Accurate measurement of blood pressure (BP) requires a BP cuff with the appropriate bladder width. Recently, based on previous findings, it is suggested that the ratio of bladder width to mid-AC should be 40–47% of the arm circumference [1,2]. Using a cuff with a bladder width that is too narrow for the mid-arm circumference (mid-AC) tends to overestimate BP, which potentially results in ‘cuff hypertension’. In contrast, a bladder width that is too wide for the mid-AC results in underestimates or incorrectly low BP readings [3–8].

This analysis is an update of previous analysis on the same topic [9,10]. Accordingly, the objectives of this analysis were two-fold. The first objective was to provide averages and selected percentiles of mid-AC (cm) for the noninstitutionalized US population aged 3 years and older by sex, age, race and Hispanic origin, BMI, high BP status (aged 8–17 years), and hypertension status (aged > 18 years). The second objective was to examine frequency distribution by selected age groups (3–5, 6–11, 12–19, and 20 years and older) of mid-AC to be cuffed by the Baum; W.A. Baum Co., Inc., Copiague, New York, USA (infant, child/small adult, adult, large adult, and extra-large adult) and by the Welch Allyn (small child, child, small adult, adult, large adult, and extra-large adult).

Table 2 presents the mean values and selected percentiles by survey periods, sex, age groups, race and Hispanic origin, and BMI. For survey years 2011–2016, the overall mean mid-AC was not statistically significantly different across the three 2-year survey periods, from 31.0 cm during 2011–2012 to 31.3 cm during 2015–2016. From 2011–2016, mean mid-AC was greater for males than for females (32.0 vs. 30.4 cm, P < 0.001). Mean mid-AC varied by age groups and was smallest among children 3–5 years (17.5 cm) and largest among adults 40–49 year (34.0 cm). Among self-reported race and Hispanic groups, non-Hispanic Black persons had the largest mean mid-AC (32.0 cm) and non-Hispanic Asian persons had the smallest mean mid-AC (28.4 cm). Among children and adolescents aged 3–19 years, those classified as having a normal weight or being underweight had the smallest mean mid-AC (22.0 cm) and those with obesity had the largest mean mid-AC (31.5 cm). Similarly, for adults aged 20 years and older, normal weight was associated with the smallest mean mid-AC (28.2 cm) and obesity with the largest mean mid-AC (37.8 cm). Among adults aged 18 years and older, adults with normal BP (< 130/80 mmHg and not medicated for hypertension) had a smaller mean mid-AC (32.4 cm) than adults with hypertension (≥130/80 mmHg or medicated for hypertension) (mean mid-AC = 44.2 cm). Similar pattern was observed for children and adolescents aged 8–17 years old, where children with normal BP had a smaller mean mid-AC (26.1 cm) than children with high BP (mean mid-AC = 28.2 cm).

Figure 1 describes the CFD plot for mid-AC for specific age groups (3–5, 6–11, 12–19, and 20 years and more). The age specific CFD are further described in Table 3 by Baum and Welch Allyn cuff systems. Among children aged 3–5 years old, an estimated 91.7% had a mid-AC ranging from 10 to less than 20 cm and required a cuff that could accommodate this range, such as the Baum infant cuff (66.6%) and Welch Allyn child cuff (88.2%). Among children aged 6–11 years old, greater than 20% needed an adult or larger cuff sizes (Baum or Welch Allyn) accommodating a mid-AC at least 25 cm. Among adolescents ages 12–19 years old, an estimated 13.0% of this age group required a 35 to less than 47 cm mid-AC range cuff, like the Baum large adult cuff; an estimated 21.7% required a 32 to less than 40 cm mid-AC range cuff, like the Welch Allyn large adult cuff; and an additional 3.7% required a 40–55 cm mid-AC range cuff, like the Welch Allyn extra-large adult cuff. For adults aged 20 years and older, greater than 33% of this age group required at least 35 cm mid-AC cuff like the Baum large adult cuff and extra-large adult cuff, and more than an estimated 50% required at least 32 cm mid-AC range cuff like the Welch Allyn large adult cuff and extra-large adult cuff.

From 2011–2012 to 2015–2016, there was no significant change in the mean mid-AC. During 2011–2016, there were significant differences in mean mid-AC by sex, age groups, race and Hispanic origin, hypertension status, and BMI status. Increased BMI represents a challenge for appropriate cuff-bladder sizes, and as previously shown that even after adjusting for all covariates, BMI was significantly associated with BP cuff sizes [10].

During 2011–2016, more than 13.0% of noninstitutionalized adolescents aged 12–19 years and older, either being cuffed by the Baum or Welch Allyn cuff system, required a large adult BP or extra-large adult cuff to be correctly cuffed. Moreover, more than 33.0% of noninstitutionalized adults aged 20 years and older, either being cuffed by the Baum or Welch Allyn cuff system, required a large adult BP or extra-large adult cuff to be correctly cuffed.

Although this analysis (2011–2016) is an update of previous analyses (1999–2010), we felt that there is a value in the update [9,10]. First, and most importantly, currently there is a wide use of single cuff in home BP devices and kiosk BP devices, also known as wide-range cuffs, which are mostly used with automated systems; therefore, additional care must be taken in validating the results obtained, but wide range cuffs should not be used for manually taken BP. One example of validating a wide-range cuff in a kiosk setting is the validation of Pharma-Smart PS-2000 cuff (Pharmasmart International Inc., Rochester, New York, USA) using AAMI/ISO validation criteria [23]. Second, this paper examines the association of the new high BP and hypertension guidelines for children and adolescents aged 8–17 years and adults aged 18 years and older and mid-AC values [18,21]. Third, the current analysis includes BMI as a covariate. Very recent analysis of obesity and severe obesity comparing NHANES survey years 2007–2008 to 2015–2016 shows that obesity and severe obesity rates have not changed for children and adolescents aged 2–19 years. However, they increased significantly in adults; thus, BMI was added to account for that increase [24]. Finally, although in the previous analyses we provided means and standard error by demographic covariates, in the current analyses (in addition to mean mid-AC and standard error), we are providing percentages (5th to 95th) to help with further data interpretation. Finally, this analysis associates mid-AC values with two known BP cuff systems.

There are a number of clinical implications to our report. Overall, in 2011–2014, ~ 36 million (17%) persons engaged in monthly or more frequent home BP monitoring [25,] and as Graves reported, of 124 websites offering home BP devices, only 53% offered more than one cuff size to measure BP [26]. Moreover, Omron (OMRON Corporation, Kyoto, Japan) upper arm home BP devices systems, 5–10 series, offer a one-cuff system with cuffing range 22–42 cm [27]. Using our current mid-AC data to assess the accuracy of BP measurements of the Omron systems suggests that as prescribed, by the manufacturer, mid-AC range, it would miss-cuff 5% of the adult population ages 20 years and older (Fig. 1) and more than 5% of the obese adult and hypertensive population (Table 2). In addition, our data suggested that children and adolescents aged 8–17 years observed to have high BP during the MEC examination on average need an adult BP cuff for accurate BP determination and large adult cuff for those with high BP and in the 95% range.

Ubiquitous in any local pharmacy is the BP kiosk with one BP cuff ‘one size fits all’. Indeed, it is speculated that ~30 000 BP kiosks are located in pharmacies and work places and the concern is what effect does the one cuff has on the accuracy of obtained BP [23]. For example, Pharma-Smart PS-2000 (one cuff system; mid-AC range 22–38 cm) and Vita-Stat 90550; Medical Screening Services, Inc., Niles, Illinois, USA (one cuff system; mid-AC range 23–33 cm), both are public-use BP monitors [23,28]. Using our current mid-AC data to assess the accuracy of BP measurements of those devices suggests that as prescribed, by the manufacturer, mid-AC ranges, the former would miss-cuff at least 15% of the adult population ages 20 years and older (Fig. 1) and more than 25% of the obese adult population (Table 2), whereas the latter would miss-cuff more than 50% of the adult population ages 20 years and older (Fig. 1) and more than 75% of the obese adult population (Table 2).

There are a number of strengths underlining this report. First, mid-AC data were collected by highly trained health technicians, observed frequently for quality control, and using the same strict protocol to measure mid-AC since NHANES III [13]. Second, previously we chose the National High Blood Pressure Education Program Working Group on High Blood Pressure in Children (4th report) and the American Heart Association (7th report) recommended BP cuff sizes as the basis of our analysis; however, those recommended BP cuff sizes are considered ‘theoretical’. Therefore, we felt that by selecting the Baum and Welch Allyn cuff system as demonstration cuff systems we provided more practical reference cuff sizes [29,30].

The limitations to our current report are the fact that we did not consider other manufacturers cuff systems, such as the Omron HEM-907XL for cuff sizes or other manufacturers, such as GE (GE Healthcare, Chicago, Illinois, USA) or Philips (Philips Healthcare, Andover, Massachusetts, USA). Moreover, we measured the mid-AC on the right arm. There is a chance that the left mid-AC may be larger or smaller than the right mid-AC.