The Accuracy of the Broselow TM Pediatric Emergency Tape for Weight Estimation in an Omani Paediatric Population *

OBJECTIVES
This study aimed to evaluate the accuracy of the Broselow™ Pediatric Emergency Tape (BT) for estimating weight in an Omani paediatric population at a tertiary care hospital.


METHODS
This retrospective cross-sectional study was conducted during July 2015. The electronic medical records of Omani outpatients <14 years old attending the Sultan Qaboos University Hospital, Muscat, Oman, between July 2009 and June 2013 were reviewed for recorded height and weight data. The BT Version 2002A was used to predict weight based on actual height measurements. Predicted weight measurements were then compared with actual weight to determine the accuracy of the estimation.


RESULTS
A total of 3,339 children were included in the study, of which 43.5% were female and 56.5% were male. The mean age was 6.4 ± 3.1 years and the mean height was 93.2 ± 23.5 cm. The mean actual weight was 13.9 ± 6.7 kg while the mean BT-predicted weight was 14.4 ± 6.9 kg. The BT-predicted weight estimations correlated significantly with actual weight measurements (intraclass correlation coefficient: 0.97; P <0.001). A Bland-Altman analysis indicated that the BT performed well when estimating weight among Omani children, with an overestimation of only 0.5 kg for the entire cohort.


CONCLUSION
The BT was found to be an effective tool for estimating weight according to body length in an Omani paediatric population. It should therefore be considered for use in emergency situations when actual weight cannot be determined.


Advances in Knowledge -To the best of the authors' knowledge, this is the first study from Oman to investigate the accuracy of the Broselow™ Pediatric Emergency
Tape (BT) in a paediatric population.

Application to Patient Care:
-The results of this study indicate that the BT can be used to accurately predict the weight of Omani paediatric patients; this may be useful in emergency situations or during resuscitation.

I
2][3] Inaccurate weight estimations may have adverse effects, such as causing the patient to become unresponsive or experience toxicity during certain interventions. 4owever, it is often extremely difficult to use traditional methods to determine a paediatric patient's weight during a critical or emergency situation.][7] The Broselow ™ Pediatric Emergency Tape (BT) was developed in the late 1970s as a tool to determine a child's weight based on height calculated by measuring the head-to-heel body length while the patient is in a supine position. 8,9Subsequently, the tool sorts children into weight categories based on their length, with each category assigned a colour that corresponds to a list of proportionally-sized equipment, appropriate doses of emergency drugs and safe levels of defibrillation energy. 8,9][12] It has been recommended by both the American College of Surgeons and the American Heart Association for use in their Advanced Trauma Life Support ® and Pediatric Advanced Life Support ® courses. 13,14lthough originally developed using data from a Western population, recent studies from the USA and Canada now indicate that the BT may significantly underestimate weight due to the rising prevalence of childhood obesity. 15,16In addition, to the best of the authors' knowledge, no studies have yet been performed to assess the accuracy of the BT in estimating weight in an Omani paediatric population.Nevertheless, the BT is still used in Oman to gauge weight during paediatric emergencies.As such, this study therefore aimed to determine the accuracy of the BT Version 2002A in estimating weight among an Omani paediatric population.

Methods
This retrospective cross-sectional study took place in July 2015 at the Sultan Qaboos University Hospital (SQUH), a tertiary care institution in Muscat which receives referrals from all over Oman.All paediatric Omani patients <14 years old who visited the general outpatient clinic at SQUH between July 2009 and June 2013 were included in the study.Children were excluded if their weight was >35 kg or their height was outside the recommended range (50-150 cm). 8,9atients were also excluded if they had any medical conditions that would substantially affect their weight and/or height, such as an amputation, dwarfism, a growth hormone deficiency, metabolic disease, severe joint contractures or neurological defects like cerebral palsy.In order to obtain an intraclass correlation (ICC) of 0.8 to compare BT weight estimations with actual weight, the required sample size was calculated to be 300 subjects, assuming an ICC of 0.7, with α and β errors at 5% and 10%, respectively, and an expected dropout rate of 10%.However, as many subgroup analyses were planned, a minimum of 3,000 subjects was deemed necessary.
Hospital electronic medical records were reviewed to collect data for all eligible subjects during their initial outpatient visit.All children were weighed on a standard calibrated paediatric scale and their height was measured using a stadiometer.For children <2 years old, recumbent length was measured using a standard medical measuring tape and weight was calculated using an electronic infant scale while the child was wearing diapers.Weight and body length/height were measured to the nearest 0.1 kg and 1 cm, respectively.During the retrospective chart review, patient information was collected using a data sheet, including their name, medical record number, age in years, gender, body length/height and weight.The 2002A version of the BT was then used to estimate the weight of each patient based on actual body length/height measurements collected from medical records.A regression analysis was used to predict weight using the BT.

Results
A total of 3,339 children were included in the study.Of these, 1,451 (43.5%) were female and 1,888 (56.5%) were male.The majority of the participants were 5-9 years old (53.9%), while 28.8% were ≤5 years old and 17.0% were ≥10 years old.The mean age was 6.4 ± 3.1 years and the mean height was 93.2 ± 23.5 cm.The mean actual weight was 13.9 ± 6.7 kg, while the mean BT-predicted weight was 14.4 ± 6.9 kg.The mean difference between actual weight and BTpredicted weight was -0.5 ± 1.9 kg.The combined average of both actual and BT-predicted weight was 14.1 ± 6.7 kg [Table 1].
The ICC of the actual weight and BT-predicted weight measurements was 0.97 (ICC coefficient: 0.97-0.98;P <0.001).A scatter plot of actual weight measurements and BT-predicted weight estimates is shown in Figure 1.The linear regression equation was as follows: where W is the actual weight.This equation implies that a 1 kg increase in actual weight results in a 0.9 kg increase in BT-predicted weight, with a near-perfect correlation between the two measurements.The Bland Altman plot is shown in Figure 2.
The mean bias was -0.49 ± 1.98 (95% confidence interval: -2.46-1.50)with the limits of agreement at -4.37 and 3.40.A total of 53.7% and 85.4% of the BT-predicted weight estimates fell within 10% and 20% of the actual weights, respectively.The precision of the BT-predicted weight estimates was 3.9% and the percentage of error was 27.9%.For children weighing <15 kg, the overestimation was negligible; however, when the actual weight was 15-25 kg, a small difference between actual and BT-predicted weight was observed.Nevertheless, as the percentage of error was within 30%, BT-predicted weight estimation was deemed very reliable and comparable to actual weight measurements. 18,19

Discussion
2][3] The BT is an accessible and unbiased tool to estimate a child's weight based on their body length/height and is useful during paediatric resuscitation. 8,9However, as the BT was not specifically developed for Omani children, it is important to determine its accuracy among this specific population.Using the BT Version 2002A, the results of this study confirmed that BTpredicted weight estimates correlate with actual weight in an Omani paediatric population.][23][24] Furthermore, other indices also suggested that the  BT sufficiently predicts actual weight.Although the BT slightly overestimated the actual weight of the population in the current study by 0.5 kg, the percentage of error still fell within acceptable limits at <30%. 18,19Nevertheless, while overestimation was negligible in smaller children weighing <15 kg, it was more pronounced in children weighing between 15-25 kg.[25][26][27] The BT has been shown to perform differently in various populations.9][30] As a result, some researchers have advised that certain adjustments be made to the BT to improve its accuracy for these particular populations. 28,29Conversely, studies from more economically developed countries have found that the BT tends to underestimate weight among paediatric populations. 5,15,25,31,32These contradictory data may reflect the increasing prevalence of obesity in more developed countries. 15,16he strengths of the current study include its large sample size and the collection of data from a tertiary care centre which receives referrals from all over the country.Therefore, the subjects included in this study may be considered representative of the general Omani paediatric population.However, a potential limitation of this study was the assumption that the sample population were of the same weight as children who require resuscitation.Furthermore, the most recent version of the BT, the BT Version 2011, was not used to estimate weight.This is because the BT Version 2002A is currently used at SQUH in emergency settings.Moreover, intra-rater reliability was also not estimated and the study was conducted in only one hospital.

Conclusion
The BT was found to accurately predict weight in an Omani paediatric population.As such, it can be considered an effective tool to estimate paediatric weight in emergency situations where the actual weight of the child cannot be easily determined.Accurate weight estimation allows physicians to administer the correct drug doses, use proportionately sized instruments and determine safe shock levels for defibrillation when treating critically ill children.
7 14.1 ± 6.7 SD = standard deviation; BT = Broselow ™ Pediatric Emergency Tape.*Difference between actual weight and BT-predicted weight.† Average of actual weight and BT-predicted weight.Computing, Vienna, Austria) and the Statistical Package of the Social Sciences (SPSS), Version 22 (IBM Corp., Armonk, New York, USA) was used for the analysis.This study received ethical approval from the Research & Ethics Committee of the College of Medicine & Health Sciences, Sultan Qaboos University (MREC #1024).

Figure 1 :
Figure 1: Scatterplot comparing actual weight measurements with predicted weight estimates using the Broselow ™ Pediatric Emergency Tape among Omani paediatric outpatients at the Sultan Qaboos University Hospital, Muscat, Oman (N = 3,339).BT = Broselow ™ Pediatric Emergency Tape.

Figure 2 :
Figure 2: Bland-Altman plot comparing the mean weights of actual weight measurements and predicted weight estimates using the Broselow ™ Pediatric Emergency Tape (BT) with the differences between actual weight and BTpredicted weight among Omani paediatric outpatients at the Sultan Qaboos University Hospital, Muscat, Oman (N = 3,339).BT = Broselow ™ Pediatric Emergency Tape; SD = standard deviation.

Table 1 :
Statistical software including R software, Version 3.2.2(R Foundation for Statistical Mean age, height and weight according to gender of Omani paediatric outpatients at the Sultan Qaboos University Hospital, Muscat, Oman (N = 3,339)