly identified that a u‐AOBP of 135/85 mm Hg corresponds to an accurately measured, observed office BP of 140/90 mm Hg, and only 47% of clinicians and 18% of MAs recognized that a u‐AOBP of 130/80 mm Hg corresponds to an accurately measured, observed office BP of 130/80 mm Hg,30 the new office hypertension threshold recommended by the 2017 American College of Cardiology/American Heart Association hypertension guideline.4 Although u‐AOBP does not entirely eliminate white‐coat BP elevation,12, 29 40% of clinicians and 14% of MAs incorrectly indicated that an elevated u‐AOBP at two or more office visits confirms the diagnosis of hypertension without the need for out‐of‐office BP monitoring. Similarly, 47% of clinicians and 7% of MAs incorrectly indicated that an elevated u‐AOBP at one office visit warranted an adjustment of therapy in treated hypertension patients without resort to out‐of‐office BP monitoring.

4. DISCUSSION

This study investigated the time efficiency and medical staff satisfaction with u‐AOBP performed without a pre‐procedure rest period in a primary care setting. One hundred thirty u‐AOBP procedures were accomplished within a median 3.58‐3.94 minutes—projected to a median 8.50‐8.94 minutes had a prior 5‐minute rest period been utilized—which compares favorably to the 8‐14 minutes required to complete two or three guideline‐quality manual BP measurements.31 MAs were objectively engaged in productive clinical activities during over 80% of the AOBP procedures, further contributing to the time efficiency of u‐AOBP. Ninety‐three percent of MAs and 100% of clinicians agreed that u‐AOBP improves hypertension control. Time efficiency and medical staff satisfaction are important since u‐AOBP is now the preferred office method recommended by multiple organizations to screen for hypertension.3, 6, 32

We did find important barriers to overcome to successfully implement u‐AOBP. One‐third of MAs and clinicians reported initial difficulty incorporating u‐AOBP into clinic workflow, not surprising as most US MAs and primary care clinicians are not familiar with u‐AOBP rationale and technique. Overcoming this barrier will require dissemination of the 2019 American Heart Scientific Association Statement3 recommending u‐AOBP as the preferred office BP measurement approach and the development of u‐AOBP educational materials; we recently posted a u‐AOBP training video.33

Despite training and more than 2 year's experience with u‐AOBP in the three clinics, we found deficiencies in medical staff knowledge about u‐AOBP performance and interpretation. Fourteen percent of MAs reported performing u‐AOBP on all or most patients rather than limiting it to patients with initially elevated observed BP; this is a time‐inefficient approach since only about 3% of patients with initial observed BP at goal will have elevated BP on subsequent measurements.27, 28 MAs and clinicians indicated that u‐AOBP was not always performed with the patient entirely alone in the exam room, and on 10 of 140 u‐AOBP measurements, MAs re‐entered the room prior to u‐AOBP completion, both of which may result in white‐coat BP elevation. Imperfect adherence to BP measurement protocols despite intensive education and monitoring has been reported previously.3, 34 Regular reinforcement of correct u‐AOBP technique is essential to minimize performance deterioration and adjust for medical staff turnover.

Many medical staff may be incorrectly interpreting u‐AOBP results. Only 60% of clinicians and 21% of MAs recognized that a u‐AOBP of 135/85 mm Hg corresponds to a guideline‐quality observed office BP of 140/90 mm Hg,6, 11, 12 and 47% of clinicians and 18% of MAs that a u‐AOBP level of 130/80 mm Hg corresponds to a guideline‐quality observed office BP of 130/80 mm Hg,30 the latter the office BP threshold considered to warrant out‐of‐office BP monitoring by the 2017 American College of Cardiology/American Heart Association hypertension guideline.4 This incorrect interpretation may result in significant under detection of patients who would benefit from out‐of‐office BP monitoring to confirm the diagnosis of hypertension. Some medical staff appear to place undue reliance on elevated u‐AOBP results as 40% of clinicians and 14% of MAs incorrectly indicated that at an elevated u‐AOBP at two or more office visits confirmed the diagnosis of hypertension. Confirmatory out‐of‐office BP monitoring is currently recommended for patients with elevated u‐AOBP.3, 25, 29

Medical staff were uncertain about the use of a rest period prior to u‐AOBP, reflecting uncertainty among current studies and recommendations.3, 6, 23, 24, 25 Recommendations that u‐AOBP may be performed with “no special period of rest”6 have been based primarily on studies with the BpTRU device (no longer available) that takes six BP measurements and averages the last five.12, 13 There are currently few studies using no rest period with u‐AOBP devices that perform and average three BP measurements.12, 13 Two studies found good concordance of three measurement u‐AOBP performed using 0‐ and 1‐minute rest periods with mean awake ambulatory BP.35, 36 A recent study of 100 hypertensive persons compared mean systolic u‐AOBP using the Omron HEM‐907 XL device, performed with and without 5 minutes rest, to mean awake ambulatory systolic BP.37 In all 100 patients, three measurement mean systolic u‐AOBP performed with and without 5 minutes of rest did not differ significantly from each other (125.7 ± 17.9 mm Hg vs 127.0 ± 18.2 mm Hg, respectively) or from systolic mean awake ambulatory BP (126.6 ± 12.9 mm Hg). Similarly, in the 37 person subset with mean systolic u‐AOBP ≥ 130 mm Hg, mean systolic u‐AOBP with and without 5 minutes of rest correlated closely (143.3 ± 14.1 mm Hg and 144.3 ± 15.3 mm Hg, respectively); however, mean awake ambulatory BP in the 37 patients was significantly lower at 135.7 ± 11.9 mm Hg, a finding at variance from recent meta‐analyses. Despite the close correlations above of mean systolic u‐AOBP values with and without 5 minutes rest, these authors recommend that if initial systolic u‐AOBP is ≥130/80 mm Hg, u‐AOBP be repeated with 5 minutes rest. This recommendation is presumably based on their other finding of an order effect in 45 persons in whom mean systolic u‐AOBP performed first with no rest was 130.0 ± 17.7 mm Hg but follow‐up systolic u‐AOBP after 5 minutes rest was 126.5 ± 16.2 mm Hg. Given the greater time efficiency using no rest period, more studies comparing three measurement u‐AOBP devices with and without 5 minutes of rest are needed.

Study strengths include using three separate primary care clinics of different sizes and at different geographic locations, a large number of MAs performing multiple u‐AOBP procedures during usual clinical activities and unaware of the study focus, time‐stamp methodology to assure accurate timing of u‐AOBP procedures, and a 90% response rate among MAs and clinicians to the anonymous questionnaire.

A study limitation is that all three clinics had been using u‐AOBP for more than 2 years, and one clinic had participated in a published study using u‐AOBP,26 so our findings may not reflect the experience and initial difficulties of clinics newly implementing u‐AOBP. Our study did not collect data about actual observed BP and u‐AOBP levels and clinician actions in response to these BP levels, so we cannot compare actual clinician behavior with their questionnaire responses.

5. CONCLUSIONS

Because it is more accurate than casual BP measurement and reduces white‐coat BP elevation, u‐AOBP has been proposed as the preferred method to screen for elevated office BP. Concerns that u‐AOBP may be impractical and time‐inefficient in routine clinical practice were not confirmed in this study which found that u‐AOBP could be accomplished in <5 minutes without a prior rest period during which time MAs engaged in productive clinical activities. u‐AOBP may be more time‐efficient to clinic workflow than performing three guideline‐quality observed BP measurements over 8‐14 minutes. MA and clinician attitudes toward u‐AOBP were positive relative to feasibility, efficiency, and contribution to BP control.

Barriers to effective implementation and ongoing utilization of u‐AOBP include knowledge deficiencies concerning performance and interpretation. Initial difficulties incorporating u‐AOBP into clinic workflow and medical staff knowledge deficiencies about u‐AOBP utilization and interpretation require the development and dissemination of effective educational materials.

CONFLICT OF INTEREST

The authors have no conflicts of interest to disclose.

AUTHOR CONTRIBUTIONS

Study concept and design: All authors. Acquisition, analysis, or interpretation of data: All authors. Drafting of the manuscript: John Doane, Barry Stults, and Molly Conroy. Critical revision of the manuscript for important intellectual content: All authors. Statistical analysis: Marcus Archibald, Molly Conroy. Obtained funding: Molly Conroy. Administrative, technical, or material support: All authors. Study supervision: Barry Stults. Final approval: All authors. Accountability for all aspects of the work: All authors.

Supporting information

Fig S1

Table S1

Doane J, Flynn M, Archibald M, Ramirez D, Conroy MB, Stults B. Unattended automated office blood pressure measurement: Time efficiency and barriers to implementation/utilization. J Clin Hypertens. 2020;22:598–604. 10.1111/jch.13840