Weight Estimation for Drug Dose Calculations in the Prehospital Setting – A Systematic Review

Background: Weight estimation is required to enable dose calculations for weight-based drugs administered during emergency care. The accuracy of the estimation will determine the accuracy of the administered dose. This is an important matter of patient safety. The objective of this systematic review was to collect, review, evaluate, and create a synthesis of the current literature focusing on the accuracy of weight estimation in the prehospital environment. Methods: This systematic review followed the PRISMA guidelines. Studies were identified and included if they were peer reviewed, full length, published in English, and contained original data. Studies utilizing any form of weight estimation methodology in the prehospital setting (in children or adults) were included. Data on the quality of the studies and accuracy of the weight estimation systems were extracted. Common themes were also identified. Results: Twenty-five studies met the inclusion criteria, with only nine studies (36.0%) containing useful weight estimation accuracy data. The overall quality of the studies was poor. The Broselow tape and paramedic estimates were the most studied methods of weight estimation, but there was insufficient evidence to support conclusions about accuracy. The major themes identified included the importance of accurate weight estimation and drug dosing as critical matters of patient safety, and the need for training to ensure these processes are performed accurately. Conclusions: There were limited robust data identified on the accuracy of different weight estimation methods used in the prehospital setting. Future high-quality clinical research in this area is of critical importance to ensure patient safety in the prehospital environment.


Introduction
The dosing of medications by Emergency Medical Services (EMS) paramedics is often based on weight, especially in children. 1 There is insufficient time and resources to weigh critically ill or injured patients in the prehospital setting, and having a safe, accurate, and reliable way to estimate weight is an integral step to ensuring positive outcomes. 2 With medication errors occurring in more than 28% of EMS encounters (when drugs are administered), and with weight estimation errors a significant contributor to these errors, a weight estimation technique that is rapid, accurate, reliable, and easy to use in the prehospital environment is essential. 3 This would facilitate dose calculations by EMS practitioners and would also allow for the receiving facility to prepare for the patient's arrival.
Weight estimation techniques in children have generally been well-described in the Emergency Medicine literature. They include parental estimations, visual estimations by health care providers, age-based formulas, length-based methods (such as the Broselow tape), and the newerand most accuratelength-and habitus-based methods (such as the Mercy method and the PAWPER XL tape). 4 In adults, the need for weight-based dosing (and therefore, weight estimation) is less common than in children, but it is still important. 5 Weight estimation methods in adults include self-estimations, estimations by family members, visual estimations by health care providers, anthropometric formulas, automated computerized methods, and the use of pediatric methods in adults. 6 Patient self-estimations have been shown to be the most accurate, but this might not be possible in an incapacitated patient, and a reliable method of estimation needs to be available.
There is very little known about what weight estimation methods are used in prehospital settings anywhere in the world. 7 There is less known about how accurate these methods are in the EMS environment. To the authors' knowledge, there has not been a systematic review on prehospital weight estimation systems and how accurate and reliable the various methods are when used by EMS practitioners. This could be an important source of information to determine the best strategies for estimating weight by EMS practitioners. This information would also be useful for policymakers and protocol boards.
The purpose of this study was to conduct a comprehensive review, analysis, and synthesis of the existing literature on weight estimation practices in the prehospital emergency medical care setting. The specific aims were to evaluate the quality of relevant published research, to identify the weight estimation methods used in prehospital medicine, to assess the evidence supporting the effectiveness and accuracy of the described methods, and to identify important themes arising from the studies.

Identifying Relevant Studies
This systematic review was based on the Preferred Reporting Items for Systematic Reviews and Meta-Analyses (PRISMA) methodology, with a protocol registered with PROSPERO (registration number CRD42021253761). No significant amendments of, or deviations from, the registered protocol were noted. A literature search was conducted for publications from January 1988 through September 2022 using PubMed (National Center for Biotechnology Information, National Institutes of Health; Bethesda, Maryland USA), Embase (Elsevier; Amsterdam, Netherlands), Web of Science (Clarivate Analytics; London, United Kingdom), and Google Scholar (Google Inc.; Mountain View, California USA) databases. The Boolean search terms "weight estimation" OR "weight prediction" AND "prehospital" OR "EMS" OR "Emergency Medical Services" OR "paramedic" OR "out of hospital" were used. Additional searches were done based on the "similar articles" section of PubMed and by reviewing the bibliographies (and the Medical Subject Headings [MeSH] terms) of the papers identified in the searches. To minimize reporting biases, broad inclusion criteria were used, and multiple databases were used for the search. This included searching for studies in the "grey literature."

Study Selection
Studies were screened if they were peer reviewed, full length, published in English, and contained original data ( Figure 1). Studies were considered for inclusion if they reported on weight estimation by any method, in any age group, and were related to EMS in any way. Studies on drug dosing accuracy were included if data pertaining to weight estimation were included. Studies were excluded if there was no reference to weight estimation. Screening was conducted by two researchers independently (BH and MW), and the identified articles were reviewed and screened by the other researcher.

Data Extraction
The following data were extracted from each included study: study information (publication date, number of patients, origin of study), study design, the methods of weight estimation used, the environment the study was performed in, the age range, the location, and the major findings. The data were extracted by one author (BH) and the accuracy independently confirmed by the other authors (MW, LG).

Data Analysis and Grading of Evidence
The approach used to assess the certainty of the evidence from the included studies included as assessment of the risk of bias (and study limitations), inconsistency in results (heterogeneity), indirectness of evidence, imprecision and statistical or methodological flaws, and publication bias. Each included study was graded for quality of evidence using a modified Newcastle-Ottawa scale, as has been described previously (Supplementary Table 1; available online only). 4 Each study could score a minimum of zero stars and a maximum of ten stars on the modified Newcastle-Ottawa scale. An assessment of selective non-reporting or under-reporting of results in the studies was included in the Newcastle-Ottawa scale. On this scale, a study with score from six to ten has high quality, four to five has a moderate risk of bias, and zero to three has a very high risk of bias. In addition, the Cochrane method for assessing bias in non-randomized studied was used, which assessed for selective reporting, incomplete outcome data, adequate control of confounders, blinding, appropriate comparability of cohorts, appropriate sample size, and appropriate selection. A formal assessment of heterogeneity was not conducted as there was an insufficient number of high-quality studies with data suitable for a pooled quantitative analysis.

Outcomes
The main outcomes of interest were the quality of the studies, the methods of weight estimation studied, and the accuracy of weight estimation by the studied methods (ideally percentage of estimates within 10% and 20% of actual measured weight [P10 and P20]). In addition, recurring themes arising in the included studies were identified and analyzed.

Quality of the Studies
On the Newcastle-Ottawa scale, 16/25 studies (64.0%) were at very high risk of bias, 6/25 studies (24.0%) were at moderate risk of bias, and 3/25 (12.0%) of studies were high-quality studies. The results of the Cochrane assessment of risk of bias are shown in Figure 2a and Figure 2b. The major risks of bias in the identified studies were incomplete reporting of outcome data, inadequate control of confounders, and weak study design or selection methodology.

Accuracy Outcomes
The data on the accuracy outcomes are shown in Table 2  Broselow tape (two studies), paramedic estimates (six studies), caregiver estimates (one study), unknown method of estimation (two studies), PAWPER XL tape (two studies), and the Mercy method (one study). The remainder of the studies (16/25; 64.0%) evaluated the combined effects of weight estimation and drug dose calculations on the final drug dose accuracy.

Major Themes
The major themes identified in the included articles, related to paramedic weight estimation, are shown in Table 3.

Discussion
Medication errors during prehospital emergency care of children are known to be the among the most frequent and most severe. 15,25 This error is caused by two factors: errors in estimating the weight and errors during the drug dose calculation process. However, weight estimation is not always given sufficient importance. 32,33 This is true not only in children, but in adults as well, when weightbased drugs are administered. 5,34 However, there remains a critical gap in knowledge concerning which weight estimation systems are the most accurate and appropriate for use in the prehospital setting. 35 As the safety and effectiveness of potentially life-saving drug therapy relies on accurate weight estimation, it is crucial to address this gap in understanding. The significance of this study lies in its contribution to addressing and highlighting this knowledge gap and providing valuable insights into improving patient safety and care in the prehospital environment.

Quality of the Studies
The current study highlights a significant gap in high-quality studies in EMS weight estimation. Despite its importance, the number of papers addressing this topic is limited, with only 25 identified, of which merely nine contained relevant data. This underscores the urgent need for further research in this area to improve the accuracy and reliability of EMS weight estimation, and ultimately enhance patient outcomes.
The study designs, sampling, and selection were found to be suboptimal. Many of the simulation studies relied on weight estimations from manikins or images rather than actual patients. These studies should only be regarded as preliminary reports, and the findings should not be generalized to clinical environments. Several of the clinical studies used a before-and-after methodology. This format is typically regarded as a low-quality study design, and the findings should also be regarded as hypothesis-generating rather than substantive. 36 The lack of prospective clinical studies was notable and is indicative of the poor quality of the available evidence. Compounding these issues were the small sample sizes, and a preponderance of convenience sampling, further limited the studies' power to provide meaningful conclusions.
The failure to describe potential confounders was also a major weakness. Most studies did not evaluate weight estimation accuracy in subgroups of body mass index/BMI (and age-groups in children). These two factors are known to have a significant impact on the accuracy of weight estimation methods. The failure to include this information makes it impossible to compare outcomes in different studies.
Some studies failed to specify the weight estimation methods used in the study. This dramatically limited the usefulness of the information obtained. There is limited value in studying weight estimation accuracy without this information.
Taken together, these findings highlight the need for more robust study designs, larger sample sizes, and the evaluation of relevant confounding factors in future investigations of weight estimation accuracy. Additionally, the standardization of weight estimation methods and data reporting across studies would greatly facilitate the comparison of results and the identification of best practices.

Accuracy Data
Studies from the Emergency Medicine literature have suggested that an acceptable standard for a weight estimation system is to achieve 70% of estimates within 10% of actual weight (P10 ≤ 70%) and 95% of estimates within 20% of actual weight (P20 ≤ 95%). 4,37 This is a standard which is generally achieved by the newer lengthand habitus-based weight estimation systems (such as the PAWPER XL tape and the Mercy method), and which is seldom reached by other methods, including the Broselow tape. 2,4 In this review, only the PAWPER XL tape used by paramedics achieved   this benchmark in two studies, and paramedic estimates of adults' weights came close in a single study. 4,9,26 None of these were clinical, prehospital studies, however. The Broselow tape was found to out-perform paramedic estimates by Vilke, et al. 10 Unfortunately, the acceptable error range used in this study was very large (±50%), which makes it difficult to compare these results with those from other studies. The only other useful study on the Broselow tape showed a poor accuracy in children and adolescent simulated patients. 26 From the studies that evaluated drug dosing accuracy with the Broselow tape, it is possible to infer that weight estimation was more accurate than paramedic "guesstimates" and age-based formulas, but there was no direct evidence presented in any study to support this. It is worth noting that the accuracy, and acceptability, of the Broselow tape has been questioned in the Emergency Medicine literature because of its inaccuracy in underweight and obese children. 2,4,38 The accuracy of visual estimates of weight by paramedics was poor, with the exception of a single study by Martin, et al. 9 It has been well-established in other research that visual estimates of weight are unreliable, and frequently very inaccurate. There is reasonable consensus that this method of weight estimation should not be used. 2 Interestingly, the single study evaluating caregiver estimates of weight showed that these estimates were substantially less accurate than reported in previous studies in the Emergency Medicine and pediatrics literature. 2,4,20 This was a pragmatic, real-world study in which 9-1-1 dispatchers obtained weight estimates from family members of sick children. This study is a useful warning that realworld scenarios may be very different to the typical settings used for weight estimation studies, and with substantially different results. This finding is significant, and important, because parental estimates of weight are considered to be the gold standard for weight estimates in children, but there have been few real-life studies to confirm this.
The only information on other methods of weight estimation showed that the PAWPER XL tape performed very well and the Mercy method moderately in studies by Wells, et al. 24,26 Age-based formulas performed poorly and were only reported in a single study by Hoyle, et al. 28 In this review, when compared against doctors and nurses, there was no convincing evidence that paramedics were better or worse at  estimating weight using the Broselow tape, the PAWPER XL tape, or the Mercy method. 4,26 However, one study did find that paramedics were significantly less accurate than doctors and nurses at visual estimation of weight in adult patients. 12 This evidence from a single study is weak, however. Since visual estimation of weight is generally condemned as a poor method of weight estimation, the relevance is limited in any event. 39,40 Many articles not included in this review have suggested that a particular weight estimation system would be suitable for use during prehospital care, without actually studying it in this environment (eg, Lubitz, et al and Park, et al). 41,42 It is not clear whether generalizing weight estimation performance data from a different environment, such as the ED, to the prehospital environment would be valid. In the Emergency Medicine and pediatric literature, the most accurate weight estimation methods in children are currently the PAWPER XL tape and the Mercy method. 2,4 While it is possible that these methods could be accurate in the prehospital environment, their usability (with a comprehensive drug dosing guide) during the circumstances of prehospital emergency care would need to be tested and confirmed. 43 Many of the included studies focused on the accuracy of drug dosing as the key outcome, but generally ignored the impact of inaccurate weight estimation on drug dose accuracy. Both weight estimation and the drug dose calculation and administration process need to be accurate and easy-to-use to ensure accurate drug delivery. 33,34 This concept needs to be included in future research.

Themes
Four major themes were identified from the included studies. Firstly, the importance of accurate weight-based drug dosing was recognized as crucial for safe and effective emergency drug therapy in the EMS environment. This highlights the critical need for further research in this field. This has already been identified by patient safety organizations as a vital aspect of ED care, and applies equally to the prehospital environment. 44,45 Secondly, both weight estimation and drug dose calculation were identified as key contributors to drug dosing errors, underscoring the critical importance of training to ensure these processes are conducted accurately and efficiently. This emphasizes the need for comprehensive guidance and on-going training for EMS personnel, using accurate weight estimation methods and guides to drug dose calculation. Thirdly, paramedic "guesstimates" of weight were deemed unreliable, with length-based tapes being a preferable alternative. However, these length-based tapes are not without their limitations, including incorrect usage and inaccuracies in underweight and obese children, as well as in older children. More accurate weight estimation methods should be considered. Lastly, weight estimation in adults has not been adequately studied, and there are currently no reliable methods for estimating weight in incapacitated adults in the EMS setting. This highlights a critical gap in the literature and the need for future research to develop accurate and reliable weight estimation methods for this population. Taken together, these findings emphasize the importance of accurate weight estimation and drug dose calculation in EMS, and the need for on-going research and training to improve patient outcomes and ensure optimal care in emergency situations.

Limitations
There were some important limitations in this study. Most importantly, there were limited data to draw firm conclusions about prehospital weight estimation because of the limited number of studies and the poor quality of many of the included research articles. Several studies were found in abstract form only and could not be included in this review, indicating a potential publication bias in this area of study.

Conclusion
Not enough is known about the practice of weight estimation in the EMS setting or by paramedics. That is the most important finding of this study. It was found that there was very limited published information available, and the quality of existing studies was suboptimal. Therefore, no real conclusions can be drawn regarding actual weight estimation practices, what methods are currently used, nor the performance of weight estimation systems in the EMS environment. This is a significant threat to patient safety in the EMS environment, and there is a critical and urgent need for high-quality research. Wide recognition was found in the literature on the importance of weight estimation accuracy, drug dosing accuracy, and training in these processes for EMS personnel. Future research should focus on real-world, high-quality clinical research identifying which methods of weight estimation are most accurate and easy-to-use, and which drug dosing guides are most accurate and easy-to-use in adults and children. In addition, appropriate training methods and protocols need to be developed and studied to determine how best to ensure competency by users, as well as adherence to best evidence practices in this regard. The authors did not examine the relationship between weight estimation errors and treatment failure or adverse effects of medications. Many of the children were older than 12, the effective limit for the Broselow tape. Adults Paramedics were able to estimate weight within 10% in 74% and within 20% in 93% of cases, but methods used were unknown. "Current EMS training usually does not specifically address techniques for the assessment of patient weights." Although an error of this magnitude could pose an increased risk of toxicity with pharmacologic agents commonly used in management of OHCA, this study suggests that scales or devices designed to assess patient weights are not needed on EMS vehicles as "these errors occur infrequently."

Supplementary Material
Only patients with ROSC were included. Large estimation errors and the resulting dose errors may have been more prevalent in non-ROSC patients. All methods were reported to be accurate with a mean error of 15%, but the Broselow tape was found to be the best. However, more than 10% of estimates had >50% error. "Use of the Broselow tape has helped to improve the accuracy of estimating the child's weight, but not all prehospital programs are consistently using it."

*
Only part of the study focused on weight estimation, and there was no confidence interval or variance reported. Very wide margins of acceptable accuracy were used; limited evaluation of Broselow tape, drug doses calculated from tape in only one scenario. Only 80% of crews were able to correctly use the tape, but there was a 60% error rate in drug dosing, partly due to inaccurate weight estimation (errors >10%).
In addition, 20% of crews used inappropriate methods of weight estimation (all were inaccurate).
Simulations only included weight estimation as a portion of the studyerrors were identified but not closely examined. No specific analysis of weight estimation. ***