Predicting body weight by measuring weight underneath the buttock in supine position: A quantitative observation study

Abstract

BACKGROUND:

Measuring body weight (BW) for bedridden patients often presents difficulty and challenge.

OBJECTIVES:

The present study aimed at providing a self-designed indirect method to predict BW by measuring weight underneath the buttock (WUB) of an individual in supine position, thereby providing an easy, safe and effective way of BW measurement for bedridden patients.

METHODS:

A total of 180 subjects participated in the present study and agreed to have their BW and WUB to be measured. BW was measured normally at the standing position through an electronic weighing machine without any special requirement. By placing the electronic weighing machine under the subject’s buttock along with an additional hard board set under the electronic weighing machine, WUB was measured in subjects who were asked to assume a supine position in beds to simulate conditions of bedridden individuals. Measurement was repeated thrice to minimise the test error.

RESULTS:

Average BW (62.7 $\pm$ 11.7 kg) was 2.0 $\pm$ 0.1 ( $\approx$ 2) folds of WUB (31.4 $\pm$ 6.0 kg). Significant linear correlation was identified between BW and WUB in all subjects with a linear equation yielded ( $y=$ 1.8 x $+$ 6.0). Further multiple regression analysis resulted in an equation of BW (kg) $=$ $-$ 36.8 $+$ 1.66*WUB (kg) $+$ 29.0*height (m). Predicted BW (PBW) was calculated out based on the results described above: the multiple relationship (2 folds), the linear equation, and the multiple regression equation, and differed from the measured BW by 3.6 $\pm$ 2.8, 3.5 $\pm$ 2.7 and 4.2 $\pm$ 3.1 kg respectively.

CONCLUSIONS:

Predicting BW through WUB in supine position is effective and reliable because the latter can be easily measured and features a strong linear relationship with BW. This method provides clinical staff with remarkable benefits in BW determination for bedridden patients.

Keywords

Bedridden body weight supine position

1. Introduction

Measuring body weight (BW) is a routine clinical work for patient assessment and helps us evaluate patients’ nutrition status, guide their food intake, detect oedema, determine medication dose and set suitable tidal volume of ventilator [1] and is thereby especially important for critically ill patients. However, most severely ill patients are bedridden, causing significant difficulty in measuring BW.

Currently available methods used to measure BW of bedridden patients include Hill-Rom bed [2], bed scale [3, 4], ceiling hoist scale [5], chair scale, DY-type multifunctional portable medical weight measurement [6, 7, 8, 9] and determination from mid-arm-circumference measurement [10, 11, 12] or by height, circumference of chest, arm and waist [13]. Most of these methods feature unique requirements, such as heavy equipment, large space or skilled staff, whereas some exhibit poor accuracies. Other disadvantages include high expense, high risks, complexities and large deviation.

Based on the current situation, the present study aimed at providing, for bedridden patients who cannot stand up, an easy and dependable way to measure their BW, which is calculating BW through measuring the weight underneath the buttock (WUB) in supine position based on the assumption of linear correlation between WUB and BW.

2. Methods

2.1 Subjects

A total of 180 independent-living individuals who can stand up (62 males and 118 females, 21 $\sim$ 88 years old) volunteered to participate in the current study.

2.2 Study protocols and methods

All subjects agreed to have their BW and WUB measured. No adverse event had happened during the test.

BW measurement was performed in the usual manner. After taking off thick coat and shoes, subjects stood firm on a weighing machine, which was placed on the smooth and hard ground, and their BW was recorded.

WUB measurement was performed when subjects lay in an ordinary sickbed at supine position, simulating conditions of bedridden individuals, with the weighing machine and hard board placed together underneath the subject’s buttock (the order from top to bottom was patient buttock, weighing machine, hard board and sickbed). The board should be hard enough to support the pressure from the subject’s buttock without deformation and large enough to support the weighing machine (its surface area should be larger than that of weighing machine). Each subject was asked to maintain a straight body, relaxed with the hip lying on the centre of the scale and with each arm lying on the side of the body. Measurement was repeated thrice.

Three kinds of weighing machines were applied. Firstly, with an ordinary electronic weighing machine (60 subjects), during WUB measurement, we often asked the subjects to tilt their body to read the numbers, especially for overweight or large-sized subjects because the machine was covered by the subject’s buttock. Secondly, with a Bluetooth electronic weighing machine (60 subjects), results can be read from a mobile device with Bluetooth connection, which made measurement easier. Thirdly, with an electronic weighing machine with digital readout retention function (60 subjects), results can be read after the machine was taking out or subjects stood up to avoid any reading problem.

2.3 Statistical analysis

Data were presented as mean $\pm$ standard deviation (SD) unless indicated otherwise and were analysed using Statistics Software SPSS 16.0. Pearson correlation analysis was applied to determine the correlation between BW and WUB. Multiple regression analysis was applied to examine the factors associated with BW and yield the regression equation of BW. Correlation distance test was performed to determine differences among thrice-repeated measurements. $P<$ 0.05 was considered statistically significant.

3. Results

3.1 BW, WUB and related data

General characteristics of 180 subjects are listed in Table 1. We found that BW was 2.0 $\pm$ 0.1 ( $\approx$ 2) folds of WUB, and significant correlation existed between BW and WUB with a linear equation of BW $=$ 1.8*WUB $+$ 6.0 in all subjects (Table 1).

Table 1
General characteristics of subjects, BW, WUB and related data

n	60	60	60	180
	[baseline=(char.base)] [shape=circle,draw,inner sep=0.2pt] (char) 1; ${}^{\rm a}$	[baseline=(char.base)] [shape=circle,draw,inner sep=0.2pt] (char) 2; ${}^{\rm a}$	[baseline=(char.base)] [shape=circle,draw,inner sep=0.2pt] (char) 3; ${}^{\rm a}$	All
Males/females (n)	22/38	24/36	16/44	62/118
Age (y)	43.3 $\pm$ 15.8	44.4 $\pm$ 14.0	43.0 $\pm$ 14.6	43.5 $\pm$ 14.8
Height (m)	1.63 $\pm$ 0.07	1.64 $\pm$ 0.08	1.64 $\pm$ 0.07	1.64 $\pm$ 0.08
BW (kg)	60.6 $\pm$ 10.5	65.5 $\pm$ 11.5	61.9 $\pm$ 12.7	62.7 $\pm$ 11.7
WUB(kg)	31.1 $\pm$ 5.5	32.6 $\pm$ 6.3	30.6 $\pm$ 6.1	31.4 $\pm$ 6.0
BW/WUB ${}^{\rm b}$	1.95 $\pm$ 0.15	2.02 $\pm$ 0.12	2.03 $\pm$ 0.14	2.02 $\pm$ 0.13
Correlation ${}^{\rm c}$ (r)	0.90 ${}^{**}$	0.95 ${}^{**}$	0.94 ${}^{**}$	0.93 ${}^{**}$
Linear equation ${}^{\rm d}$	y $=$ 1.7 x $+$ 7.0	y $=$ 1.7 x $+$ 9.0	y $=$ 1.9 x $+$ 2.9	y $=$ 1.8 x $+$ 6.0

${}^{\rm a}$ : [baseline=(char.base)] [shape=circle,draw,inner sep=0.2pt] (char) 1;, [baseline=(char.base)] [shape=circle,draw,inner sep=0.2pt] (char) 2; and [baseline=(char.base)] [shape=circle,draw,inner sep=0.2pt] (char) 3; represent three different kinds of weighing machines were used respectively (same in all other tables). ${}^{\rm b}$ : BW is divided by WUB. ${}^{\rm c}$ : Correlation between BW and WUB. ${}^{\rm d}$ : y represents BW and x represents WUB. ${}^{**}$ : $P<$ 0.05.

3.2 Multiple linear regression equation

A multiple linear regression equation for BW was obtained from multiple regression analysis: BW (kg) $=$ $-$ 36.8 $+$ 1.66*WUB (kg) $+$ 29.0*Height (m) (Table 2).

Table 2
Predicting model for BW.

Caption	Coefficient (SE)	$P$	95% Confidence interval
$r^{2}=$ 0.886
Height	29.0 (4.4)	$<$ 0.05	20.4 to 37.6
WUB	1.66 (0.05)	$<$ 0.05	1.55 to 1.76
Constant	$-$ 36.8 (6.6)	$<$ 0.05	$-$ 49.9 to -23.7

3.3 Predicted BW (PBW) and its differences from BW

Based on the results above, we can see the PBW could be deduced in 3 ways: firstly, simply by WUB times 2 according to their 2 times multiple relationship, secondly, based on the linear equation of BW and WUB, and thirdly on the multiple regression equation. The detailed calculated results based on these three methods are listed in Table 3.

Table 3
Calculated PBW and its differences from measured BW

Items	Methods ${}^{\rm b}$	[baseline=(char.base)] [shape=circle,draw,inner sep=0.2pt] (char) 1;	[baseline=(char.base)] [shape=circle,draw,inner sep=0.2pt] (char) 2;	[baseline=(char.base)] [shape=circle,draw,inner sep=0.2pt] (char) 3;	All
PBW (kg)	2-folds	62.2 $\pm$ 11.1	65.2 $\pm$ 12.5	61.1 $\pm$ 12.3	62.8 $\pm$ 12.0
	Linear	60.6 $\pm$ 9.7	65.5 $\pm$ 10.8	61.9 $\pm$ 11.9	62.7 $\pm$ 10.9
	Regression	62.0 $\pm$ 10.6	64.8 $\pm$ 11.4	61.2 $\pm$ 11.4	62.6 $\pm$ 11.0
BW-PBW (kg) ${}^{\rm a}$	2-folds	3.9 $\pm$ 3.1	3.2 $\pm$ 2.4	3.6 $\pm$ 2.8	3.6 $\pm$ 2.8
	Linear	3.5 $\pm$ 2.9	2.8 $\pm$ 2.4	3.5 $\pm$ 2.8	3.5 $\pm$ 2.7
	Regression	3.6 $\pm$ 2.6	2.7 $\pm$ 2.3	3.1 $\pm$ 2.2	3.1 $\pm$ 2.4

${}^{\rm a}$ : BW minus PBW: the difference between measured BW and calculated PBW. ${}^{\rm b}$ : Methods are the 3 ways described in the text above.

4. Repeated measurements of WUB

Errors of the thrice-repeated WUB measurements were calculated in all subjects (0.48 $\pm$ 0.36 kg). Correlation distance analysis showed strong correlations (0.97 $\sim$ 0.99) among each twice-repeated measurement for each group, indicating no difference among thrice-repeated measurements.

5. Discussions

In the present study, by measuring BW of 180 healthy subjects in standing position and WUB while in supine position, we found that BW could be predicted from WUB in three ways: 1) BW $\approx$ WUB*2 (the two times multiple relationship); 2) BW $=$ 1.8*WUB $+$ 6.0 (the linear equation); 3) BW (kg) $=$ $-$ 36.8 $+$ 1.66*WUB (kg) $+$ 29.0*Height (m) (the multiple regression equation).

In our study, the first 60 healthy subjects had their weight measured with an ordinary electronic weighing machine, and the data presented a significant correlation between BW and WUB. However, difficulties were encountered in reading the numbers from the weighing machine because the reading frame was often covered by the subject’s buttock, especially for overweight and large-sized subjects. Portable weighing machines currently available in market are generally small and exquisite. To read the results, we often asked subjects to tilt their bodies, thus possibly resulting in significant measurement error.

Several solutions were introduced to overcome this difficulty. Firstly, a new type of weighing machine of larger size should be designed to easily obtain the readings; in this machine, the reading frame is out of subject’s buttock, or the reading frame position is modified by leaving it protruding outside of the body part covering the weighing machine. With this idea, a patent has been applied to and approved by the State Patent Bureau of China. Secondly, we figured out other methods including using two different kinds of weighing machines: a Bluetooth electronic weighing machine and an electronic weighing machine with readings remaining for at least 30 s after removing the weighing material. Both weighing machines solved the reading problem. With each weighing machine, WUB was measured thrice. Errors of repeated measurements were small, and distance correlations analysis indicated strong correlations between each time of measurement.

Some limitations were noted in the current study. Firstly, our subjects are all Chinese, and their weight and height may be smaller than those of Western Caucasians or Black Africans. Secondly, sample size was small, and we believe that a larger number of subjects will yield more dependable results. Thirdly, although not specifically examined in this study, this method features some contraindications, including high muscle tension or muscle spasticity, emotional agitation or pelvic fracture, all of which may cause difficulty in body movement or obeying instructions.

In summary, a 2-fold multiple and a linear relationships are existed between BW in standing position and WUB in the supine position, and a multiple linear equation can be deduced from their relationship. Our self-designed method of predicting BW from WUB in supine position is a simple, effective and reliable indirect manner of determining BW for bedridden patients.

Footnotes

Acknowledgments

We thank all participants for their time and efforts in this study. This work was supported by the National Natural Science Foundation of China (grant number 81501944) and the Natural Science Foundation of Jiangsu Province of China (grant number BK20150353).

Conflict of interest

None declared.

Ethical approval

This study was approved by the Medical Ethics Committee of The First Affiliated Hospital of Soochow University and conducted in compliance with the guidelines stated in the WMA Declaration of Helsinki (project identification code: 2017025).

References

Stehman

Buckley

Dos Santos

Riffenburgh

Swenson

Mulligan

, et al. Bedside estimation of patient height for calculating ideal body weight in the emergency department. The Journal of Emergency Medicine. 2011; 41(1): 97-101.

Schneider

Thorpe

Dellbridge

Matalanis

Bellomo

. Electronic bed weighing vs daily fluid balance changes after cardiac surgery. J Crit Care. 2013; 28(6): 1113e1-5..

Yu Ping

Wei Hen

Dan

. a simple mattress with weighting system to support pressure ulcer prevention. Journal of Nursing (China). 2016; (8): 77-8.

Gerl

Miko

Nelson

Godaire

. Are In-Bed Electronic Weights Recorded in the Medical Record Accurate? Medsurg Nursing: Official Journal of the Academy of Medical-Surgical Nurses. 2016; 25(3): 177-81, 201.

Bloomfield

Steel

MacLennan

Noble

. Accuracy of weight and height estimation in an intensive care unit: Implications for clinical practice and research. Crit Care Med. 2006; 34(8): 2153-7.

Yang

Dai

el al., Development and Application of DY type multifunctional portable medical weight. Nurs J Chin PLA. 2005; 22(05): 101.

Hou

Yan

. Design of a weight measuring device for bedridden patient. China Medical Equipment. 2014; (11): 48-9.

Guo

Xiang

et al., Medical combination weighing device. Industrial Metrology. 2000; (S1): 256-7.

. Design and investigation for determine the weight of the clinical emergency patient. Medical Equipment. 2007; 20(12): 16-7.

10.

Freitag

Edgecombe

Baldwin

Cottier

Heland

. Determination of body weight and height measurement for critically ill patients admitted to the intensive care unit: A quality improvement project. Australian critical care: official journal of the Confederation of Australian Critical Care Nurses. 2010; 23(4): 197-207.

11.

Nwankwo

Ostchega

Zhang

Hughes

. Validating prediction equations for mid-arm circumference measurements in adults: National Health and Nutrition Examination Survey, 2001–2012. Blood Pressure Monitoring. 2015; 20(3): 157-63.

12.