A Study of Weight Approximation by Practicing EMS Providers

Abstract

Emergency medical service (EMS) providers are often required to accurately estimate a patient’s weight via visual assessment with little to no input from the patient. While this is an important skill, it is often not taught at any point in the initial or continuing education process. It is a skill that must largely be self-taught by comparing estimates made by providers with a patient’s self-reported weight. Weight approximation skills are especially important for paramedics administering medications in the field. The goals of this analysis were to quantify accuracy of weight estimation skills for current providers and to determine if this skill improved with years of experience. 183 EMS providers (paramedics and EMT-Bs) from the Irving (TX) Fire Department (IFD) were tasked with approximating the weights of at least five other providers to calculate the accuracy of weight estimation. Adjusting the acceptable percent error led to the following conclusions: Providers’ weight estimates were within 5% of the measured weights in 57% of the patients, within 10% of measured weights in 79% of the patients, and within 20% of measured weights in 99% of the patients. The authors conclude that significant variability exists in the abilities of practicing EMS providers to estimate patient weight in the non-emergency setting.

Introduction

In emergency medicine, medications are often prescribed and administered based on the patient’s weight. Medications prescribed by weight-based dosing tend to be those where a “one-size-fits-all” dose is less than ideal. Tailoring such dosing to the patient is always the best approach. Therefore, an accurate weight for the patient is a necessary basis for targeting treatment.

When administered in a controlled environment (such as the emergency department), weight-based administration is a relatively straightforward task aided by validated patient weight from direct measurement utilizing a scale. Such measurements provide the patient weight in kilograms (kg) corresponding to the units in which the medications are prescribed: Typically weight of the medication (such as milligrams) per kilogram of the patient’s documented weight.

Related

In an uncontrolled environment (such as in the ambulance, a patient’s home, or in an outdoor or public area), generally no scale is available. The provider must rely on a patient’s self-reported weight or make an estimation of the patient’s weight. Sometimes the provider must do both, such as when patients do not know their weight or obviously report it inaccurately. As a compounding factor, weights are almost universally reported (or estimated) in pounds, which must then be converted to kilograms. This confounding factor is not considered in the present study but can be an additional source of error.

Therefore, accuracy of weight estimation is critical in delivering accurate medication dosing. In the recently-reported high-profile arrest, sedation, and subsequent death of an unarmed man in Colorado, inaccurate weight estimation may have contributed in a significant way to the accidental overdose of ketamine for the patient. Case details revealed that the paramedic had estimated the subject’s weight to be around 200 pounds (90.7 kg) and had administered a 500 mg dose of ketamine. However, it was revealed during the trial that the subject actually weighed 143 pounds (65 kg), making a 325 mg dose suitable for him instead. Consequently, a forensic pathologist stated that the high dosage of ketamine might have played a role in the death of the patient.¹

This case engendered the creation of a Colorado state law limiting ketamine administration for sedation to medical professionals and not police officers. The new regulations also mandate monitoring for patients under ketamine sedation. Moreover, ketamine can no longer be used on aggressive, paranoid, agitated, or disoriented patients due to the subjective nature of the diagnosis without getting a waiver from the Colorado Department of Public Health and Environment.

A study conducted in San Diego (CA) measured the comfort level and accuracy of paramedics in making pediatric weight estimates. It was found that despite receiving no formal training in the area, the paramedics were somewhat accurate in estimating pediatric patient weight.² Another study compared paramedic weight estimates of cardiac patients to their actual weight, thereby measuring the accuracy of the approximations. This study utilized a Pearson correlation analysis and concluded that the paramedic weight estimations correlated well with actual patient weights.³ A study conducted in two Australian Universities found that undergraduate nurses and paramedics have varying abilities to accurately estimate patient weight.⁴

Interestingly, two other studies seem to contradict the above conclusions. These separate studies compared paramedics’ ability to estimate patient weight to that of emergency department (ED) personnel. Both studies found that paramedics were least likely to make accurate weight approximations as compared to ED staff members. They also concluded that, when compared to current weight estimation methods, patients’ self-reported weights are the most accurate.^5,6 However, in emergency situations, the option of gaining the reported weight from a patient may be difficult or impossible.

If accuracy of medication administration is based upon estimates of patient weight, how are providers in the field trained to accomplish this task? Even if trained, how accurate are these skills? The answers to these questions become more important when considering the potential negative consequences for the patient related to potential under- or overdosing of a patient on certain medications.

Hypothesis

The authors proposed that experienced EMS providers could estimate a patient’s weight within 20% of the actual measured weight, and that the estimates would be more accurate based upon years of experience.

Methods

One-hundred and three EMS providers from the Irving Fire Department (IFD) were tasked with estimating the weights of their colleagues and themselves. Providers’ estimates of body weights were compared to actual measured weights. Samples were included in the analysis only if a provider’s experience, weight estimates, and corresponding measured weights were documented.

Between 6 to 25 estimates of weights were made per provider participant. For each provider, the number of weight estimates that deviated by 5%, 10%, and 20% from actual value were determined to measure accuracy. The Pearson’s Correlation Coefficient, R, between estimated weight and actual weight was calculated for each estimate, which was then evaluated against the number of years of experience of each provider.

Results

The weight estimates fell within 5% of the measured weights in 57% of the target individuals, within 10% of measured weights in 79%, and within 20% of measured weights in 99%.

Pearson’s Correlation Coefficient, R, between estimated and actual weights was calculated for each provider with the following results: 99 out of 183 providers (54%) had R≥0.9, and 157 out of 183 (86%) of the providers had R≥ 0.8.

Pearson’s Correlation Coefficient (between estimated and actual weights) for each provider was plotted against the years of experience of the provider. The authors expected to find a linear relationship between years of experience and weight estimation accuracy. However, no such relationship was found. (Figure 1 and Figure 2)

Figure 1: 5% Weight Estimation Accuracy vs Years of Experience

Figure 2: Pearson’s Correlation Coefficient, R, vs Provider Experience in Years

Discussion

This study was initiated by the authors in consideration of an in-custody death of a Colorado patient who received ketamine and subsequent statutory regulations which followed in that state. Since the introduction of ketamine into the IFD formulary, mandatory training on this medication has been in place. All paramedics must receive live classroom training on ketamine utilization prior to being cleared to administer it.

When ketamine is used, IFD protocols require continuous monitoring and documentation of the following vital signs, ideally beginning prior to administration and continuing to bedside at the receiving emergency department:

Blood pressure every 5 minutes
Continuous electrocardiographic monitoring
Continuous end-tidal capnography monitoring
Continuous pulse oximetry monitoring

It should be noted that when ketamine is administered in the setting of a patient with an acute agitated behavioral health emergency, these monitoring findings may not be available until after sedation is administered.

The results of the present study indicate that variations of drug dosing may be caused by weight estimation and further study is warranted. Considerations for further study should include variations in dosing attributable to the precision and accuracy of equipment used for measuring and administration of medications.

Limitations

The limitations of the present study include:

All subjects in this study were young, healthy males between 5’7” and 6’3” in height.
The scale used to obtain the actual weight measurements was an inexpensive home model without calibration. To mitigate this limitation, all weights were obtained with the same scale.
Each provider estimated the weights of different people. The number of estimates made by each provider were not equal. To account for this limitation, provider data was only included if at least 5 weight approximations were made by any of the subjects.
Any samples with missing information were excluded from the analysis.
The study environment was not typical of an emergency scene. It had good lighting, no distractions, and no time constraints. Many, if not most, of these conditions are not representative of a true emergency, especially one involving a critically ill or injured patient.
Years of experience may not correlate with estimation accuracy because after 2- 5 years paramedics in the IFD are frequently promoted to officer level and seldom deal with direct patient care, thereby losing practice for the skill of weight estimation.

Conclusion

In a controlled study outside the emergency setting, EMS personnel in a suburban fire department were generally accurate in estimating weights of other providers within some 20% of actual measured weight. Further study of this subject in actual patient emergencies is necessary. The authors are conducting a follow-up study to add diversity of body habitus to the subjects whose weight will be estimated.

References

1. Statewide Grand jury returns 32-count indictment against Aurora Police Officers and Aurora Fire Rescue Paramedics in the death of Elijah Mcclain. Colorado Attorney General. https://coag.gov/press-releases/9-1-21/. Published September 1, 2021.

2. Gary M Vilke, Allen Marino, Roger Fisher, Theodore C Chan, Estimation of pediatric patient weight by EMT-PS, The Journal of Emergency Medicine, Volume 21, Issue 2, 2001, Pages 125-128, ISSN 0736-4679, https://doi.org/10.1016/S0736-4679(01)00354-7.

3. Martin, D., Soria, D., Brown, C., Pepe, P., Gonzalez, E., Jastremski, M., . . . Cummins, R. (1994). Agreement Between Paramedic-Estimated Weights and Subsequent Hospital Measurements in Adults with Out-of-Hospital Cardiac Arrest. Prehospital and Disaster Medicine, 9(1), 54-56. doi:10.1017/S1049023X00040851

4. Williams, B., Boyle, M., & O’Meara, P. (2010). Can Undergraduate Paramedic and Nursing Students Accurately Estimate Patient Age and Weight? Prehospital and Disaster Medicine, 25(2), 171-177. doi:10.1017/S1049023X00007937

5. William L. Hall, Gregory L. Larkin, Mauricio J. Trujillo, Jackie L. Hinds, Kathleen A. Delaney, Errors in weight estimation in the emergency department: Comparing performance by providers and patients, The Journal of Emergency Medicine, Volume 27, Issue 3, 2004, Pages 219-224, ISSN 0736-4679, https://doi.org/10.1016/j.jemermed.2004.04.008.

6. Bradley L. Anglemyer, Chris Hernandez, Jane H. Brice, Bin Zou, The accuracy of visual estimation of body weight in the ED, The American Journal of Emergency Medicine, Volume 22, Issue 7, 2004, Pages 526-529, ISSN 0735-6757, https://doi.org/10.1016/j.ajem.2004.09.002.