Home Blood Pressure Monitoring Devices: Device Performance in an Alaska Native and American Indian Population

Abstract

Objectives: Home blood pressure monitoring (HBPM) is an important component of blood pressure (BP) management. We assessed performance of two HBPM devices among Alaska Native and American Indian people (ANAIs). Methods: We measured BP using Omron BP786 arm cuff, Omron BP654 wrist cuff, and Baum aneroid sphygmomanometer in 100 ANAIs. Performance was assessed with intraclass correlation, paired t-tests, and calibration models. Results: Compared to sphygmomanometer, average BP was higher for wrist cuff (systolic = 4.8 mmHg and diastolic = 3.6 mmHg) and varied for arm cuff (systolic = −1.5 mmHg and diastolic = 2.5 mmHg). Calibration increased performance from grade B to A for arm cuff and from D to B for wrist cuff. Calibration increased false negatives and decreased false positives. Discussion: The arm HBPM device is more accurate than the wrist cuff among ANAIs with hypertension. Most patients are willing to use the arm cuff when accuracy is discussed.

Keywords

blood pressure home blood pressure monitoring device performance Alaska Native American Indian

Introduction

Alaska Native and American Indian (ANAI) people experience major health disparities related to hypertension and its health consequences, such as cardiovascular disease and stroke (Rhoades et al., 2007; Redwood et al., 2010). Hypertension prevalence is estimated at upward of 40% among ANAI adults (Howard et al., 2010; Jernigan et al., 2010; Jolly et al., 2013, 2015; Rhoades et al., 2007), compared to 32% among whites (Howard et al., 2018; Muller et al., 2019). Controlling blood pressure (BP) is a key factor in preventing hypertension-related morbidity and mortality, but successful control depends on sustained patient engagement with a clinical care team (Lee & Park, 2016; Milani et al., 2016). Among people with hypertension, ANAIs have a lower prevalence of adequate BP control; one study reported that 34% of ANAIs with hypertension were in good control (defined as BP ≤ 140/80 mmHg), compared to 47% of Blacks and 61% of whites (Muller et al., 2019; Zhang & Moran, 2017). Home blood pressure monitoring (HBPM), in which patients use portable automated devices to measure their BP at home, has been a key strategy to engage patients and clinicians in increasing medication adherence and, thus, improving BP control over time. HBPM avoids errors due to in-office measurements including white-coat hypertension, not recognizing “masked hypertension,” or BP variability over the course of the day (in relation to time of day or time since antihypertensive medications; O’Brien et al., 2013). HBPM is also especially important for achieving good BP control among patients who experience barriers to accessing high-quality health care, including underserved minorities and rural residents.

In addition to clinical performance, the value of HBPM for BP management is influenced by behavioral and healthcare utilization factors that can vary by race and culture (Joe, 2003), yet no research has studied HBPM for hypertension management in ANAIs. Tribal and clinic stakeholders were concerned about the lack of evidence-based information on HBPM device performance in ANAIs to inform choice and use of upper arm or wrist cuff HBPM devices. In response to community interest in using wrist rather than upper arm (brachial) HBPM devices, we conducted a longitudinal crossover study in preparation for a larger group-randomized clinical trial. Our goal was to evaluate performance of two HBPM devices, each previously validated according to British Hypertension Society or International protocols—one with an upper arm cuff and one with a wrist cuff—for identifying uncontrolled BP in urban ANAI adults with hypertension (Whelton et al., 2018). This study was conducted as part of “Blood Pressure-Improving Control among Alaska Native People” (BPICAN), to test the effect of a multilevel intervention on improving BP control among ANAIs with hypertension (BPICAN, 2019). We hypothesized that (1) upper arm cuff devices would perform better than wrist cuff devices, but that values from the latter could be calibrated with patient data to perform adequately based on existing standards and (2) hypertension misclassification would differ by the HBPM device based on relative proportion of false positives (erroneously identifying high BP) and false negatives (failing to identify high BP) in ANAIs with self-reported hypertension.

Methods

Setting

Southcentral Foundation (SCF) is a nonprofit, tribally owned, and operated healthcare organization that provides prepaid primary care services to ∼65,000 ANAI people living in Anchorage, the rural Matanuska-Susitna Borough, and 55 remote villages (Eby, 2007; Gottlieb, 2013). Affiliated tribal and institutional review boards approved this study and dissemination (Hiratsuka et al., 2017).

Eligibility and Recruitment

Study participants were recruited in-person at SCF or by electronic media advertisement from February through April 2018. Eligibility required (1) at least one visit with an SCF provider in the past year; (2) self-identification as ANAI; (3) aged 18 years or older; and (4) self-reported hypertension. Self-reported hypertension was used as an eligibility criterion because it reflects a patient population that would likely use HBPM. Limiting eligibility criteria to individuals with diagnosed hypertension was rejected due to the high prevalence of undiagnosed hypertension in ANAIs and because people without a hypertension diagnosis may use HBPM in ways that influence how they utilize healthcare services if they have concerns about their BP. Pregnancy was an exclusion criterion due to common differences in HBPM device accuracy in pregnant women (Bello et al., 2018). Although we did not intentionally exclude rural patients, recruitment was conducted at the primary SCF facility in Anchorage, and all participants lived in urban or suburban locations. Written informed consent was obtained at the first study visit.

Study Design and Procedures

Study participants completed three visits evenly spaced over 2 weeks in a longitudinal crossover study. At each visit, systolic BP (SBP) and diastolic BP (DBP) were measured using three devices: (1) a calibrated aneroid sphygmomanometer as the gold standard (Baum-Desk Aneroid 0910), (2) automated upper arm cuff oscillometric BP monitor (Omron 10-Series BP786), and (3) automated wrist cuff oscillometric BP monitor (Omron X-Series BP654). Clinical standards for BP measurement (O’Brien et al., 2003; Williams et al., 2009) and standard American Heart Association arm cuff size thresholds were used for all participants (Pickering et al., 2005). Device order was randomized, and three measures were taken on each device before moving to the next one.

All study staff were trained and certified by study physicians, using a standardized protocol (O’Brien et al., 2003). Before certification, staff had to demonstrate their ability to identify 83% of Korotkoff sounds within 2 mmHg by sphygmomanometer, to adhere to clinical standards for BP measurement by auscultation, and to properly operate each HBPM device (MONICA Manual, 1997; Williams et al., 2009).

Because the primary purpose of this crossover study was to estimate the reliability and validity of the automated upper arm and wrist HBPM devices, the power calculation was based on SBP—the primary outcome in the main trial, BPICAN. Using SDs found in the literature (8 mmHg) and with 100 participants, we expect at least 90% power for a minimum detectable effect of 2 mmHg. These estimates are conservative given that with repeated measures per participant, power is expected to be higher.

Measures

At each study visit, BP for each device was recorded as the average of the second and third readings (O’Brien et al., 2003; Pickering et al., 2005). Other characteristics assessed included patient demographics, mid-upper arm circumference, upper arm length (acromion process to olecranon processes), and wrist circumference. To measure misclassification, we created variables indicating qualitatively opposite determinations of uncontrolled BP from the HBPM device compared to the sphygmomanometer. We created separate variables for the pre-2017 threshold (BP ≥ 140/90 mmHg) and for the 2017 guidelines released by the American Heart Association (≥130/80 mmHg; Reboussin et al., 2018). False positive readings reflected HBPM values higher than the threshold and sphygmomanometer value under the threshold (HBPM erroneously identifying high BP); false negative readings reflected HBPM value under the threshold and sphygmomanometer value higher than the threshold (HBPM failing to identify high BP).

Statistical Analysis

We calculated descriptive statistics for each variable using means (standard deviation (SD)) for continuous variables and counts (percentage) for categorical variables. Precision of BP measured by each device was assessed using intraclass correlation coefficients (ICCs), which were estimated via linear mixed effect models adjusted for device order. Intraclass correlation coefficient values could range from 0 to 1, with higher numbers reflecting more precision for within-device measurements. Accuracy was assessed by three methods. We used a paired t-test to compare the mean signed difference (tendency to measure high or low compared to sphygmomanometer) and mean absolute difference (deviation from the sphygmomanometer without respect to directionality) for the two HBPM devices (Stergiou et al., 2018); we created Bland–Altman scatterplots to assess whether accuracy differed across the range of BP values represented in the cohort (Bland & Altman, 1986, 1995); and we assigned a letter grade based on British Hypertension Society thresholds for automated HBPM devices used in research (O’Brien et al., 1993). Grades reflect the percent of values that fall within 5 mmHg, 10 mmHg, and 15 mmHg of the sphygmomanometer readings as follows: 60%, 85%, and 95% = A; 50%, 75%, and 90% = B; and 40%, 65%, and 85% = C.

We used linear regression to develop calibration models with the crude HBPM value (model 1) as the independent variable and the sphygmomanometer BP value as the dependent variable (Cao et al., 2015). Another regression was done with the crude HBPM value plus patient age, sex, and arm or wrist circumference (model 2) as independent variables and the sphygmomanometer BP was the dependent variable (Cao et al., 2015). Calibration was implemented by estimating each patient’s predicted value of their true BP based on coefficients for independent variables. We built the models with a randomly selected subset of 67 participants (192 study visits) and reserved the remaining 33 (91 study visits) for validation (Hastie et al., 2009). Lastly, we created bar charts to depict differences in misclassification of BP control by each HBPM device based on the pre-2017 (140/90 mmHg) and 2017 (130/80 mmHg) guidelines. All analyses were conducted using StataCorp statistical software (version 15.1).

Results

We identified 204 potential participants over 3 months, of which 100 met eligibility criteria. Of the 100 eligible participants, 94 and 89 participants returned for the second and third study visits, respectively. Descriptive statistics are shown in Table 1. On average, participants were 51 years old and 60% were female. Six participants had upper arm circumferences that exceeded the manufacturer’s recommendations (≤43.2 cm), and seven participants had wrist circumferences that surpassed manufacturer’s reference values (≤21.5 cm). Three participants exceeded both upper arm and wrist device reference values for appropriate fit. We did not exclude individuals who exceeded manufacturer thresholds for arm or wrist circumference because these individuals may still choose to use HBPM as a BP management strategy.

Table 1.

Demographic Characteristics of Participating ANAI Patients, Southcentral Foundation (2018).

Characteristic	(N = 100)
Age in years, mean (SD)	51 (12)
Female, %	60
Education, %
< High school graduate	6
High school graduate or GED	30
Some college	44
≥ College graduate	20
Married or living with partner, %	50
Yearly household income, %
< $15,000	12
$15,000–24,999	16
$25,000–34,999	16
$35,000–59,999	32
≥ $60,000	24
Upper arm length in cm, mean (SD)	34 (3)
Arm circumference in cm, mean (SD)	35 (6)
Wrist circumference in cm, mean (SD)	18 (2)
Sphygmomanometer BP in mmHg, mean systolic/diastolic (SD)	133/80 (14/11)
Wrist cuff BP in clinic, mean systolic/diastolic (SD)	139/85 (20/15)
Upper arm cuff BP in clinic, mean systolic/diastolic (SD)	131/84 (17/12)

Notes. ANAI = Alaska Native and American Indian; SD = standard deviation; GED = General Equivalency Diploma; BP = blood pressure.

As shown in Table 2 (left), precision of HBPM readings for SBP and DBP was similar for the upper arm (both ICC = 0.91) and wrist cuff (ICC = 0.91 and 0.93, respectively) devices. The clinical aneroid sphygmomanometer with trained and certified operator had the best precision for both SBP (ICC = 0.98) and DBP (ICC = 0.97).

Table 2.

Precision and Accuracy of Automated Home Blood Pressure Monitoring (HBPM) Devices in ANAI Patients, Southcentral Foundation (2018).

Device			Accuracy^b
	Precision^a		Signed Difference		Absolute Difference
	ICC	(95% CI)	Mean^c	(95% CI)	Mean^c	(95% CI)
Wrist cuff
Systolic	0.91	(0.88, 0.93)	4.8	(3.1, 6.6)	9.4	(8.2, 10.5)
Diastolic	0.91	(0.88, 0.93)	3.6	(2.3, 5.0)	7.0	(6.1, 7.9)
Upper arm cuff
Systolic	0.91	(0.88, 0.93)	−1.5	(−2.8, −0.3)	6.6	(5.9, 7.4)
Diastolic	0.93	(0.91, 0.95)	2.5	(1.6, 3.4)	5.0	(4.3, 5.5)
Sphygmomanometer
Systolic	0.98	(0.97, 0.98)	Gold standard		Gold standard
Diastolic	0.97	(0.96, 0.98)

^aPrecision: intraclass correlation coefficients, intra-device across second and third measurements at three visits; values represent ICC point estimates.

^bAccuracy: comparing automated HBPM devices to the sphygmomanometer across second and third measurements at three visits; within-person comparisons for blood pressure measurements between devices.

^cSigned difference calculated by averaging all differences (includes negative and positive values, below, and above the sphygmomanometer, respectively); absolute difference calculated by averaging the absolute value of all differences.

Note. ICC = intraclass correlation coefficient.

The analysis of HBPM device accuracy (Table 2, right) showed that, on average, wrist cuff SBP measures were higher than from the sphygmomanometer (mean signed difference = 4.8 mmHg, 95% CI: [3.1, 6.6]), whereas the upper arm cuff SBP measures were lower on average (mean signed difference = −1.5 mmHg, 95% CI: [−2.8, −0.3]). The mean absolute difference was 9.4 mmHg (95% CI: [8.2, 10.5]) for the wrist cuff and 6.6 mmHg (95% CI: [5.9, 7.4]) for the upper arm cuff. For DBP, mean values for both the wrist and upper arm cuff devices were higher than the sphygmomanometer, and the mean absolute difference was larger for the wrist cuff than the upper arm cuff. No differences in accuracy were found in a sensitivity analysis when analyses excluded the subset of participants with upper arm or wrist circumference that exceeded manufacturer recommendations. Bland–Altman scatterplots showed the wrist cuff tended to underestimate BP at lower values and overestimate BP at higher values for both SBP and DBP (Figure 1). In contrast, the upper arm cuff had a relatively consistent difference from the sphygmomanometer across the range of values for SBP and DBP.

Figure 1.

Scatterplot showing signed difference versus average of the sphygmomanometer and automated home blood pressure monitoring device. Horizontal lines on the plot represent the mean signed difference and the limits agreement, which are defined as (mean difference ±1.96*SD of the difference).

With no calibration, the HBPM wrist cuff device performed at grade D by the British Hypertension Society standards (Table 3; O’Brien et al., 1993). Based on absolute mean difference values, performance improved to grade C with calibration model 1 and to grade B with calibration model 2. For SBP, the upper arm cuff performed at grade B; calibration resulted in no change or improved performance to grade A, depending on the model. For DBP, the wrist cuff improved from grade C to grade A with calibration, whereas the upper arm cuff device performed at grade A throughout.

Table 3.

Home Blood Pressure Monitoring Device Systolic and Diastolic Performance grade^b With Respect to British Hypertension Society Standards for Models Without and With Calibration, Southcentral Foundation (2018).

	Systolic Blood Pressure				Diastolic Blood Pressure
	≤5 mmHg (%)	≤10 mmHg (%)	≤15 mmHg (%)	Grade^a	≤5 mmHg (%)	≤10 mmHg (%)	≤15 mmHg (%)	Grade^a
Wrist cuff, %
No calibration^b	38	64	81	D	48	76	88	C
Model 1^b	49	79	91	C	63	89	95	A
Model 2^b	51	79	92	B	56	90	96	B
Upper arm cuff, %
No calibration^b	51	81	92	B	61	88	97	A
Model 1^b	57	82	95	B	62	90	99	A
Model 2^b	62	85	97	A	65	93	99	A

^aFull sample used for model with no calibration model (n = 283 measures; n = 17 lost to follow-up); test sample (n = 91 measures) used for calibrated models.

^bGrades are derived from percentages of readings within 5, 10, and 15 mmHg of the sphygmomanometer readings as follows: 60%, 85%, and 95% = A; 50%, 75%, and 90% = B; 40%, 65%, and 85% = C. All percentages must be equal to or greater than the categorized values.

Misclassification of uncontrolled hypertension was generally lower with the upper arm cuff device than the wrist cuff device (Figure 2). Before calibration, the upper arm device also yielded lower percentage of false positives for both the pre-2017 (14% vs. 26%) and 2017 (19% vs. 28%) thresholds. Calibration reduced the percentage of false positives for both devices. In contrast, the uncalibrated wrist cuff device values yielded the lowest percentage of false negatives for the pre-2017 (14% vs. 21%–41%) and 2017 (7% vs. 10%–12%) thresholds. For current threshold, for uncontrolled BP (130/80 mmHg), the joint levels of misclassification were minimized with calibration model 2 of the upper arm cuff device (false positive = 9% and false negative = 11%).

Figure 2.

Misclassification of hypertension status for arm and wrist HBPM devices compared to the sphygmomanometer among ANAI patients, for pre-2017 and 2017 American College of Cardiology/American Heart Association guidelines, Southcentral Foundation, 2018. 1. Systolic blood pressure ≥140 mmHg or diastolic blood pressure ≥90 mmHg. 2. Systolic blood pressure ≥130 mmHg or diastolic blood pressure ≥80 mmHg. 3. Model 1 includes crude HBPM value as only predictor in the calibration model. 4. Model 2 includes crude HBPM value, age, sex, and arm circumference as predictors in the calibration model. 5. False positive: HBPM erroneously identifying high BP compared to the sphygmomanometer; false negative: HBPM failing to identify high BP compared to the sphygmomanometer. Notes. HBPM = home blood pressure monitoring; ANAI = Alaska Native and American Indian.

Discussion

We examined the performance of two HBPM devices for detecting uncontrolled BP in ANAIs with self-reported hypertension. We found that precision was high for both devices. The upper arm cuff was more accurate overall, but after calibration with data on age, sex, and wrist circumference, the wrist cuff device achieved at least a B grade based on existing standards. These findings indicate that the upper arm cuff performs adequately for use in health research with ANAIs and that the wrist cuff performs adequately after calibration with easily obtained patient data. Our results are consistent with validation studies showing that upper arm HBPM devices have better performance metrics (Cao et al., 2015; European Society of Hypertension-European Society of Cardiology Guidelines Committee, 2003; Jung et al., 2015; Melville et al., 2018; Kikuya et al., 2002; Stergiou et al., 2008; Weber et al., 2014), but most of the study samples comprised Asian or white participants. HBPM device performance is influenced by factors that may vary across culture and race, such as cuff fit secondary to obesity prevalence and preference (Ogden et al., 2015; Zamora-Kapoor et al., 2019). Thus, it is important that minority groups, including ANAIs, are represented in HBPM device performance literature given devices are distributed in clinical settings independent of race or ethnicity status.

The COVID-19 pandemic has catalyzed critical attention to persistent health inequities in marginalized communities (Wiemers et al., 2020). Prevalent hypertension is a common comorbidity in individuals hospitalized with COVID-19, with current prevalence estimates ranging between 17 and 31% (Guan et al., 2020a, 2020b; Wang et al., 2020). Controlled hypertension prevalence decreased to 44% in 2018 (Muntner et al., 2020) and has continued to decrease throughout 2020 (Curfman et al., 2020). Alaska Native and American Indian communities have been disproportionately affected by the COVID-19 pandemic; as of late January 2021, COVID-19 hospitalization rates were 6.4 times higher in non-Hispanic ANAI adults 18–49 years of age compared to non-Hispanic whites (CDC, 2021). Wiemers et al. (2020) succinctly concluded: “vulnerability based on pre-existing health conditions collides with long-standing U.S. health disparities” by race-ethnicity and socioeconomic status (p. 1). Concerns about COVID-19, reduced healthcare access, and financial uncertainty have led many people to either avoid or delay medical care for acute and chronic health conditions, including hypertension (Czeisler et al., 2020). International experts strongly recommend careful attention to BP control, in addition to COVID-19 precautions, to minimize risk for individuals with hypertension (Adams & Wright, 2020; Clark et al., 2020). Given that HBPM is a successful strategy to controlling BP, our results add timely reinforcement that arm cuff devices perform well in an ANAI population and may be one successful approach to BP control during the COVID-19 pandemic.

Conventional approaches to validating device performance treat all deviations from the gold standard as equal, regardless of directionality. In our analysis, the upper arm cuff misclassified fewer BP values than the wrist cuff overall. Because the health and economic consequences of false positives and false negatives are different, clinicians and researchers should consider the relative percentages of both before choosing the optimal HBPM device for any specific purpose. For example, the uncalibrated wrist cuff values yielded the lowest percentage of false negatives (7% based on current treatment guidelines), suggesting the lowest risk of failing to detect uncontrolled BP in ANAIs with hypertension. The uncalibrated wrist device also yielded the highest percentage of false positives (38%), which could increase adverse consequences such as unnecessary healthcare visits and patient anxiety.

This is the first study to examine the performance of HBPM devices for BP measurement in ANAIs with hypertension. Alaska Native and American Indian people experience persistent health disparities in conditions, such as cardiovascular disease and stroke, that are strongly influenced by BP (Fryar et al., 2017; National Center for Health Statistics, 2017), and their access to clinic-based health care is negatively affected by factors such as lack of transportation and lack of specialized medical providers within reasonable driving distance (Joe, 2003). HBPM can be extended to telehealth operations that have drastically increased with COVID-19 (Omboni et al., 2020). Indeed, a recent study found that when physicians used HBPM values to titrate antihypertensive medication, their patients achieved better BP control than when physicians relied solely on BP measured during clinic visits (Tucker et al., 2017). Thus, HBPM has the potential to improve BP control while minimizing the need for clinic visits in ANAIs with hypertension, especially those residing in the many remote regions of Alaska. The success of HBPM for BP control also depends on other factors such as patient–provider communication and sustained patient engagement with regular home monitoring. These factors can differ by race and ethnicity (Wisniewski & Walker, 2020; Woo et al., 2020), yet most minority populations are underrepresented—or entirely absent—from research on HBPM as a tool for managing hypertension (Uhlig et al., 2013). A 2017 systematic review and meta-analysis concluded that the overall effect of HBPM on BP control is heavily influenced by the presence and intensity of additional “co-interventions” such as antihypertensive medications and lifestyle modifications (Tucker et al., 2017). The ongoing BPICAN trial that motivated the current study will be the first to address these topics in a large cohort of ANAIs with hypertension.

Our findings have potential importance for ANAIs living in rural communities, where access to healthcare services (e.g., medication titration visits) and opportunities for behavioral change (e.g., gyms and healthy food options) may be especially limited (Carey et al., 2018; Doyle et al., 2019; Havranek et al., 2015). Notably, a higher percentage of ANAIs live in rural areas (54%; Dewess & Marks, 2017) compared with the general population (19%; US Census Bureau, n.d.); therefore, the feasibility and effectiveness of HBPM for BP control for rural ANAIs is worth careful consideration. Nonetheless, even in urban and suburban areas such as Anchorage, purchasing healthy foods can be cost prohibitive, and weather has a sustained impact on physical activity opportunities in the winter. Furthermore, the extreme climate, rugged terrain, and expanse in Alaska, combined with limited infrastructure and a population density that is dramatically lower than in the contiguous United States (1 vs. 84 people per square mile), greatly hinders access to health care in rural communities (Golnick et al., 2012; Murphy et al., 1997). Sixty percent of Alaska residents are medically underserved, and in 75% of Alaskan communities, comprehensive healthcare services are accessible only by air or water (Gottlieb, 2013). Not surprisingly, cardiovascular disease risk factors, such as hypertension, are more prevalent in rural communities and are less likely to be controlled (CDC, 2017; Chow et al., 2013; He et al., 1991; Kapral et al., 2019). Accordingly, we are currently conducting a pilot study to evaluate the feasibility of a BPICAN intervention tailored to the needs of rural ANAI patients at SCF.

There are several limitations to this study. First, it was conducted exclusively in Anchorage with urban and suburban patients at SCF. While the Anchorage population is generally representative of the larger Alaska population, these results may not be generalizable to other ANAI populations or other racial groups. Second, selection bias is a concern. For example, participants who sought care at SCF may have better access to health care than the overall population of ANAIs with hypertension and, therefore, may have more controlled hypertension. It is uncertain whether HBPM devices would have the same performance characteristics in a hypertensive population with higher average BP. Third, BP readings for each device were measured sequentially instead of alternating readings between devices. As a result, we were unable to account for change in BP over time at each visit, independent of device. However, device order was randomized across participants, reducing concerns about bias introduced by this study design feature. Fourth, BP for all devices were taken by trained study staff using a standardized research protocol, and results may not accurately reflect the HBPM measurements taken by patients at home where strict protocol adherence is challenging. Thus, this should be viewed as an efficacy study that informs the potential for HBPM to improve BP self-management and health outcomes among ANAIs with hypertension. Future studies, such as the ongoing BPICAN trial, can evaluate and strive to optimize HBPM effectiveness in the home setting. Fifth, our study evaluated HBPM device performance at the population level and should not be interpreted as reflecting how devices would perform for any individual patient. Nevertheless, population-level data are needed for decisions related to public health, such as which interventions and medical devices to offer within a healthcare system and how best to assess health status and outcomes in research studies.

Strengths of this study include its focus on ANAIs, a population completely absent from the extant literature about the accuracy of HBPM devices. Second, repeated BP measures from three study visits over 2 weeks increase the stability of the intraindividual variation when assessing device performance. Third, BP measurement error was low because all staff who obtained BP readings were trained and certified by research physicians to ensure proper use of equipment and interpretation of sphygmomanometer readings. Fourth, device order was randomized to control for attenuating BP levels when measured repeatedly in a single sitting. Fifth, we included evaluation of two calibration models and considered findings in terms of clinical value, rather than treating all misclassification as equal from a healthcare perspective. Our findings can help healthcare systems make decisions about integrating HBPM into routine patient care and can provide decision support to providers and patients who are making initial choices between available devices.

In conclusion, HBPM should be studied as a tool for improving BP control and eliminating health disparities in ANAIs with hypertension. Our findings suggest that the upper arm HBPM cuffs perform adequately for use in health research with ANAIs, and that the wrist HBPM cuff performs adequately after calibration with easily obtained patient data. Future efforts should consider cultural, environmental, behavioral, and personal preference factors that may affect performance and usability of HBPM among ANAIs in Alaska and in other communities across the United States.

Footnotes

Acknowledgments

The authors are grateful to the members of the Southcentral Foundation and Alaska Native Tribal Health Consortium research review committees for their continued review of research at the Alaska Native Medical Center campus and to the Community Advisory Board for their guidance on this study. The authors thank the Alaska Native and American Indian participants in the study.

Declaration of Conflicting Interests

The author(s) declared no potential conflicts of interest with respect to the research, authorship, and/or publication of this article.

Funding

The author(s) disclosed receipt of the following financial support for the research, authorship, and/or publication of this article: This work was supported by the National Institute on Minority Health and Health Disparities of the National Institutes of Health [U54MD011240]. The content is solely the responsibility of the authors and does not necessarily represent the official views of the National Institutes of Health.

ORCID iD

Krista R. Schaefer

References

Adams

J. M.

Wright

J. S.

(2020). A national commitment to improve the care of patients with hypertension in the US. Journal of the American Medical Association, 324(18), 1825-1826. doi:10.1001/jama.2020.20356.

Bello

N. A.

Woolley

J. J.

Cleary

K. L.

Falzon

Alpert

B. S.

Oparil

Cutter

Wapner

Muntner

Tita

A. T.

Shimbo

(2018). Accuracy of blood pressure measurement devices in pregnancy: A systematic review of validation studies. Hypertension, 71(2), 326-335. doi:10.1161/hypertensionaha.117.10295.

Bland

Altman

(1986). Statistical methods for assessing agreement between two methods of clinical measurement. Lancet (London, England), 1(8476), 307-310. doi:10.1016/S0140-6736(86)90837-8.

Bland

J. M.

Altman

D. G.

(1995). Comparing methods of measurement: Why plotting difference against standard method is misleading. Lancet, 346(8982), 1085-1087. doi:10.1016/s0140-6736(95)91748-9.

Blood Pressure-Improving Control Among Alaska Native People (BPICAN) (2019). ClinicalTrials.gov. https://clinicaltrials.gov/ct2/show/NCT03872856.

Cao

Song

Guo

Yang

Deng

Chen

Sapa

W. B.

Wang

(2015). Quality control and validation of oscillometric blood pressure measurements taken during an epidemiological investigation. Medicine, 94(37), e1475. doi:10.1097/md.0000000000001475.

Carey

R. M.

Muntner

Bosworth

H. B.

Whelton

P. K.

(2018). Prevention and control of hypertension: JACC health promotion series. Journal of the American College of Cardiology, 72(11), 1278-1293. doi:10.1016/j.jacc.2018.07.008.

Centers for Disease Control and Prevention (CDC) (2017). About rural health. Rural Health. https://www.cdc.gov/ruralhealth/about.html.

Centers for Disease Control and Prevention (CDC) (2021). Key updates for week 4, ending January 30, 2021. COVIDView: A weekly surveillance summary of U.S. COVID-19 activity. COVID-19. https://www.cdc.gov/coronavirus/2019-ncov/covid-data/covidview/index.html.

10.

Chow

C. K.

Teo

K. K.

Rangarajan

Islam

Gupta

Avezum

Bahonar

Chifamba

Dagenais

Diaz

Kazmi

Lanas

Wei

Lopez-Jaramillo

Fanghong

Ismail

N. H.

Puoane

Rosengren

Szuba

, … Yusuf

(2013). Prevalence, awareness, treatment, and control of hypertension in rural and urban communities in high-, middle-, and low-income countries. Journal of the American Medical Association, 310(9), 959-968. doi:10.1001/jama.2013.184182.

11.

Clark

C. E.

McDonagh

S. T.

McManus

R. J.

Martin

(2020). Covid-19 and hypertension: Risks and management. The BMJ Opinion. https://blogs.bmj.com/bmj/2020/04/15/covid-19-and-hypertension-risks-and-management/.

12.

Curfman

Bauchner

Greenland

(2020). Treatment and Control of Hypertension in 2020. Journal of the American Medical Association, 324(12), 1166-1167. doi:10.1001/jama.2020.13322.

13.

Czeisler

M. É.

Marynak

Clarke

K. E. N.

Salah

Shakya

Thierry

J. M.

Ali

McMillan

Wiley

J. F.

Weaver

M. D.

Czeisler

C. A.

Rajaratnam

S. M. W.

Howard

M. E.

(2020). Delay or avoidance of medical care because of COVID-19-Related concerns - United States, June 2020. MMWR. Morbidity and Mortality Weekly Report, 69(36), 1250-1257. doi:10.15585/mmwr.mm6936a4.

14.

Dewees

Marks

(2017). Twice invisible: Understanding rural Native America. Longmont, CO: First Nations Development Institute. https://www.usetinc.org/wp-content/uploads/bvenuti/WWS/2017/May%202017/May%208/Twice%20Invisible%20-%20Research%20Note.pdf.

15.

Doyle

S. K.

Chang

A. M.

Levy

Rising

K. L.

(2019). Achieving health equity in hypertension management through addressing the social determinants of health. Current Hypertension Reports, 21(8), 58. doi:10.1007/s11906-019-0962-7.

16.

Eby

(2007). Primary care at the Alaska Native Medical Center: A fully deployed "new model" of primary care. International Journal of Circumpolar Health, 66(Suppl 1), 4-13. https://pubmed.ncbi.nlm.nih.gov/18154227/.

17.

European Society of Hypertension-European Society of Cardiology Guidelines Committee (2003). 2003 European Society of Hypertension-European Society of Cardiology guidelines for the management of arterial hypertension. Journal of Hypertension, 21(6), 1011-1053. doi:10.1097/00004872-200306000-00001.

18.

Fryar

Ostchega

Hales

Zhang

Kruszon-Moran

(2017). Hypertension prevalence and control among adults: United States, 2015-2016. NCHS Data Brief, 289, 1-8. https://www.cdc.gov/nchs/data/databriefs/db289.pdf.

19.

Golnick

Asay

Provost

Van Liere

Bosshart

Rounds-Riley

Cueva

Hennessy

T. W.

(2012). Innovative primary care delivery in rural Alaska: A review of patient encounters seen by community health aides. International Journal of Circumpolar Health, 71, 18543. doi:10.3402/ijch.v71i0.18543.

20.

Gottlieb

(2013). The Nuka System of Care: Improving health through ownership and relationships. International Journal of Circumpolar Health, 72(1), 21118. doi:10.3402/ijch.v72i0.21118.

21.

Guan

W.-j.

Liang

W.-h.

Zhao

Liang

H.-r.

Chen

Z.-s.

Y.-m.

Liu

X.-q.

Chen

R.-c.

Tang

C.-l.

Wang

C.-q.

Chen

P.-y.

Sang

Wang

J.-f.

C.-c.

L.-m.

Cheng

, … He

J.-x.

(2020a). China Medical Treatment Expert Group for COVID-19.Comorbidity and its impact on 1590 patients with COVID-19 in China: A nationwide analysis. The European Respiratory Journal, 55(5), 2000547. doi:10.1183/13993003.00547-2020.

22.

Guan

W.-j.

Z.-y.

Liang

W.-h.

C.-q.

J.-x.

Liu

Shan

Lei

C.-l.

Hui

D. S. C.

L.-j.

Zeng

Yuen

K.-Y.

Chen

R.-c.

Tang

C.-l.

Wang

Chen

P.-y.

Xiang

, … Zhong

N.-s.

(2020b) Clinical characteristics of coronavirus disease 2019 in China. The New England Journal of Medicine, 382, 1708–1720. doi:10.1056/NEJMoa2002032.

23.

Havranek

E. P.

Mujahid

M. S.

Barr

D. A.

Blair

I. V.

Cohen

M. S.

Cruz-Flores

Davey-Smith

Dennison-Himmelfarb

C. R.

Lauer

M. S.

Lockwood

D. W.

Rosal

Yancy

C. W.

, & American Heart Association Council on Quality of Care and Outcomes Research, Council on Epidemiology and Prevention, Council on Cardiovascular and Stroke Nursing, Council on Lifestyle and Cardiometabolic Health, and Stroke Council. (2015). Social determinants of risk and outcomes for cardiovascular disease: A scientific statement from the American Heart Association. Circulation, 132(9), 873-898. doi:10.1161/cir.0000000000000228.

24.

Klag

M. J.

Whelton

P. K.

Chen

J.-Y.

J.-P.

Qian

M.-C.

P.-S.

G.-Q.

(1991). Migration, blood pressure pattern, and hypertension: The Yi Migrant Study. American Journal of Epidemiology, 134(10), 1085-1101. doi:10.1093/oxfordjournals.aje.a116012.

25.

Hiratsuka

Beans

Robinson

Shaw

Sylvester

Dillard

(2017). Self-determination in health research: An Alaska Native example of tribal ownership and research regulation. International Journal of Environmental Research and Public Health, 14(11), 1324. doi:10.3390/ijerph14111324.

26.

Howard

B. V.

Comuzzie

Devereux

R. B.

Ebbesson

S. O. E.

Fabsitz

R. R.

Howard

W. J.

Laston

MacCluer

J. W.

Silverman

Umans

J. G.

Wang

Weissman

N. J.

Wenger

C. R.

(2010). Cardiovascular disease prevalence and its relation to risk factors in Alaska Eskimos. Nutrition, Metabolism, and Cardiovascular Diseases: NMCD, 20(5), 350-358. doi:10.1016/j.numecd.2009.04.010.

27.

Howard

Cushman

Moy

C. S.

Oparil

Muntner

Lackland

D. T.

Manly

J. J.

Flaherty

M. L.

Judd

S. E.

Wadley

V. G.

Long

D. L.

Howard

V. J.

(2018). Association of clinical and social factors with excess hypertension risk in Black compared with White US adults. Journal of the American Medical Association, 320(13), 1338-1348. doi:10.1001/jama.2018.13467.

28.

Jernigan

V. B. B.

Duran

Ahn

Winkleby

(2010). Changing patterns in health behaviors and risk factors related to cardiovascular disease among American Indians and Alaska Natives. American Journal of Public Health, 100(4), 677-683. doi:10.2105/AJPH.2009.164285.

29.

Joe

(2003). The rationing of healthcare and health disparity for the American Indians/Alaska Natives. In Smedley

B. D.

Stith

A. Y.

Nelson

A. R.

(Eds.), Unequal treatment: Confronting racial and ethnic disparities in health care. National Academy Press. https://www.ncbi.nlm.nih.gov/books/NBK220367/.

30.

Jolly

S. E.

Howard

B. V.

Umans

J. G.

(2013). Cardiovascular disease among Alaska Native peoples. Current Cardiovascular Risk Reports, 7(6), 1-13. doi:10.1007/s12170-013-0362-5.

31.

Jolly

S. E.

Koller

K. R.

Metzger

J. S.

Day

G. M.

Silverman

Hopkins

S. E.

Austin

M. A.

Boden-Albala

Ebbesson

S. O. E.

Boyer

B. B.

Howard

B. V.

Umans

J. G.

(2015). Prevalence of hypertension and associated risk factors in Western Alaska Native people: The Western Alaska Tribal Collaborative for Health (WATCH) Study. The Journal of Clinical Hypertension, 17(10), 812-818. doi:10.1111/jch.12483.

32.

Jung

M.-H.

Kim

G.-H.

Kim

J.-H.

Moon

K.-W.

Yoo

K.-D.

Rho

T.-H.

Kim

C.-M.

(2015). Reliability of home blood pressure monitoring: In the context of validation and accuracy. Blood Pressure Monitoring, 20(4), 215-220. doi:10.1097/mbp.0000000000000121.

33.

Kapral

M. K.

Austin

P. C.

Jeyakumar

Hall

Chu

Khan

A. M.

Jin

A. Y.

Martin

Manuel

Silver

F. L.

Swartz

R. H.

J. V.

(2019). Rural-urban differences in stroke risk factors, incidence, and mortality in people with and without prior stroke. Circulation. Cardiovascular Quality and Outcomes, 12(2), e004973. doi:10.1161/circoutcomes.118.004973.

34.

Kikuya

Chonan

Imai

Goto

Ishii

, & Research Group to Assess the Validity of Automated Blood Pressure Measurement Devices in Japan (2002). Accuracy and reliability of wrist-cuff devices for self-measurement of blood pressure. Journal of Hypertension, 20(4), 629-638. doi:10.1097/00004872-200204000-00019.

35.

Lee

C. J.

Park

(2016). The role of home blood pressure telemonitoring for blood pressure control. Pulse, 4(2-3), 78-84. doi:10.1159/000448375.

36.

Melville

Teskey

Philip

Simpson

J. A.

Lutchmedial

Brunt

K. R.

(2018). A comparison and calibration of a wrist-worn blood pressure monitor for patient management: Assessing the reliability of innovative blood pressure devices. Journal of Medical Internet Research, 20(4), e111. doi:10.2196/jmir.8009.

37.

Milani

R. V.

Lavie

C. J.

Bober

R. M.

Milani

A. R.

Ventura

H. O.

(2016). Abstract 12057: Improving hypertension control and patient engagement using digital tools. Circulation, 134(Suppl 1), A12057. doi:10.1161/circ.134.suppl_1.12057.

38.

MONICA Manual (1997). Part III: Population survey, section 1: Population survey data component. WHO MONICA Project. https://www.thl.fi/publications/monica/manual/part3/iii-1.htm.

39.

Muller

C. J.

Alonso

Forster

Vock

D. M.

Zhang

Gottesman

R. F.

Rosamond

Longstreth

W. T.

Jr. MacLehose

R. F.

(2019). Stroke incidence and survival in American Indians, Blacks, and Whites: The Strong Heart Study and Atherosclerosis Risk in Communities Study. Journal of the American Heart Association, 8(12), e010229. doi:10.1161/jaha.118.010229.

40.

Muntner

Hardy

S. T.

Fine

L. J.

Jaeger

B. C.

Wozniak

Levitan

E. B.

Colantonio

L. D.

(2020). Trends in blood pressure control among US adults with hypertension, 1999-2000 to 2017-2018. The Journal of the American Medical Association, 324(12), 1190-1200. doi:10.1001/jama.2020.14545.

41.

Murphy

N. J.

Schraer

C. D.

Theile

M. C.

Boyko

E. J.

Bulkow

L. R.

Doty

B. J.

Lanier

A. P.

(1997). Hypertension in Alaska Natives: Association with overweight, glucose intolerance, diet and mechanized activity. Ethnicity & Health, 2(4), 267-275. doi:10.1080/13557858.1997.9961835.

42.

National Center for Health Statistics (2017). Health, United States, 2016: With chartbook on long-term trends in health. Hyattsville, MD: National Center for Health Statistics. https://www.ncbi.nlm.nih.gov/books/NBK453378/.

43.

O'Brien

Asmar

Beilin

Imai

Mallion

J.-M.

Mancia

Mengden

Myers

Padfield

Palatini

Parati

Pickering

Redon

Staessen

Stergiou

Verdecchia

(2003). European Society of Hypertension recommendations for conventional, ambulatory and home blood pressure measurement. Journal of Hypertension, 21(5), 821-848. doi:10.1097/00004872-200305000-00001.

44.

O'Brien

Parati

Stergiou

Asmar

Beilin

Bilo

Clement

de la Sierra

de Leeuw

Dolan

Fagard

Graves

Head

Imai

Kario

Lurbe

Mallion

Mancia

Mengden

, … Zhang

(2013). European Society of Hypertension position paper on ambulatory blood pressure monitoring. Journal of Hypertension, 31(9), 1731-1768. doi:10.1097/HJH.0b013e328363e964.

45.

Ogden

Carroll

Fryar

Flegal

(2015). Prevalence of obesity among adults and youth: United States, 2011-2014. NCHS Data Brief, 219, 1-8. https://www.cdc.gov/nchs/data/databriefs/db219.pdf.

46.

Omboni

McManus

R. J.

Bosworth

H. B.

Chappell

L. C.

Green

B. B.

Kario

Logan

A. G.

Magid

D. J.

Mckinstry

Margolis

K. L.

Parati

Wakefield

B. J.

(2020). Evidence and recommendations on the use of telemedicine for the management of arterial hypertension. Hypertension, 76(5), 1368-1383. doi:10.1161/HYPERTENSIONAHA.120.15873.

47.

Peng

Tibshirani

Friedman

(2009). The elements of statistical learning: Data mining, inference, and prediction (2nd ed.). Springer Science+Business Media, LLC.

48.

Pickering

T. G.

Hall

J. E.

Appel

L. J.

Falkner

B. E.

Graves

Hill

M. N.

Jones

D. W.

Kurtz

Sheps

S. G.

Roccella

E. J.

(2005). Recommendations for blood pressure measurement in humans and experimental animals: Part 1: Blood pressure measurement in humans: A statement for professionals from the subcommittee of professional and public education of the American Heart Association Council on High Blood Pressure Research. Circulation, 111(5), 697-716. doi:10.1161/01.Cir.0000154900.76284.F6.

49.

Reboussin

D. M.

Allen

N. B.

Griswold

M. E.

Guallar

Hong

Lackland

D. T.

Miller

E. R.

3rd Polonsky

Thompson-Paul

A. M.

Vupputuri

(2018). Systematic Review for the 2017 ACC/AHA/AAPA/ABC/ACPM/AGS/APhA/ASH/ASPC/NMA/PCNA Guideline for the prevention, detection, evaluation, and management of high blood pressure in adults: A report of the American College of Cardiology/American Heart Association Task Force on Clinical Practice Guidelines. Hypertension, 71(6), e116-e135. doi:10.1161/hyp.0000000000000067.

50.

Redwood

Lanier

Johnston

Asay

Slattery

(2010). Chronic disease risk factors among Alaska Native and American Indian people, Alaska, 2004-2006. Preventing Chronic Disease, 7(4), A85. http://www.cdc.gov/pcd/issues/2010/jul/09_0168.htm.

51.

Rhoades

D. A.

Welty

T. K.

Wang

Yeh

Devereux

R. B.

Fabsitz

R. R.

Lee

E. T.

Howard

B. V.

(2007). Aging and the prevalence of cardiovascular disease risk factors in older American Indians: The Strong Heart Study. Journal of the American Geriatrics Society, 55(1), 87-94. doi:10.1111/j.1532-5415.2006.01018.x.

52.

Stergiou

G. S.

Alpert

Mieke

Asmar

Atkins

Eckert

Frick

Friedman

Graßl

Ichikawa

Ioannidis

J. P.

Lacy

McManus

Murray

Myers

Palatini

Parati

Quinn

Sarkis

, … O’Brien

(2018). A universal standard for the validation of blood pressure measuring devices: Association for the Advancement of Medical Instrumentation/European Society of Hypertension/International Organization for Standardization (AAMI/ESH/ISO) Collaboration Statement. Journal of Hypertension, 36(3), 472-478. doi:10.1097/hjh.0000000000001634.

53.

Touyz

P. K.

Carey

R. M.

Aronow

W. S.

Casey

D. E.

Jr. Collins

K. J.

Dennison Himmelfarb

DePalma

S. M.

Gidding

Jamerson

K. A.

Jones

D. W.

MacLaughlin

E. J.

Muntner

Ovbiagele

Smith

S. C.

Jr. Spencer

C. C.

Stafford

R. S.

Taler

S. J.

Thomas

R. J.

Williams

K. A.

Sr. , … Wright

J. T.

Jr. (2018). 2017 ACC/AHA/AAPA/ABC/ACPM/AGS/APhA/ASH/ASPC/NMA/PCNA Guideline for the prevention, detection, evaluation, and management of high blood pressure in adults: Executive summary: A report of the American College of Cardiology/American Heart Association Task Force on Clinical Practice Guidelines. Hypertension, 71(6), 1269-1324. doi:10.1161/hyp.0000000000000066.

(2017). Self-monitoring of blood pressure in hypertension: A systematic review and individual patient data meta-analysis. PLoS Medicine, 14(9), e1002389. doi:10.1371/journal.pmed.1002389.

(2013). Self-Measured blood pressure monitoring in the management of hypertension: A systematic review and meta-analysis. Annals of Internal Medicine, 159(3), 185-194. doi:10.7326/0003-4819-159-3-201308060-00008.

56.

US Census Bureau (n.d.). Rural America. https://mtgis-portal.geo.census.gov/arcgis/apps/MapSeries/index.html?appid=49cd4bc9c8eb444ab51218c1d5001ef6.

57.

Wang

Zhu

Liu

Zhang

Wang

Xiang

Cheng

Xiong

Zhao

Wang

Peng

(2020). Clinical characteristics of 138 hospitalized patients with 2019 novel coronavirus-infected pneumonia in Wuhan, China. Journal of the American Medical Association, 323(11), 1061-1069. doi:10.1001/jama.2020.1585.

58.

Weber

M. A.

Schiffrin

E. L.

White

W. B.

Mann

Lindholm

L. H.

Kenerson

J. G.

Flack

J. M.

Carter

B. L.

Materson

B. J.

Ram

C. V. S.

Cohen

D. L.

Cadet

J.-C.

Jean-Charles

R. R.

Taler

Kountz

Townsend

R. R.

Chalmers

Ramirez

A. J.

Bakris

G. L.

, … Harrap

S. B.

(2014). Clinical practice guidelines for the management of hypertension in the community: A statement by the American Society of Hypertension and the International Society of Hypertension. The Journal of Clinical Hypertension, 16(1), 14-26. doi:10.1111/jch.12237.

59.

Wiemers

E. E.

Abrahams

AlFakhri

Hotz

V. J.

Schoeni

R. F.

Seltzer

J. A.

(2020). Disparities in vulnerability to complications from COVID-19 arising from disparities in preexisting conditions in the United States. Research in Social Stratification and Mobility, 69, 100553. doi:10.1016/j.rssm.2020.100553.

60.

Williams

J. S.

Brown

S. M.

Conlin

P. R.

(2009). Videos in clinical medicine. Blood-Pressure measurement. The New England Journal of Medicine, 360(5), e6. doi:10.1056/NEJMvcm0800157.

61.

Wisniewski

J. M.

Walker

(2020). Association of simulated patient race/ethnicity with scheduling of primary care appointments. JAMA Network Open, 3(1), e1920010. doi:10.1001/jamanetworkopen.2019.20010.

62.

Woo

K. K.

Can

Chang

D. W.

(2020). Racial differences in the utilization of guideline-recommended and life-sustaining procedures during hospitalizations for out-of-hospital cardiac arrest. Journal of Racial and Ethnic Health Disparities, 7(3), 403-412. doi:10.1007/s40615-019-00668-8.

63.

G. S.

Christodoulakis

G. R.

Nasothimiou

E. G.

Giovas

P. P.

Kalogeropoulos

P. G.

(2008). Can validated wrist devices with position sensors replace arm devices for self-home blood pressure monitoring? A randomized crossover trial using ambulatory monitoring as reference. American Journal of Hypertension, 21(7), 753-758. doi:10.1038/ajh.2008.176.

64.

Zamora-Kapoor

Sinclair

Nelson

Lee

Buchwald

(2019). Obesity risk factors in American Indians and Alaska Natives: A systematic review. Public Health, 174, 85-96. doi:10.1016/j.puhe.2019.05.021.

65.

Zhang

Moran

A. E.

(2017). Trends in the prevalence, awareness, treatment, and control of hypertension among young adults in the United States, 1999 to 2014. Hypertension, 70(4), 736-742. doi:10.1161/hypertensionaha.117.09801.

66.

Zhang

Petrie

Littler

de Swiet

Padfield

Altman

Bland

Coats

Atkins

(1993). An outline of the revised British Hypertension Society protocol for the evaluation of blood pressure measuring devices. Journal of Hypertension, 11(6), 677-679. doi:10.1097/00004872-199306000-00013.