How accurate must blood pressure measurements be to prevent diagnostic errors? According to published research, blood pressure measurement errors of 5 mmHg and above can affect clinical decision-making. There is a vulnerable range between 130 – 150 mmHg systolic and 75 – 85 mmHg diastolic, where small errors can result in patients being incorrectly evaluated and diagnosed.
This paper reports the results of our investigation and evaluation of blood pressure measurement accuracy, research, and studies published in peer-reviewed journals. Human factors and decision-making principles were applied to assess existing (legacy) technologies and practices, and their implications for published hypertension prevalence and related factors. We also considered clinician overconfidence [1],[2] in their own and their staff's ability to accurately measure blood pressure despite acknowledged problems with legacy instruments and methods.
Investigation
Measurement accuracy is central to the credibility of scientific research, analysis, practices, and decision-making. Accurate measurements of blood pressure, therefore, are indispensable to blood pressure assessments, hypertension diagnosis, treatment strategy development, efficacy evaluations, and long-term management. Published research is clear that even small errors in blood pressure measurements (5 – 10 mmHg) have disproportionately large impacts on clinical decision-making:
“population-wide, small inaccuracies in blood pressure measurement can have considerable consequences. Underestimating true blood pressure by 5 mm Hg would mislabel more than 20 million Americans with prehypertension when true hypertension is present. It has been predicted that the consequences of an untreated 5 mm Hg of excessive systolic blood pressure would be a 25% increase over current levels of fatal strokes and fatal myocardial infarctions for these individuals. Conversely, overestimating true blood pressure by 5 mm Hg would lead to inappropriate treatment with anti-hypertension medications in almost 30 million Americans, with attendant exposure to adverse drug effects, the psychological effects of misdiagnosis, and unnecessary cost.[3]”
Previous efforts to improve blood pressure measurement accuracy have proven ineffective, which raises questions about population estimates of hypertension prevalence, treatment efficacy, and disease management outcomes. A recent post on the American Medical Association website acknowledged that:
“BP measurement is often suboptimally performed in clinical practice, which can lead to errors that inappropriately alter management decisions in 20% to 45% of cases. This inaccuracy has persisted despite extensive education and efforts to raise awareness on the adverse consequences of incorrect clinic BP measurement, according to the Lancet Commission on Hypertension Group position statement.[4]”
Accurately measuring representative blood pressure involves more than instrument accuracy and human performance. Blood pressure is affected by many variables, which are difficult to identify and compensate for. These include mundane factors such as patient movement and interactions between patients and clinicians, which can promote white coat syndrome, a condition where patient blood pressure becomes artificially high in clinical settings. Researchers have suggested limiting clinician-patient interactions during blood pressure measurements as a compensatory strategy:
“…white coat effect (the tendency of blood pressure to rise during a medical visit) and talking effect were analyzed in 42 patients with essential hypertension… Measures of systolic/diastolic blood pressure were higher during stressful talking than during relaxed talking. The talking and its emotional contents seemed to explain 70% of the white coat phenomenon. To minimize the white coat phenomenon in the clinic, physicians, nurses, and clinicians are advised to measure blood pressure during an initial period of silence.[5]”
Patient self-measurements are significantly affected by equipment performance, poor training, and limited experience. Poor device accuracy has been reported in nearly 20% of home blood pressure instruments, particularly among less expensive models that are not FDA-listed. Researchers also report widespread errors committed by patients in measuring their blood pressure:
“Only 3% of patients measured their BP without error; 60% made three or more errors. The most frequent error, made by 76% of subjects, was incorrect sphygmomanometer cuff placement (above or below heart level, or/and the indicator mark was not aligned with the brachial artery). Regarding patients’ previous instruction for the correct use of their devices, 36% of patients referred to their monitor’s user manual, 22% did not receive any prior assistance, and only 29% were adequately counseled by physicians on how to measure their BP correctly.[6]”
A systematic review of peer-reviewed studies
Research on popular subjects like hypertension is broad and often noisy due to conflicting findings, differing methods, and differences in interpretation. A common strategy for reducing noise and improving quality is to perform comprehensive reviews of published findings. A recently published systematic study of hundreds of published papers identified twenty-five sources of inaccuracies, which the authors divided into seven categories and stages (before measurement, selecting the device, positioning the patient, attaching the device to the patient, taking the measurement, and interpreting the measurement) [7]. The authors concluded that:
“…though many of the identified sources of inaccuracy may not individually affect a reading of BP enough to alter a clinical decision, the cumulative effect of multiple patient, procedure, device and/or observer factors could have a substantial impact – as could some of the more influential sources of inaccuracy in isolation. To make sound interpretations of BP measurements, clinicians should recognize the extent of variability that can exist between any two measurements… [their findings should] act as a reminder that any single BP measurement includes the potential for substantial inaccuracy and should not be treated as a meaningful clinical indicator on its own.[8]”
Limited baselines
The systematic study also highlighted the lack of studies on normal and hypotensive patients. Measurements from patients with controlled blood pressure can help improve baselines and benchmarks. Hypotension is a concern with patients who suffer recurring dizzy spells, a frequent condition among older patients. As a result, the researchers did not illuminate the influence of inaccurate blood pressure measurements on hypotension diagnoses:
"hypotensive patients were seldom studied in the reviewed literature, with most participants being normotensive or hypertensive. The extent to which the identified sources of inaccuracy generalize to hypotensive deteriorating patients is not apparent, and replications utilizing varying populations would help to clarify this [9].”
Human factors considerations [10]
Human Factors Engineering (HFE) is the discipline that focuses on the interactions between systems and human operators, including the human-machine interface. HF engineers analyze how well instruments, equipment, and systems account for human strengths and limitations. Designs that do not incorporate human factors are prone to poor performance in the real world of human users and operators [11].
One hundred years of sphygmomanometer improvements
have not translated into more accurate measurements.
Sensors like Biobeat's cuffless monitors are introducing
innovative alternatives to legacy cuff-based technologies and methods.
In the context of blood pressure measurements, the literature points to poor human factors in instrument design, operations, and practices as contributors to measurement inaccuracy. The design of the venerable, century-old sphygmomanometer and its newer digital versions are major contributors to poor accuracy in clinical and home use settings. Known issues include proper cuff size, cuff position, patient body and arm position, and scale readability. Social components, including patient-clinician interactions during measurement and patient reactions to clinical environments, are also factors. The systematic review of the literature discussed above identified over two dozen factors that affect blood pressure measurement accuracy, many of which are traceable to recognized human factors deficiencies.
The literature points to training as the primary remedy to offset factors that undermine measurement quality. While improving training is usually beneficial, it is impractical as a solution for busy nurses and nurses’ assistants, much less patients, to effectively address poor measurement quality. In practice, while training has been used as a stop-gap measure, it is not a practical strategy for overcoming human factor weaknesses. We concluded that legacy equipment designs and practices virtually guarantee inaccurate blood pressure readings.
Implications
The acknowledged inaccuracy of blood pressure measurements raises profound questions about what is verifiably known about hypertension, including its prevalence, diagnostic reliability, treatment strategies, and efficacy assessments. The first cautionary principle of quantitative analysis and informed decision-making is “garbage in – garbage out.” Measurements lacking sufficient, repeatable accuracy are unacceptable for use in applications with human safety implications.
Science and engineering are sometimes hampered by technological and instrument limitations. In those instances, good practices require acknowledging their limits and implications for known use. They also require compensatory steps to improve overall performance and reduce their potential impacts, particularly when human safety and health are involved. Experience and research suggest that in the case of blood pressure, compensatory strategies have not produced the desired improvements in measurement accuracy, interpretation, and clinical decision-making.
Conclusion
Research and recommendations across the literature highlight the limitations of single blood pressure measurements as inputs into clinical decision-making. Averaging multiple readings is a recommended improvement. Using ambulatory blood pressure monitoring (ABPM) studies that measure blood pressure four times per hour over 24 hours is a better strategy for mitigating measurement errors. Unfortunately, factors including costs, limited availability of equipment and trained personnel, and patients' reactions to cuff discomfort and pain limit its use [12].
Clinicians continue to rely on legacy blood pressure measuring equipment and practices to inform their decision-making. Averaging multiple readings is a common but not required compensatory strategy. ABPM studies are less common despite published recommendations by researchers and organizations, including the American Heart Association. As a result, millions of hypertensive patients are not accurately diagnosed and treated by their care providers. These findings raise profound concerns about the basis of decision-making in hypertension diagnosis, treatment development, treatment efficacy assessment, and disease management.
References
[1] Lowenstein EJ. Patient safety and the mother of all biases: Overconfidence. Int J Womens Dermatol. 2019 Nov 7;6(2):127-128. doi: 10.1016/j.ijwd.2019.09.005. PMID: 32258349; PMCID: PMC7105692. [2] Cassam, Q. Diagnostic error, overconfidence and self-knowledge. Palgrave Commun 3, 17025 (2017). https://doi.org/10.1057/palcomms.2017.25 [3] Pathan MK, Cohen DL. Resistant Hypertension: Where are We Now and Where Do We Go from Here? Integr Blood Press Control. 2020 Aug 5;13:83-93. doi: 10.2147/IBPC.S223334. PMID: 32801854; PMCID: PMC7415451. [4] Sara Berg, 4 big ways bp measurement goes wrong, and how to tackle them, January 3, 2023, https://www.ama-assn.org/delivering-care/hypertension/4-big-ways-bp-measurement-goes-wrong-and-how-tackle-them [5] Le Pailleur C, Vacheron A, Landais P, Mounier-Véhier C, Feder JM, Montgermont P, Jais JP, Metzger JP. Talking effect and white coat phenomenon in hypertensive patients. Behav Med. 1996 Fall;22(3):114-22. doi: 10.1080/08964289.1996.9933772. PMID: 9116382. [6] Nessler, K., Krztoń-Królewiecka, A., Suska, A. et al. The quality of patients’ self-blood pressure measurements: a cross-sectional study. BMC Cardiovasc Disord 21, 539 (2021). https://doi.org/10.1186/s12872-021-02351-5 [7] Kallioinen N, Hill A, Horswill MS, Ward HE, Watson MO. Sources of inaccuracy in the measurement of adult patients' resting blood pressure in clinical settings: a systematic review. J Hypertens. 2017 Mar;35(3):421-441. doi: 10.1097/HJH.0000000000001197. PMID: 27977471; PMCID: PMC5278896. [8] Ibid, Kallioinen. [9] Ibid, Kallioinen. [10] Human factors evaluation performed in collaboration with AC “Dean” Macris, A. C. Macris Consultants, Mystic, CT 06355 [11] Ozzie Paez, The increasingly confusing language of automation, May 17, 2018, Ozzie Paez Research, https://www.ozziepaezresearch.com/post/2018/05/17/the-growing-language-of-automation
[12] Ozzie Paez, Hypertension and Patient Compliance: Part I, OPRHealth, June 20, 2023, https://www.oprhealth.com/post/hypertension-and-patient-compliance-part-i
Comments