a b s t r a c t

Objective: Although noise is known to negatively affect Blood pressure measurements, its impact on different BP measurement methods remains unclear. The aim of this study is to compare the agreement of oscillometric and auscultatory BP measurement methods under in-ambulance noise levels.

Methods: This method-comparison study was conducted on 50 healthy volunteers in a tertiary emergency department (ED). Participants were divided into two groups of 25, and BP was measured using auscultatory and oscillometric methods in noisy and ambient environments by 2 emergency medicine technicians (EMT). The primary object of the study was to compare the agreement of auscultatory mercury sphygmomanometers and automated auscillometric BP measurements in ambient and noisy environments.

Results: We examined the agreement between auscultative and oscillometric measurements of BP conducted in an ambient environment (46.75 [IQR (41.2-55.18)] dB) and found that both systolic and Diastolic BP were within the level of agreement (LoA) established before the study (systolic BP [-13.96 to 8.48 mmHG], diastolic BP [-7.44 to 8.08 mmHg]); whereas, in noisy environment (92.35 [IQR 88-96.55] dB) both systolic and diastolic BP were outside the range of LoA (systolic BP [-37.77 to 9.94 mmHg], diastolic BP [-21.73 to 16.37 mmHg]). Additionally, we found that in ambient environments, concordance Correlation coefficients were higher than in noisy environments (0.943 [0.906-0.966], 0.957 [0.93-0.974]; 0.574 [0.419-0.697], 0.544 [0.326-0.707]; systolic

and diastolic BP, respectively).

Conclusion: The results of this study demonstrate that noise significantly affects the agreement between oscillometric and auscultatory blood pressure measurement methods.

(C) 2023

1. Introduction

In emergency medicine practice, blood pressure (BP) is a cardinal vital sign that reflects the physiologic Hemodynamic response caused by the medical condition of the individual [1]. For an accurate assess- ment of this sign, the measurement must also be precise [2]. Several criteria are included in the recommended protocol of BP measurement, but their application in ED and prehospital units is uncertain [3]. When factors such as the patient and the environment in which the measure- ment is performed deviate from the ideal, the results and quality of these measurements may be adversely affected.

An accurate assessment of BP requires the use of an appropriate measurement technique. The first of two commonly used non- invasive BP measurement methods is the auscultatory method, which

* Corresponding author at: Department of Emergency Medicine, University of Health Sciences, Kartal Dr. Lutfi Kirdar City Hospital, Kartal, Istanbul, Turkey.

E-mail address: [email protected] (S. Yilmaz).

relies on hearing the “tapping” sounds known as Korotkoff sounds com- posed of five phases as heard with a stethoscope; the oscillometric method involves the measurement of small pressure oscillations in the cuff surrounding the upper arm as a result of the heart pumping blood [4,5]. Moreover, an oscillometric test is performed by a machine while an auscultatory test is conducted by a human and is, like all human measurement tasks, highly dependent on the user. When the arm moves under the cuff or there is an irregular pulse, the oscillometric method may yield misleading results [6]. To date, there have been many studies in the literature examining the success and compatibility of BP measurement methods, but in these studies, ideal BP measurement con- ditions were first and foremost realized by ensuring silence [4,7-10]. In studies based on hearing voices, it is essential that silence be main- tained. However, studies prepared in such a manner have limited relevance to emergency medicine because both prehospital and EDs are noisy environments [11]. Noise is known to negatively affect the success of BP measurement, however, there are not enough data available on different methods of measuring BP under noise [12].

https://doi.org/10.1016/j.ajem.2023.02.022

0735-6757/(C) 2023

The aim of this study is to compare the agreement of oscillometric and auscultatory BP measurement methods under in-ambulance noise levels.

1. Methods
   1. Study design and setting

This method-comparison study was conducted on 50 volunteers in a Tertiary ED between November 1, 2022, and November 15, 2022. Ethics committee approval was obtained from Medipol University Non- Interventional Clinical Research Ethics Committee on 13.10.2022 (E-10840098-772.02-6162). Written informed consent was obtained from all participants before the study. All studies were conducted in accordance with the Declaration of Helsinki.

• 1. Selection of participants

Healthy volunteers aged 18 and older of both sexes participated in this study. The BP of each participant was measured between 9 and 10 am on different days of the week. Before the BP measurement, pa- tients rested on a chair in an isolated room in the ED for 10 min. We di- vided the volunteers into two groups and measured 25 volunteers in a noisy environment and the other 25 in an ambient environment. We chose the environment in which the volunteers would be measured at random. Randomization was conducted online using a program. (https://www.graphpad.com/quickcalcs/).

• 1. Interventions and outcomes

In this study, the primary objective was to compare auscultatory mercury sphygmomanometers and automated auscillometric BP mea- surements in ambient and noisy environments. We simulated a noisy environment in the ambulance by running an ambulance siren sound recording through a wireless speaker, while operating two hair dryers at the same time in the room. As the noise level inside ambulances has been determined to be over 80 dB in previous studies, we tried to generate noise at this level. We measured ambient noise only in an iso- lated single room in ED with the door closed. A digital decibel meter (Decibel X app) was used to measure and record noise levels in both conditions. For the comparison of agreement between automatic and manual auscultatory BP measurements, the Passing-Bablok, Bland- Altman, and Lin’s concordance correlation methods were chosen. We planned 50 measurements for each of the two environments in order to obtain accurate results in this method [13,14]. EMTs with more than five years of experience who recently underwent a hearing test and did not have hearing loss performed manual auscultatory blood pressure measurement in both arms. Each measurement was recorded on the form as separate measurements. The measurements from two different arms were recorded as two separate records. Next, oscillometric measurements were conducted and noted.

Each volunteer’s BP was measured by 1 EMT, and the EMTs per- formed an equal number of manual auscultatory measurements. A com- parison of the agreement between the BP measurements of the two EMTs before the study began was conducted by performing 20 consec- utive measurements on 10 volunteers who had not been included in the study with a manual auscultatory method. (The agreement analysis re- sults are included in the results section). It was not disclosed to the EMTs who performed the measurements what the primary outcome of the study was.

• 1. Measurements

During the study, manual measurements were conducted with an Erka wall type mercury sphygmomanometer (Istanbul, Turkey) and au- tomatic measurements were conducted with a Philips SureSigns VS2+

monitor (Andover, MA, USA). The devices were calibrated prior to the study. A cuff of the size recommended by the manufacturer suitable for the arm circumference was used for both manual and automatic BP measurements. In order to ensure that no piece of clothing was under the cuff or that the clothing did not wrap around the cuff, the vol- unteers wore only short-sleeved clothing during the measurement. Age, gender, systolic BP and diastolic BP results with manual auscultatory and automatic BP monitors were recorded.

• 1. Statistical analysis

Analysis of the statistical data was carried out with IBM SPSS Statis- tics (version 27) and MedCalc Statistical Software (version 20.104). Numerical variables were expressed as mean +- standard deviation or median (IQR 25th - 75th) and categorical variables as number (%). Kolmogorov-Smirnov and histograms were used to verify that the data conformed to a normal distribution. For comparisons between in- dependent groups, the student’s t-test or Mann-Whitney U test was used and for categorical data, the Pearson Chi-square test was applied. By using intraclass correlation, the prestudy agreement analysis was evaluated. Bland-Altman, Passing-Bablok, and Lin’s concordance corre- lation were used to evaluate the agreement of BP measurements taken by different methods. Clinical significance was defined as a level of agreement (LoA) >20 mmHg for systolic BP and >10 mmHg for dia- stolic BP for the Bland-Altman test. Correlation analysis was conducted using Lin’s concordance correlation coefficient (r_ccc). Using Pearson’s ? and bias corrected factor (b_c) values, this analysis was used to determine the strength of agreement between measurements [15]. A value of <=0.90 was considered poor agreement, 0.90-0.95 moderate agreement, 0.95-0.99 substantial agreement and >=0.99 near perfect agreement [16]. Passing-Bablok analysis data included intercept, slope, and residual standard deviation (RSD). The residual standard deviation indicates random bias in this analysis. A slope 95% CI including 1 indi- cates no proportional bias, and an intercept 95% CI including 0 indicates no constant bias [17]. The presence of systematic bias was determined by using a Bland-Altman analysis. An intercept of 0 for the Bias 95% CI indicates no systematic bias in the data. We ensured that the difference between two measurements was normally distributed before perform- ing a Bland-Altman analysis. The LoA are defined as the mean differ- ence +- the standard deviation of the differences [18]. Statistical significance was defined as a p value of <=0.05 for all tests.

1. Results
   1. Intraclass-correlation of EMTs

Before the study, 10 volunteers who were not included in the study were measured consecutively by both EMTs using their right and left arms, and their agreement was compared. This was used to assess the agreement of the EMTs who would perform manual BP measurements. Intraclass correlation was 0.932 for systolic BP (95% CI 0.812-0.976) and 0.957 for diastolic BP (95% CI 0.878-0.986).

• 1. Main results

We recruited 50 volunteers for our study. For noisy and ambient en- vironments, we divided the volunteers into two groups of 25 each. Table 1 shows no significant differences between groups based on age or gender. Automatic BP measurements showed no significant differ- ence between the groups in the proportion of people with elevated BP defined as systolic BP above 140 mmHg or diastolic BP above 90 mmHg (p = 0.387).

The median noise level in the ambient environment was 46.75 (IQR 41.2-55.18) dB, whereas the noisy environment was 92.35 (IQR 88-96.55) dB (p < 0.001). The mean systolic BP measured in ausculta- tory method in the ambient environment was 112.02 +- 17.86 mmHg,

Table 1

Demographic characteristics of participants.

Table 4

Lin’s concordance correlation analysis.

	Ambient (n = 25)	Noisy (n = 25)	p value		Method	Noise	Lin’s rccc (95% CI)	Precision (?)	Accuracy (Cb)
Age	30 (24-46.25)	34 (29.75-41.50)	0.341		SBP	Ambient	0.943 (0.906-0.966)	0.958	0.985
Gender (female)	17 (68%)	17 (68%)	0.99			Noisy	0.574 (0.419-0.697)	0.778	0.738
Noise Level (dB)	46.75 (41.2-55.18)	92.35 (88-96.55)	<0.001		DBP	Ambient	0.957 (0.93-0.974)	0.966	0.99
High BP?	8 (16%)	6 (12%)	0.387			Noisy	0.544 (0.326-0.707)	0.567	0.959

BP: Blood pressure.

* High BP is defined as either systolic BP >140 mmHg or diastolic BP >90 mmHg.

SBP: Sistolic blood pressure; DBP: Diastolic blood pressure; r_ccc: Lin’s concordance correlation coefficient.

* A value of <=0.90 was considered poor agreement, 0.90-0.95 moderate agreement,

0.95-0.99 substantial agreement and >= 0.99 near perfect agreement for Lin’s concordance analysis [16].

Table 2

Baseline characteristics of the measurements.

	Method	Environment	n	Min	Max	Mean	Median	SD
SBP	Oscillometric	Ambient	50	90	173	114.76	107.5	19.68
	Auscultatory		50	88	170	112.02	108	17.86
DBP	Oscillometric		50	51	111	71.74	67.5	14.4
	Auscultatory		50	54	106	72.06	70	12.55
SBP	Oscillometric	Noisy	50	78	161	112.96	111.5	19.38
	Auscultatory		50	72	135	99.04	97	15.46
DBP	Oscillometric		50	54	99	72.82	72	9.7
	Auscultatory		50	54	90	70.14	72	11.04

SBP: Sistolic blood pressure; DBP: Diastolic blood pressure; Oscillometric: Automated oscillometric blood pressure measurement method; Auscultatory: Manual oscultative sphyngometric blood pressure measurement method.

while the mean oscillometric BP measurement was 114.76 +-

19.68 mmHg (Table 2). According to Bland-Altman analysis, the mean bias was -2.74 (95% confidence interval: -4.37-1.11) mmHg (p = 0.001). The lower limit of LoA was -13.96 (-16.76 to -11.16) mmHg and the upper limit was 8.48 (5.68 to 11.28) mmHg, which were both below the pre-established acceptable limit of agreement (Table 3). A concordance analysis (Table 4) revealed moderate agree- ment between the methods (r_ccc 0.943, 95% CI 0.906-0.966, ? 0.958, c_b 0.985). Based on the Passing-Bablok analysis, neither intercept value (7.4 [95% CI -4 to 20.5]) nor slope value (0.9 [95% CI 0.8 to 1.01]) exhibited proportional or constant bias (Table 5).

In the noisy environment, we calculated a mean oscillometric sys- tolic BP value of 112.96 +- 19.38 mmHg, and a mean auscultatory BP value of 99.04 +- 15.46 mmHg. Using the Bland-Altman analysis, we found a bias of -13.92 (95% CI -17.38 to -10.46) mmHg (Fig. 1). We found that the lower limit of LoA was -37.77 (95% CI -43.73 to

-31.83) mmHg and the upper limit of LoA was 9.94 (95% CI 3.99 to 15.89) mmHg, which were higher than the pre-study limit of 20 mmHg. The concordance correlation analysis showed poor agree- ment between the measurement methods (r_ccc 0.574 [95% CI 0.419 to 0.697], ? 0.778, c_b 0.738). There was no constant bias between methods in the Passing-Bablok analysis (intercept 10.44 [95% CI

-7.15 to 24.96). The slope value of 0.78 (95% CI 0.64 to 0.93) showed a proportional bias between methods (Fig. 2).

In the ambient noise environment, the mean oscillometric diastolic BP value was 71.74 +- 14.4 mmHg, while the mean auscultatory BP value was 72.06 +- 12.55 mmHg. According to the Bland-Altman analy- sis, no systematic bias existed between measurements (mean bias 0.32 [95% CI -0.81 to 1.45]). Lower limit of LoA of -7.44 (95% CI -9.38 to

-5.51) mmHg and an upper limit of LoA of 8.08 (95% CI 6.15 to 10.02) mmHg were both below study set values. According to the con- cordance correlation analysis, the approaches were in substantial agree- ment (r_ccc 0.957 [95% CI 0.93 to 0.974], ? 0.966, b_c 0.99). As a result of Passing-Bablok analysis of the methods, the intercept 7.44 (95% CI 1.5 to 12.36) indicated a constant bias, while the slope 0.91 (95% CI 0.83 to 1.01) showed no proportional bias.

The mean oscillometric diastolic BP value in the noisy environment was 72.82 +- 9.7 mmHg whereas the mean auscultatory BP was 70.14 +-

11.04 mmHg. According to the Bland-Altman analysis, there was no sys- tematic bias between measurements (mean bias -2.68 [95% CI -5.44 to 0.08] mmHg). It was found that the lower limit of LoA was -21.73 (95% CI -26.48 to -16.98) mmHg and upper limit of LoA was 16.37 (95% CI 11.62 to 21.12) mmHg, which exceeded the pre-study limits. There was poor agreement between the methods based according to concordance correlation analysis (r_ccc 0.544, ? 0.567, b_c 0.959).

Intercept (-20.01 [95% CI -59.38 to 2.59]) and slope (1.22 [95% CI

0.91 to 1.76]) showed no constant or proportional bias between methods, according to a Passing-Bablok analysis of the methods.

1. Discussion

The results of this comparative simulation study indicate that oscillometric and auscultatory BP measurements have poor agreement under noise.

In this study, we used three different statistical methods to examine the agreement between the different methods of measuring BP, as well as examining the individual sub-features of the agreement. As a base- line, we set a level of agreement for the Bland-Altman analysis at 20 mmHg or above for systolic BP and 10 mmHg or above for diastolic BP before starting the study. A noise effect adversely affected the method agreement in our study and resulted in higher LoA than set

Table 3

Bland-Altman analysis.

Method	Noise	p	Mean Bias (95% CI)	Lower LoA (95% CI)	Upper LoA (95% CI)
SBP	Ambient	0.001	-2.74	-13.96	8.48
			(-4.37 to -1.11)	(-16.76 to -11.16)	(5.68 to 11.28)
	Noisy	<0.001	-13.92	-37.77	9.94
			(-17.38 to -10.46)	(-43.73 to -31.83)	(3.99 to 15.89)
DBP	Ambient	0.57	0.32	-7.44	8.08
			(-0.81 to 1.45)	(-9.38 to -5.51)	(6.15 to 10.02)
	Noisy	0.0569	-2.68	-21.73	16.37
			(-5.44 to 0.08)	(-26.48 to -16.98)	(11.62 to 21.12)

SBP: Sistolic blood pressure; DBP: Diastolic blood pressure; LoA: level of agreement.

* The pre-study LoA limit was set as above 20 mmHg for SBP and above 10 mmHg for DBP for the Bland-Altman analysis.

Table 5 Passing-Bablok analysis.
		Intercept (95% CI)	Slope (95%CI)	RSD (+- 1.96 RSD)	Cusum Test
SBP	Ambient Noisy	7.4 (-4 to 20.5) 10.44	0.9 (0.8 to 1.01) 0.78	4 (-7.84 to 7.84) 8.13	0.68 0.14
		(-7.15 to 24.96)	(0.64 to 0.93)	(-15.94 to 15.94)
DBP	Ambient Noisy	7.44 (1.5 to 12.36) -20.01	0.91 (0.83 to 1.01) 1.22	2.55 (-4.99 to 4.99) 6.91	0.14 0.01
		(-59.38 to 2.59)	(0.91 to 1.76)	(-13.55 to 13.55)

SBP: Sistolic blood pressure; DBP: Diastolic blood pressure; RSD: Residual standard deviation.

* A slope 95% CI including 1 indicates no proportional bias, and an intercept 95% CI including 0 indicates no constant bias [17].

values pre-study. The Passing-Bablok method also revealed propor- tional bias in our study. It is evident from this result that the difference between the results becomes greater as BP increases. We found that the measurement methods are correlated both in noisy environments and in ambient environments. A statistically significant correlation, how- ever, is natural for methods designed to measure the same data, so it would not be correct to conclude from this correlation alone that these methods are equivalent.

The auscultatory and oscillometric methods of measuring BP are generally considered to produce well-matched results [6]. However, some studies have reported inaccurate results due to technological or

methodological differences. Oscillometric device validation guidelines state that at least 45 of 50 systolic and diastolic BP measurements should differ by <5 mmHg and at least 48 of 50 measurements should differ by <10 mmHg [12]. In the ambient environment, our BP measure- ments complied with Guideline recommendations.

First BP values are indicators of patient hemodynamics and may pro- vide early information about prehospital, emergency department and even intensive care unit processes [19-21]. Therefore, to achieve the most accurate result, the most accurate measurement should be used. During a study conducted with healthy adult volunteers at the scene, in an ambulance, in an emergency department, and in a quiet

Fig. 1. Comparison of auscultatory and auscillometric blood Pressure measurements in ambient and noisy environments using a Bland-Altman analysis.

Fig. 2. Comparison of auscultatory and auscillometric blood pressure measurements in ambient and noisy environments using a Passing-Bablok analysis.

environment, Hunt et al. found that there were differences in BP mea- surements only in transportation and controlled environments [22]. In our study, we assessed the noise levels of both pre-hospital ambulances and emergency departments to determine the impact of noise on BP measurements in different environments.

In spite of its simulation nature, our study differs fundamentally from that of the environment it simulates. Aside from being noisy, am- bulances are also mobile structures. Although we were able to simulate ambulance noise, we conducted our study in a stationary, immobile en- vironment that differs from an ambulance. However, in a previous study, Talke et al. showed that the motion of the environment in BP measurement was not significantly different compared to the motion in a stationary environment [23]. As a result of this data, it still remains true that the main limitation of BP measurement in ambulances is the noise. This issue may be advanced through further studies with real patients in which both noise and motion are considered confounding factors.

1. Limitations

There are several limitations to our study. First, this study aimed to determine the effect of noise on the agreement of BP measurement methods. It is important to note, however, that this is a simulation study conducted on healthy volunteers. It is possible that simulation studies may differ from real-life studies, even though they seem to min- imize the effect of confounding factors. Second, we conducted our study in a single center with one device for each measurement method and

two measurers. Since the auscultatory BP measurement method is likely to be affected by individual differences, we think that increasing the number of measurers may be more compatible with real-life data. Because previous studies used different devices for data collection, there is a need for studies that utilize more devices. A limitation of our study is that only decibel calculations were made without evaluating sound frequency values. Although the study captured the noise level in- side the ambulance, it is unclear whether the same frequency as the sound was captured.

1. Conclusion

This simulation study performed on healthy volunteers indicates that oscillometric and auscultatory methods of measuring BP are in good agreement in an ambient environment, but that this agreement is significantly impaired by noise. It is possible that these findings may suggest that auscultatory methods for measuring BP should be avoided in loud environments if real-life studies support them.

Meetings

The article has not been presented in any organization.

Grant or other financial support

None.

Author contributions

ACT, and SY conceived the study, designed the trial. ACT, and SY supervised the conduct of the trial and data collection. ACT, and SY undertook recruitment of participating centers and patients and man- aged the data, including quality control. ACT and SY provided statistical advice on study design and analyzed the data; ACT chaired the data oversight committee. SY drafted the manuscript, and all authors con- tributed substantially to its revision. ACT and SY takes responsibility for the paper as a whole.

CRediT authorship contribution statement

Ali Cankut Tatliparmak: Software, Resources, Project administra- tion, Methodology, Investigation, Formal analysis, Data curation, Con- ceptualization, Validation, Visualization, Writing - review & editing. Sarper Yilmaz: Visualization, Validation, Supervision, Software, Meth- odology, Investigation, Formal analysis, Data curation, Conceptualiza- tion, Project administration, Resources, Writing - original draft, Writing - review & editing.

Declaration of Competing Interest

The authors declare that there is no conflict of interest regarding the publication of this article.

References

1. Sapra A, Malik A, Bhandari P. Vital sign assessment. StatPearls. StatPearls Publishing;

2022 Accessed December 18, 2022. http://www.ncbi.nlm.nih.gov/books/ NBK553213/.

1. Schulze MB, Kroke A, Saracci R, Boeing H. The effect of differences in measurement procedure on the comparability of blood pressure estimates in multi-Centre studies. Blood Press Monit. 2002;7(2):95-104. https://doi.org/10.1097/00126097- 200204000-00002.
2. Tolonen H, Koponen P, Naska A, et al. Challenges in standardization of blood pres- sure measurement at the population level. BMC Med Res Methodol. 2015;15:33. https://doi.org/10.1186/s12874-015-0020-3.
3. James GD, Gerber LM. Measuring arterial blood pressure in humans: Auscultatory and automatic measurement techniques for human biological field studies. Am J Hum Biol. 2018;30(1). https://doi.org/10.1002/ajhb.23063.
4. Babbs CF. The origin of Korotkoff sounds and the accuracy of auscultatory blood pressure measurements. J Am Soc Hypertens. 2015;9(12):935-950.e3. https://doi. org/10.1016/j.jash.2015.09.011.
5. Lewis PS. British and Irish hypertension Society’s blood pressure measurement working party. Oscillometric measurement of blood pressure: a simplified explana- tion. A technical note on behalf of the British and Irish hypertension society. J Hum Hypertens. 2019;33(5):349-51. https://doi.org/10.1038/s41371-019-0196-9.
6. Amadasun FE, Isa JI. A comparison of sphygmomanometric and oscillometric methods of blood pressure measurements in adult in-patients. Niger J Clin Pract. 2005;8(2):86-9.
7. Eliasdottir SB, Steinthorsdottir SD, Indridason OS, Palsson R, Edvardsson VO. Com- parison of aneroid and oscillometric blood pressure measurements in children. J Clin Hypertens (Greenwich). 2013;15(11):776-83. https://doi.org/10.1111/jch. 12196.
8. Liu J, Ou Y, Yan BP, Sodini C, Zhao N. Comparisons of Oscillometric blood pressure measurements at different sites of the upper limb. Annu Int Conf IEEE Eng Med Biol Soc. 2018;2018:1168-71. https://doi.org/10.1109/EMBC.2018.8512349.
9. Landgraf J, Wishner SH, Kloner RA. Comparison of automated oscillometric versus auscultatory blood pressure measurement. Am J Cardiol. 2010;106(3):386-8. https://doi.org/10.1016/j.amjcard.2010.03.040.
10. Filus W, de Lacerda ABM, Albizu E. Ambient noise in emergency rooms and its health hazards. Int Arch Otorhinolaryngol. 2015;19(3):205-9. https://doi.org/10.1055/s- 0034-1387165.
11. Stergiou GS, Palatini P, Asmar R, et al. Recommendations and practical guidance for performing and reporting validation studies according to the universal standard for the validation of blood pressure measuring devices by the Association for the Advancement of medical instrumentation/European Society of Hypertension/Inter- national Organization for Standardization (AAMI/ESH/ISO). J Hypertens. 2019;37 (3):459-66. https://doi.org/10.1097/HJH.0000000000002039.
12. Ludbrook J. Linear regression analysis for comparing two measurers or methods of measurement: but which regression? Clin Exp Pharmacol Physiol. 2010;37(7): 692-9. https://doi.org/10.1111/j.1440-1681.2010.05376.x.
13. Bablok W, Passing H. Application of statistical procedures in analytical instrument testing. J Autom Chem. 1985;7(2):74-9. https://doi.org/10.1155/ S1463924685000177.
14. Lin LI. A concordance correlation coefficient to evaluate reproducibility. Biometrics. 1989;45(1):255-68.
15. Mahon G. A Proposal for Strength-of-Agreement Criteria for Lin’s Concordance Cor- relation Coefficient. Accessed December 18, 2022. https://www.semanticscholar. org/paper/A-Proposal-for-Strength-of-Agreement-Criteria-for-Mahon/0bed3 bdb75247d09d4c5f1398cd309272fd26ad5; 2005.
16. Bilic-Zulle L. Comparison of methods: Passing and Bablok regression. Biochem Med (Zagreb). 2011;21(1):49-52. https://doi.org/10.11613/bm.2011.010.
17. Dogan NO. Bland-Altman analysis: a paradigm to understand correlation and agree- ment. Turk J Emerg Med. 2018;18(4):139-41. https://doi.org/10.1016/j.tjem.2018. 09.001.
18. Jouffroy R, Saade A, Ellouze S, et al. prehospital triage of septic patients at the SAMU regulation: comparison of qSOFA, MRST, MEWS and PRESEP scores. Am J Emerg Med. 2018;36(5):820-4. https://doi.org/10.1016/j.ajem.2017.10.030.
19. Jouffroy R, Gilbert B, Gueye PN, et al. Prehospital hemodynamic optimisation is asso- ciated with a 30-day mortality decrease in patients with septic shock. Am J Emerg Med. 2021;45:105-11. https://doi.org/10.1016/j.ajem.2021.02.060.
20. Tazarourte K, Cesareo E, Sapir D, et al. Update on prehospital emergency care of se- vere trauma patients. Ann Fr Anesth Reanim. 2013;32(7-8):477-82. https://doi.org/ 10.1016/j.annfar.2013.07.005.
21. Hunt RC, Allison EJ, Whitley TW, Robinson MJ, Smart TL. Comparison of EMT blood pressure measurements with an automated blood pressure monitor: on scene, dur- ing transport, and in the emergency department. Ann Emerg Med. 1985;14(9): 871-5. https://doi.org/10.1016/s0196-0644(85)80636-3.
22. Talke PO. Measurement of systolic blood pressure using pulse oximetry during helicopter flight. Crit Care Med. 1991;19(7):934-7. https://doi.org/10.1097/ 00003246-199107000-00018.