Table 1 Studies included in the qualitative review and quantitative meta-analysis
Author and date | Study size (N) | Country | Design | Patient ages | Estimation techniques evaluated | Target | Arm | Risk of bias | Major findings; comments; major limitations |
|---|---|---|---|---|---|---|---|---|---|
Traub 1983 [69] | > 20,000 | USA | R | 0 to 18 years | Formula to estimate IBW (Traub-Kichen formula) | None | 1 | Low | Findings: Height was a good predictor of weight; IBW is only useful for a handful of drugs; TBW must be used in low weight-for-height children. Comments: Derivation study for Traub-Kichen formula. IBW predicted (actually 50th centile weight-for-length) by formula. Limitations: Incomplete presentation of data. Limited validation of formula. |
Garland 1986 [67] | 258 | USA | P | 0 to 19 years | DWEM, weight table | < 10%* | 1, 2 | Low | Findings: DWEM performed best of methods tested. Body habitus accurately assessed by evaluators. Comments: First ever report of evaluation of weight estimation systems in the literature. None of the systems tested were very accurate. Limitations: Only children up to 170 cm were included. Incomplete presentation of data. |
Lubitz 1988 [70] | 937 | USA | P | 0 to 12 years | Broselow tape | None | 1, 2 | Low | Findings: Broselow tape better than healthcare provider guesses and similar accuracy to DWEM. Accuracy of Broselow tape falls off sharply in children > 25 kg. Comments: Original study of Broselow tape. Authors recommended that an assessment of body habitus in children > 25 kg should be considered. Limitations: No formal, prospective comparison with other methodologies or indication of desired accuracy. |
Oakley 1988 [71] | – | UK | – | – | – | None | 1 | N/A | Findings: Reference chart needed to aid rapid and accurate management. Comments: Weight estimation table derived from averaged boy-girl 50th centile weight-for-height (source not mentioned). Limitations: No validation of methodology. |
Losek 1989 [72] | – | – | – | – | – | None | 1 | N/A | Findings: “Body habitus + height = accurate weight estimate”. Comments: Letter claiming superior performance of the DWEM over the Broselow tape. No original data. Limitations: No mention of desired accuracy. |
Haftel 1990 [73] | 100 | USA | P | 2 months to 15 years | Hanging-leg weight | None | 1, 2 | Low | Findings: System accurate in children > 10 kg and more so > 25 kg. Comments: Good results never evaluated in subsequent studies. Limitations: Small sample size. Incomplete presentation of data. |
Hughes 1990 [74] | 139 | UK | P | 0 to 10 years | Broselow tape, healthcare provider guesses | None | 1 | N/A | Findings: Broselow tape performed substantially better than nurses’ guesses. Comments: First validation study of Broselow tape in the UK. Limitations: Small sample size. Incomplete presentation of data. |
Greig 1997 [36] | 75 | UK | P | 0 to 12 years | healthcare provider guesses | None | 1, 2 | Low | Findings: Guesses of weight are very inaccurate; children should be weighed whenever possible. Comments: Authors suggest that accurate weight estimation is required for most drugs administered in emergency situations. Age-based formulas were wrongly considered acceptable. Limitations: Incomplete presentation of data, very small sample. |
Leffler 1997 [75] | 117 | USA | P | 0 to 5 years | Parental estimates, Leffler | < 10%† | 1, 2 | Low | Findings: Parental estimates performed much better than formula. Distraught parents may be unreliable. Comments: Small sample size. Only children < 6 years included. Limitations: Over- or underestimation not recorded. Incomplete presentation of data. |
Dearlove 1999 [76] | 50 | UK | P | 1 to 16 years | Parental estimates, Broselow tape, EPLS, Argall | < 10%† | 1, 2 | Low | Findings: Broselow tape performed best, far better than parental estimates and age-based formulas. Comments: The target of 10% accuracy chosen for children was deliberately less than the 20% that the authors considered would be appropriate for adults. Limitations: Incomplete presentation of data and small sample size. |
Goldman 1999 [77] | 233 | Israel | P | – | Parental estimates | < 10%† | 1, 2 | Low | Findings: Parents, especially mothers, can accurately estimate their children’s weights. Comments: Those parents that had weighed their children an average of 5 weeks previously had the best results. The authors defined highly accurate weight estimations as < 5% error, accurate as < 10% error and semi-accurate as < 20% error. Limitations: Incomplete presentation of data. Misinterpretation of bias as indicative of accuracy. |
Harris 1999 [78] | 100 | USA | P | 0 to 8 years | Parental estimates, healthcare provider guesses | None | 1, 2 | Low | Findings: Weight estimates by parents, nurses and doctors were significantly unreliable. The error is “so great and so frequent that clinically significant untoward effects can be anticipated”. Comments: Broselow tape recommended by authors. Limitations: Incomplete presentation of data. |
Molyneux 1999 [53] | 142 | Malawi | P | 8 months to 5 years | Blantyre tape, healthcare provider guesses | <20%† | 1, 2 | Low | Findings: Healthcare provider guesses were very inaccurate; Blantyre tape better than guesses. A 20% error considered an acceptable target. Comments: Very young study population, mostly under 5 years. Limitations: Incomplete presentation of data. |
Kun 2000 [79] | 909 | Hong Kong | P | 0 to 12 years | Broselow tape | <10%* | 1, 2 | Low | Findings: Broselow tape most accurate in children from 10 to 25 kg, but acceptable for all children. Adjustment for habitus would be advantageous. Comments: Accuracy of Broselow tape outside of the 10–25 kg range was actually poor. The accuracy in this range was reasonable, but not as good as the authors suggest. Limitations: Poor interpretation of statistics. Broselow tape version not reported. |
Carroll 2001 [80] | 169 | UK | P | – | EPLS, novel methods | None | 1 | N/A | Findings: MAC and shoe size were better indicators of weight than age. Comments: Abstract. Interesting concept, frequently cited abstract. Limitations: Incomplete presentation of data. |
Vilke 2001 [81] | 80 | USA | P | – | Broselow tape, healthcare provider guesses | <50% | 1 | N/A | Findings: 95% of estimates within acceptable error range. Comments: Unrealistic target range, with no evidence basis. Tenfold errors in drug doses in 10% of cases; Broselow tape more accurate than guesses. Limitations: Incomplete presentation of data. |
Black 2002 [52] | 495 | Australia | P | 0 to 18 years | EPLS, Broselow tape, DWEM, Oakley, TJ, TK | None | 1, 2 | Low | Findings: Broselow tape and DWEM were more accurate than formulas. These methods should be used if weighing not possible. Comments: EPLS worst performer but poor accuracy of all systems. Good reproducibility of assessment of body habitus. Limitations: Incomplete presentation of data. Broselow tape version not reported. |
Hofer 2002 [82] | 585 | Switzerland | R | 6 months to 11 years | Broselow tape | <10%* | 1, 2 | Low | Findings: Broselow tape was accurate but underestimated weight in older children. Comments: Nearly 25% of sample excluded because they were too tall for the tape. Limitations: Broselow tape not actually used and version not reported. Incomplete presentation of data. |
Uesegi 2002 [83] | 48 | Japan | P | – | Healthcare provider guesses | <20% | 1, 2 | Low | Findings: Doctors’ guesses of children’s weight were not accurate—drug doses should therefore be titrated in small paediatric patients. Comments: Wide variation in different doctors’ accuracy, not related to seniority. All estimators were very inaccurate; worst estimations occurred in children < 20 kg. Limitations: Incomplete presentation of data; conclusion that underestimation of weight may “not be a serious problem” was not supported by the evidence. |
Argall 2003 [84] | 300 | UK | P | 1 to 10 years | EPLS, Broselow tape | None | 1 | N/A | Findings: Both methods performed poorly and worsened with increasing age. Comments: Difficult to draw any conclusions from this study, but Broselow tape marginally better than formula. Authors suggest that methods of weight estimation not keeping up with increasing obesity. Limitations: Broselow tape version not reported. Incomplete presentation of data. |
Potier 2003 [85] | – | – | – | – | EPLS, Broselow tape | None | 1 | N/A | Findings: EPLS “may” be losing accuracy with increasing obesity; Broselow tape “may indeed be” more accurate. Comments: Mini-PICO analysis. Limitations: Limited qualitative-only evaluation. No comment on acceptable degree of accuracy. |
Hohenhaus 2004 [86] | – | – | – | – | Broselow tape | None | 1 | N/A | Findings: Broselow tape may cause significant weight estimation errors if used incorrectly. Broselow tape more intended for equipment size determination than for weight estimation. Limitations: Broselow tape proposed as best instrument with minimal discussion. No targets for weight estimation. |
Moore 2004 [87] | 144 | USA | P | – | Broselow tape | None | 1 | N/A | Findings: Despite using Broselow tape, only 56% correct medication doses delivered. Comments: Despite improving weight estimation, the Broselow tape did not decrease medication errors. Limitations: Incomplete presentation of data. |
Luscombe 2005 [88] | 237 | UK and New Zealand | P | 1 to 10 years | EPLS, Leffler | None | 1, 2 | Low | Findings: The Leffler formula was more accurate than the EPLS formula. Comments: A new formula was promoted but the evidence was limited: only bias was evaluated, not accuracy. Both formulas underestimated weight significantly The Luscombe formula, as known today, was the result of a later study. Limitations: Incomplete presentation of data. |
Theron 2005 [89] | 909 | New Zealand | P | 1 to 10 years | EPLS, Shann, Leffler, Oakley, Broselow tape | None | 1, 2 | Low | Findings: Formulas and Broselow tape underestimated the weight of Pacific Island and Maori children. Comments: Broselow tape was the best performer of the systems tested, despite the fact that the authors reported the contrary. Limitations: Broselow tape version not reported. Incomplete presentation of data. |
Bavdekar 2006 [90] | 500 | India | P | 0 to 2 years | Novel formula based on foot length | None | 1 | N/A | Findings: Foot length can be used for emergency drug calculation Comments: Only 10% of population > 1 year old. very poor statistics. Method of data analysis made findings unreliable and uninterpretable - bias confused with accuracy. |
Nieman 2006 [37] | 7813 | USA | R | 0 to 12 years | Broselow tape | < 10%* | 1, 2 | Low | Findings: One third of children had inaccurate weight estimations. The 1998 tape performed better than the 2002 tape. A measurement of obesity (e.g. MAC) should be added to Broselow tape to increase accuracy. Comments: Broselow tape version 1998 and 2002A used. The study population BMI was 16.8. Good statistics. Limitations: Broselow tape not actually used. Nearly 7% of sample excluded because too tall for the tape. |
Varghese 2006 [91] | 500 | India | P | 1 to 12 years | Argall, EPLS, Nelson, Broselow tape | None | 1 | N/A | Findings: Formulas overestimated weight in this developing-world study. The Broselow tape was the most accurate. Comments: More than half the study population was under 6 months of age and only 8% > 5 years. Limitations: Incomplete presentation of data. |
Anderson 2007 [92] | – | – | – | – | – | None | 1 | N/A | Comments: Letter. The author comments on benefits of under- vs. overestimation of weight. No quantification of permissible or safe error was suggested. |
Dieckman 2007 [93] | – | – | – | – | – | None | 1 | N/A | Comments: Editorial on weight estimation and drug dosing. LBW might be better than TBW. No evidence offered for opinion; no weight estimation target suggested (for TBW or LBW). |
Du Bois 2007 [94] | 400 | USA | P | 0 to 19 years | Broselow tape, DWEM | MPE < 5% | 1, 2 | Low | Findings: DWEM better than Broselow tape, but both systems underestimated weight especially in children > 20 kg. Comments: Good inter-rater reliability for habitus assessment. Performance of DWEM much better than Broselow tape, but both systems performed poorly. Limitations: Incomplete presentation of data. Broselow tape version not reported. Statistically inappropriate accuracy target. |
Hashikawa 2007 [32] | 1207 | USA | R | 0 to 12 years | Broselow tape | < 20%† | 1 | N/A | Findings: Approximately 60% accuracy of colour zones assignment (accurate drug dosing). Weight underestimated in obese and older children. Comments: Rising prevalence of obesity blamed for poor performance. Broselow tape version 2002B. The average study population BMI was 17. Limitations: Broselow tape not actually used. Assessment of correct zone assignment only, weight not measured. Incomplete presentation of data. |
Im 2007 [95] | 454 | Korea | P | – | Broselow tape | None | 1 | High | Findings: Broselow tape accurate in children of normal weight-for-length Comments: Only children falling within 3rd to 97th weight-for-height centiles included. Very young study population. Broselow tape only recommended by authors for “normal-growth” children < 20 kg and < 120 cm. Limitations: Incomplete presentation of data. Conclusions not supported by findings. |
Jang 2007 [96] | 665 | Korea | R | – | Broselow tape | < 10% | 1, 2 | Low | Findings: Broselow tape reasonably accurate in this population, but less so in children > 25 kg. Comments: Overall underestimation of weight. Performance not very good and on par with most other studies. Limitations: Broselow tape not actually used and version not reported. Incomplete presentation of data. |
Kelly 2007 [97] | 410 | Australia | P | 1 to 11 years | BG | < 20% | 1, 2 | Low | Findings: BG performed moderately well, but overestimated weight in low BMI children. Comments: Multiple papers on same data. BMI was 17 in study population. Significant number of children had large errors of weight estimation. Limitations: Incomplete presentation of data. |
Krieser 2007 [98] | 410 | Australia | P | 1 to 10 years | Parental estimates, Broselow tape, BG, Argall, EPLS | < 10%* | 1, 2 | Low | Findings: Parental estimates performed best, followed by Broselow tape. Only 11% of parents could not provide an estimate. Formulas performed much worse than other methods. Comments: Multiple papers on same data. Study population BMI was 17.1. Limitations: Broselow tape not actually used. Broselow tape version not reported. Incomplete presentation of data. |
Luscombe 2007 [1] | 13,988 | UK | R | 1 to 10 years | EPLS, Luscombe | None | 1, 2 | Low | Findings: The authors commented that since few children with high-acuity conditions are actually weighed in clinical practice, weight estimation essential. The EPLS formula significantly underestimated weight, which may lead to under-resuscitation. The Luscombe formula was more accurate. Comments: Both formulas actually performed poorly. Limitations: Incomplete presentation of data. Mean bias used incorrectly. |
Luten 2007 [99] | – | – | – | – | Broselow tape | None | 1 | N/A | Findings: No substantiation for setting acceptable accuracy of weight estimation at 10%. Comments: Editorial comment; no evidence provided. |
Nguyen 2007 [100] | 410 | Australia | P | 1 to 11 years | Argall | < 10%* | 1, 2 | Low | Findings: Argall formula performed poorly, especially in children > 35 kg. Comments: Study population BMI was 17. Limitations: Incomplete presentation of data. |
Patel 2007 [101] | 360 | USA | P | 1 month to 10 years | Parental estimates, EPLS, Broselow tape | None | 1 | N/A | Findings: Parental estimates performed best, then Broselow tape, then EPLS. Comments: Abstract. Conclusions based on correlation and rudimentary statistical analysis. Limitations: Incomplete presentation of data. |
Pollock 2007 [48] | 100 | Malawi | P | 1 to 7 years | EPLS, Luscombe | None | 1, 2 | Low | Findings: The Luscombe formula was less accurate than EPLS with greater overestimation of weight. The authors suggested length-based systems should rather be used. Comments: Scientific letter. Both formulas performed very poorly with significant overestimation of weight. Limitations: Incomplete presentation of data. |
Thompson 2007 [34] | 1843 | Australia | R | 0 to 14 years | BG, EPLS, ARC | None | 1, 2 | Low | Findings: The BG formula performed better than EPLS and ARC formulas. Authors advised cautious use in infants. Comments: Data obtained from high acuity patients. The BG did not actually perform that well in this study. Limitations: Incomplete presentation of data. Only measures of bias used for comparison. |
Tinning 2007 [102] | 67,363 | Australia | R | 0 to 14 years | BG development study | None | 1 | N/A | Findings: New formulas developed with no target accuracy and no validation sample. Comments: Limited indication of accuracy of formulas as only measures of bias reported. Limitations: Incomplete presentation of data. |
Zikos 2007 [103] | Comments: Identical abstract and data to Patel 2007 | ||||||||
Gardner 2008 [104] | not reported | UK | P | 1 to 15 years | EPLS, Luscombe, healthcare provider guesses | None | 1 | N/A | Findings: Age-based formulas often used as basis for weight guess, but the formulas were better than guesses. Comments: Abstract. Limitations: Incomplete presentation of data. |
Luten 2008 [3] | – | – | – | – | – | None | 1 | N/A | Comments: This was a commentary on Broselow tape—that IBW might be better than TBW as a goal; “the Broselow tape is a tool that was not designed to be used without clinical judgement”. No evidence offered to support opinion; no goal target suggested (for TBW or IBW). |
Ramarajan 2008 [56] | 548 | India | P | 0 to 12 years | Broselow tape | < 10%† | 1, 2 | Low | Findings: Broselow tape overestimated weight by > 10% in Indian children over 10 kg. A correction factor was developed, but not validated. Comments: One of the few studies to show overestimation of weight by Broselow tape. Limitations: Broselow tape version not reported. Incomplete presentation of data. |
Wells 2008 [105] | – | – | – | – | – | None | 1 | N/A | Comments: Letter. The Broselow tape has not been shown to accurately predict IBW. TBW should be used for dose calculation in emergencies. |
Zink 2008 [106] | 127 | USA | P | 0 to 17 years | Parental estimates, healthcare provider guesses, DWEM, Broselow tape | None | 1 | High | Findings: Broselow tape and DWEM were the least accurate methods. Comments: No conclusions can be drawn from this study because of the methodology. Limitations: The data appears to favour healthcare provider and parent guesses, but the statistical methodology is flawed. |
Anstett 2009 [107] | 545 | Ireland | R | – | Broselow tape | None | 1, 2 | Low | Findings: The Broselow tape was often inaccurate and tended to underestimate weight. Comments: Abstract. The Broselow tape actually performed better in this study than in many other studies. Limitations: Broselow tape version not reported. Broselow tape not actually used. |
Cattamanchi 2009 [108] | 15,000 | India | P | 2 months to 12 years | Broselow tape | < 10%* | 1, 2 | Low | Findings: The Broselow tape performed well, especially in children < 10 kg but underestimated all others, especially in children > 18 kg. Comments: Abstract. Very large prospective study. The authors recommended a new version of Broselow tape for Indian children because of underestimation of weight. Limitations: Broselow tape version not reported. Incomplete presentation of data. |
Cattermole 2009 [109] | 1368 | Hong Kong | P | 1 to 12 years | MAC, Broselow tape, foot length | None | 1 | N/A | Findings: Estimates of weight can be based on MAC. A special colour-coded MAC tape could be produced to aid drug dosing. The authors recommended habitus modified use of Broselow tape. Comments: Abstract. No data presented. Broselow tape performed better in younger children, MAC better in older children. Limitations: No data presentation. |
Partridge 2009 [110] | 777 | USA | P | 0 to 20 years | Parental estimates, healthcare provider guesses | < 10%† | 1, 2 | Low | Findings: Parents were better than nurses at estimating weight; nurses were very inaccurate. Comments: Guessed weights most often underestimations. The longer the time from last weighing, the greater the error. All nurses, regardless of training and experience, were poor estimators. Limitations: Incomplete presentation of data. |
Paw 2009 [111] | 791 | UK | P | 1 to 12 years | EPLS | None | 1 | High | Findings: Very poor performance of EPLS formula. The authors recommended the Broselow tape or an alternative formula. Comments: Abstract. Uniformly abysmal accuracy across different ethnic groups. Limitations: Incomplete presentation of data. |
Sandell 2009 [112] | 846 | UK | P | 1 to 11 years | EPLS, age-based estimates vs. length-based estimates | None | 1, 2 | Low | Findings: Length-based and age-based systems are suitable in emergencies, but length-based were better; new formulas more accurate than EPLS; “one size fits all” approach not likely to be successful. Comments: Unique method of analysing data—does not allow comparisons with other studies in this format. Age-based methods less accurate than suggested; biological variability less in length-based than age-based systems. Limitations: Incomplete presentation of data. |
So 2009 [113] | 1011 | USA | P | 0 to 10 years | Broselow tape, Leffler, Theron | None | 1, 2 | Low | Findings: Broselow tape most accurate weight estimation method with best accuracy in normal BMI children. Comments: Study population BMI was 17.8. New formula developed, but not tested. Limitations: Incomplete presentation of data. Bias mistaken for accuracy. Broselow tape not actually used and Broselow tape version not reported. |
Stewart 2009 [30] | 475 | Australia | P | 0 to 10 years | Broselow tape | < 20%† | 1, 2 | Low | Findings: Best performance of Broselow tape between 10 and 25 kg but very inaccurate in children >25 kg. More accurate than age-based formulas, however. Comments: Masters dissertation. Good statistics. Limitations: Broselow tape version not reported. |
Yamamoto 2009 [12] | 542 | Hawaii | P | – | Broselow tape, novel habitus-specific equation | None | 1 | High | Findings: Adding body habitus data to length increases accuracy of weight estimation. Comments: Useful data unable to be imputed from poorly presented statistics; overweight children were preferentially selected in this study Limitations: Incomplete presentation of data. Broselow tape version not reported and Broselow tape not actually used. |
Bicer 2010 [114] | 2319 | Turkey | P | 3 to 18 years | EPLS, Argall, novel formulas | None | 1 | High | Findings: Argall formula performed best. Comments: Article in Turkish; many formulas studied, often applied outside of the age-range for which they were intended. Limitations: Incomplete presentation of data. Measures of bias mistaken for measures of accuracy. |
Casey 2010 [115] | 1426 | Australia | P | 0 to 14 years | EPLS, BG, Broselow tape | < 20%† | 1, 2 | Low | Findings: BG was accurate in children 1–4 years of age. The EPLS formula was the least accurate of all methods. The authors recommend that the accuracy and ease-of-use of the Broselow tape mandates its use in the ED and pre-hospital. Comments: The authors’ suggestions that the BG was more “accurate” than the Broselow tape are misleading: the Broselow tape actually significantly outperformed the BG formula. BMI was 17.9 in study population. Limitations: Incomplete presentation of data. Broselow tape not actually used and Broselow tape version not reported. |
Cattermole 2010 [116] | 1370 | Hong Kong | P | 0 to 11 years | EPLS, Broselow tape, MAC formula | 20–30% | 1, 2 | Low | Findings: MAC formula outperformed Broselow tape in children > 5 years. Comments: MAC actually only more accurate in children from 9 to 11 years who were too tall for the Broselow tape. Poor accuracy of all systems demonstrated. Limitations: Broselow tape not actually used, and 1998 edition values used. No evidence provided for desired accuracy targets. |
Lulic 2010 [117] | 209 | Croatia | P | 0 to 14 years | EPLS, Argall, BG, Luscombe, Broselow tape | < 10%† | 1, 2 | Low | Findings: Broselow tape was more accurate than age-based formulas but tended to slightly underestimate weight. Comments: Abstract; none of the systems estimated weight with an acceptable degree of accuracy. Limitations: Broselow tape version not reported. Incomplete presentation of data. |
Rosenberg 2010 [118] | 401 | USA | P | 0 to 14 years | healthcare provider guesses | None | 1 | Low | Findings: Large errors in adrenaline doses were possible, based on doctors’ guesses of weight. Comments: Abstract. Poor weight estimates translated into potentially harmful medication errors. Limitations: Incomplete presentation of data. |
Williams 2010 [35] | 468 | Australia | P | 4 and 6 years | healthcare provider guesses | None | 1, 2 | Low | Findings: Both age and weight were estimated poorly by paramedics, with no association with experience or training. Comments: Estimation of weight and age from images of two children by 234 estimators. Limitations: Incomplete presentation of data. |
Bourdeau 2011 [119] | 243 | Canada | R | 1 to 10 years | Broselow tape | < 10%† | 1, 2 | Low | Findings: Broselow tape was not accurate in First Nations’ children and may need to increase weight estimates by 12%. Although ideal method of dosing is still unknown, it is best practice to eliminate any possible contributing inaccuracies. Comments: First Nations’ children selected with high incidence of obesity in this population. Broselow tape did not perform equally well in all ethnic/population groups. Limitations: Incorrect use of some statistics. Broselow tape not actually used and Broselow tape version not reported. |
Cattermole 2011 [45] | 1248 | Hong Kong | P | 1 to 10 years | EPLS, ARC, Nelson, Shann, Leffler, BG, Argall, Luscombe, Theron, CAWR-1, CAWR-2 | < 30% | 1, 2 | Low | Findings: CAWR-1 should be used in younger Chinese children, but other methods of weight estimation should be used in older children as age formulas are inaccurate. Comments: Comprehensive statistics. Findings suggested that no age formula performed well at any age. Limitations: No evidence to support suggested target of weight estimation. |
Cattermole 2011 [120] | – | – | – | – | – | < 40% | 1 | N/A | Comments: Letter. “Age-based paediatric weight estimation is imprecise”. The author reflects that an error of ± 40% might not be considered acceptable by most clinicians, but no alternative suggested. |
Costelloe 2011 [121] | 62 | UK | P | 6 months to 6 years | Parental estimates, parent vs. nurse measurement | < 2 kg | 1 | High | Findings: Parents can estimate weight accurately based on home measurements, but very few actually weigh children at home. Comments: Authors used doses of ibuprofen and paracetamol as endpoints: the applicability in emergency situations is uncertain. Limitations: Incomplete presentation of data. Very small sample size. |
Geduld 2011 [122] | 2832 | South Africa | P | 0 to 10 years | EPLS, Luscombe, BG, Broselow tape | < 10%* | 1, 2 | Low | Findings: Broselow tape and EPLS formula most accurate in this population. Careful titration of drugs and use of clinical judgement most important in using medications safely. Comments: Data from a poor community in South Africa. The accuracy of EPLS formula was the best ever reported while the accuracy of Broselow tape was on par with other reports. Only 4% of children excluded as too tall for the tape. The authors question whether the differences in accuracy of any weight estimation system are likely to affect outcomes. Limitations: Broselow tape not actually used and Broselow tape version not reported. |
Huybrechts 2011 [123] | 275 | Belgium | P | 3 to 7 years | Parental estimates, parent vs. nurse measurement | None | 1 | Low | Findings: Parental estimates were most accurate when based on measurements made at home, rather than onguesses. Comments: Application to emergency weight estimations is uncertain as parental stress may negate this effect. Limitations: Targets were estimations of overweight or underweight. Incomplete presentation of data. |
Kelly 2011 [124] | 410 | Australia | P | 1 to 10 years | Luscombe, EPLS, Argall, BG | < 10%* | 1, 2 | Low | Findings: The Luscombe formula performed best of all the formulas. Comments: Study population BMI was 17. Same population as Nguyen 2007 and several other publications; all formulas performed poorly. Limitations: Incomplete presentation of data. |
Knight 2011 [125] | 657 | USA | R | – | Broselow tape | None | 1, 2 | Low | Findings: Broselow tape performed poorly, potentially leading to under-resuscitation in all weight categories, especially in younger children. Drug doses correct in only about 50% of cases. Consensus opinion required whether to use IBW or TBW during resuscitation. Comments: Broselow tape 2007B. High incidence of obesity in study population. Limitations: Broselow tape not actually used. Incomplete presentation of data. No direct assessment of accuracy of weight estimation. |
Luscombe 2011 [126] | 64,197 | UK | R | 1 to 16 years | EPLS, Luscombe | None | 1, 2 | Low | Findings: The Luscombe formula outperformed the EPLS formula. Weight estimation is of paramount importance in resuscitation, therefore remembering one formula better than several. Comments: While the bias of the Luscombe formula was smaller, both formulas performed poorly. Limitations: Incomplete presentation of data. An inappropriate age range was used for formulas (up to 16 years). |
Marlow 2011 [27] | 140,314 | UK | R | 0 to 16 years | EPLS, Argall, Luscombe, BG | < 10%* | 1, 2 | Low | Findings: The EPLS formula was least accurate of commonly used formulas. The BG and Luscombe formulas were very similar and the best performers. No formulas showed acceptable accuracy, however. Comments: Abstract, with additional data supplied by author. This was a very large retrospective database study with good descriptive statistics. Limitations: Some incomplete data. |
Rosenberg 2011 [38] | 372 | USA | P | 0 to 14 years | healthcare provider guesses, Broselow tape | < 10%* | 1, 2 | Low | Findings: The Broselow tape was better than guesses by doctors, but not in obese children. Poor assessment of habitus by doctors. IBW suggested as the best target for estimation in obese kids. Comments: Broselow tape 2007B. 35% of study population obese or overweight. Mean BMI was 17.4. It is a reflection of how poorly the Broselow tape performed in obese children that doctor estimates were better; overall findings of Broselow tape accuracy similar to other studies. Limitations: Incomplete data presentation. Incorrect use of some statistics. |
Wells 2011 [127] | – | – | – | – | – | < 10%* | 1 | N/A | Findings: More than 50% of study population required habitus-modified weight estimation. Length-based systems should be used for weight estimation in emergencies. Comments: Editorial comment. Insufficient evidence exists to choose between IBW and TBW for drug dose calculations. Limitations: No reference standard for weight estimation included. |
Abdel-Rahman 2012 [15] | 1938 | USA | R | 0 to 16 years | EPLS, ARC, Argall, BG, Broselow tape, Cattermole, Leffler, Luscombe, Nelson, Shann, TJ, TK, Mercy method | None | 1, 2 | Low | Findings: The Mercy method performed significantly better than the age-based, length-based and MAC-based systems. Comments: This was the original development and validation study of the Mercy method. Good mix of overweight children, but few underweight kids in sample. Broselow tape unable to be used in more than one third of the sample (too tall for the tape). Limitations: Broselow tape not actually used and Broselow tape version not reported. |
Ali 2012 [128] | 1723 | Trinidad | R | 1 to 5 years | EPLS, Luscombe, new formula | < 10%† | 1, 2 | Low | Findings: All formulas performed similarly and all poorly, even the new formula derived from the study population. Comments: Only children aged 1 to 5 are included in study. Limitations: No validation sample for derived formula. Incomplete presentation of data. |
Cantle 2012 [129] | 2253 | USA | R | 1 to 10 years | EPLS, new APLS, CAWR-1, CAWR-2, MAC | None | 1, 2 | Low | Findings: CAWR-2 performed as well as other formulas. Age formulas performed less well in older children. MAC performed well in older children but poorly in younger children. Comments: Conference presentation. Study of CAWR in a Western population. No age formula actually performed well in any age group. Limitations: Some errors in presented data and calculations. |
Elgie 2012 [130] | 188 | UK | P | 10 weeks to 10 years | APLS, EPLS, Luscombe formulas based on clothing label age | None | 1 | High | Findings: Using the age on clothing label was more accurate than using actual age. Comments: This clever study used clothing size as a surrogate marker for habitus. The Luscombe formula performed slightly better than the EPLS formula. Limitations: Incomplete presentation of data. |
Garwood 2012 [131] | 1252 | UK | R | 1 to 16 years | EPLS, new formula | None | 1 | High | Findings: The Garwood formula was better than the EPLS formula. Comments: Abstract. Authors used assessment of mean bias only. Limitations: Limited statistical analysis made findings difficult to interpret. No validation of new formula. |
Heyming 2012 [132] | 491 | USA | P | IQR 10 to 49 months | Broselow tape EMS vs. Broselow tape ED | None | 1 | High | Findings: The Broselow tape was an accurate tool to estimate weight. Comments: Broselow tape agreement between pre-hospital personnel and ED personnel only 70.1%. Limitations: Most of study population < 4 years. Incomplete presentation of ldata. Broselow tape version not reported. |
Kelly 2012 [26] | – | – | – | – | – | None | 1 | N/A | Findings: Best systems to be used, in order of accuracy: parental estimates, Broselow tape, age-based formulas. Comments: Book chapter. Limitations: No mention of acceptable targets for weight estimation or weaknesses of age formulas. Subjective assessment of studies only. |
Meguerdician 2012 [2] | – | – | – | – | – | None | 1 | N/A | Findings: The Broselow tape is the most consistent and reliable tool for weight estimation, but habitus-based weight adjustment for Broselow tape is logical. Parental estimates may rival the Broselow tape, but parents may be absent or uncertain, especially under stress. Underdosing might be prudent in emergencies. Comments: The author acknowledges that the degree of acceptable error is difficult to define. Limitations: No evidence provided of efficacy of Broselow tape in reducing medication errors. No suggestion of goal target accuracy required. Some controversial opinions about drug dosing. |
Milne 2012 [33] | 6361 | Canada | R | IQR 1.6 to 7.7 years | Broselow tape | < 30% | 1, 2 | Low | Findings: Broselow tape effective, although significantly underestimated weight. Similar findings in urban and rural children. Comments: Broselow tape edition 2002A. The authors suggested that an error range of 30% is reasonable and that, ideally, technology should be developed to estimate weight. Limitations: Broselow tape not actually used—data taken from anthropometric measurements. |
Park 2012 [28] | 124,095 | Korea | R | 0 to 14 years | EPLS, Shann, Leffler, Nelson, BG, Broselow tape, Park | None | 1, 2 | Low | Findings: All methods underestimated weight, possibly because of a secular trend towards increasing BMI. The Park formula performed best and was accurate in children < 1 year of age. Comments: All methods assessed showed poor accuracy. Limitations: Incomplete presentation of data. Poor interpretation of bias vs. accuracy. Broselow tape was not actually used and Broselow tape version not reported. |
Seddon 2012 [133] | 599 | UK | P | 1 month to 12 years | EPLS, new APLS, Argall, BG, Luscombe | None | 1, 2 | Low | Findings: Mixed racial study population, no major racial differences found in performance of formulas. There was increased underestimation of weight with increasing age. The Luscombe formula performed better than EPLS and APLS in children 1 to 10 years of age; BG and APLS better outside these ages. Limitations: Limited statistical analysis and incomplete data—only measures of bias reported. Some data values were not credible. |
Sinha 2012 [134] | 118 | USA | P | 0 to 14 years | Broselow tape | None | 1, 2 | Low | Findings: It was possible to weigh children during trauma resuscitation. Comments: Broselow tape weight was compared to stretcher weight. Nearly 40% of children in the study could not be weighed—more often the sicker kids (not fully explained). There was no validation of the accuracy of weight measured during resuscitation. Poor performance by Broselow tape with 18% of children too tall for the tape. Limitations: Incomplete presentation of data. Broselow tape version not reported. |
Trakulsrichai 2012 [135] | 595 | Thailand | P | 0 to 12 years | Broselow tape, parental estimates, growth charts | < 10%† | 1, 2 | Low | Findings: Family member estimation was most accurate, and the Broselow tape the most accurate of other weight estimation methods. Comments: Equal underestimation and overestimation by Broselow tape while family estimates tended to overestimate. Limitations: Incomplete presentation of data. Broselow tape version not reported. |
Winship 2012 [42] | – | – | – | – | – | None | 1 | N/A | Findings: As average weight of children has increased so the accuracy of formulas has decreased; age-based methods will be unable to adjust to deal with future rises in average weights of children; no age-based weight formula accurate—might as well continue with EPLS formula. Comments: Semi-systematic review. Limitations: No suggestion of goal target accuracy required. |
Wozniak 2012 [136] | 777 | Botswana | P | 18 months to 12 years | EPLS, Luscombe, Theron, Cattermole, Broselow tape | < 10%† | 1, 2 | Low | Findings: Prediction models incorporating MAC and either tibia or ulna length performed extremely well. Age-based formulas were very inaccurate. Comments: Masters dissertation. Weight was markedly overestimated by formulas in this population with a high prevalence of HIV. Limitations: Incomplete statistics. |
Abdel-Rahman 2013 [18] | 624 | USA | P | 0 to 16 years | Mercy method two-dimensional, 3D, Broselow tape | None | 1, 2 | Low | Findings: Both two-dimensional and 3D Mercy methods outperformed the Broselow tape. Comments: Study population BMI was 17.3. Children much older than most weight-estimation studies included. Multicentre study—3 in the USA. One third of children were too tall for the tape. The 3D tape was significantly less accurate than the two-dimensional method. Limitations: Broselow tape not actually used and Broselow tape version not reported. |
Abdel-Rahman 2013 [19] | 976 | USA | P | 0 to 16 years | Mercy method, EPLS, Broselow tape, Luscombe, Nelson | < 20% | 1, 2 | Low | Findings: The Mercy method was accurate across a wider age range than other methods. Comments: Study population BMI was 17.6 with very few underweight children. Inter-rater assessment was generally reasonable, but two of the raters were inferior to the others. Limitations: Broselow tape not actually used and Broselow tape version not reported. |
Akabarian 2013 [137] | 403 | Iran | P | 0 to 14 years | Broselow tape, parental estimates | < 10%* | 1, 2 | Low | Findings: The Broselow tape was more accurate than parental estimates. Comments: Article in Arabic. Very good performance of Broselow tape compared to previous studies. Limitations: Exclusion criterion of weight > 35 kg limited the assessment of Broselow tape accuracy. Limited statistical analysis and data presentation. Broselow tape version not reported. |
Cattermole 2013 [138] | 171 | Hong Kong | P | 7 to 11 years | Broselow tape | None | 1 | High | Findings: Children “too tall for the tape” do not have full adult weight—this assumption would lead to an average 30% overestimation of weight in this study. About 40% of children were too tall for the tape. Comments: Selective evaluation of children who were too tall for Broselow tape. Interesting data which highlights a flaw in the Broselow tape methodology. Broselow tape was not practically useful over the age of 10. An overestimation of 30% would be unacceptable; no mention of desirable target. Limitations: Relatively small sample size. |
Graves 2013 [139] | 37,114 | USA | R | 0 to 14 years | EPLS, new APLS, BG, Luscombe, Broselow tape | < 10%† | 1, 2 | Low | Findings: New APLS formula was best for infants, BG best for older children. Broselow tape wrong zone in up to 60% of children. Comments: Broselow tape 2007B. Despite criticisms of Broselow tape, it outperformed the formulas in every analysis. This study had some of the poorest performances of aged-based formulas in any study. A post hoc study population BMI was calculated to be 18 to 22. Limitations: Incomplete presentation of data. Broselow tape not actually used. |
Hegazy 2013 [140] | 508 | Egypt | P | 1 to 16 years | EPLS, Shann, Garwood formula | < 10%† | 1, 2 | Low | Findings: Garwood formula performed best, especially in older children. Comments: Sample population of cancer patients. Very poor performance of all formulas tested—none were close to acceptable accuracy. Limitations: Formulas used for children older than intended. Poor interpretation of findings. |
House 2013 [141] | 967 | Kenya | P | 0 to 14 years | Broselow tape, EPLS, Nelson | MPE < 10% | 1, 2 | Low | Findings: Broselow tape performed better than formulas and a measure of habitus assessment (e.g. MAC) was suggested. Broselow tape should be used rather than formulas. Comments: Broselow tape 2007B. Underestimation of weight predominated. Limitations: Incomplete statistical analysis and data presentation. Poor interpretation of results from previous studies; flawed outcome measures used (indicators of bias only). |
Lim 2013 [39] | 199 | USA | R | 0 to 17 years | healthcare provider guesses | < 20%† | 1, 2 | Low | Findings: The authors suggested that EMS personnel were generally accurate in estimating weights of children. Comments: The more severe the condition, the worse the weight estimation. Weight estimation appeared better than it actually was. Limitations: Incomplete presentation of data. |
Loo 2013 [142] | 875 | Singapore | P | 1 to 10 years | EPLS, Luscombe, Broselow tape | < 10%† | 1, 2 | Low | Findings: Broselow tape more accurate than formulas. EPLS underestimated, Luscombe overestimated weight, but EPLS performed better overall than Luscombe. Comments: Broselow tape 2007B. Mean BMI of study population was 15.9. Good statistics. Limitations: Broselow tape not actually used. |
Suh 2013 [143] | 105,072 | Korea | R | – | Broselow tape 2005A, 2007B, 2011A | < 10%* | 1, 2 | Low | Findings: Broselow tape 2011A more accurate than older versions. Comments: Abstract. Large database sample. Limitations: Incomplete presentation of data. Broselow tape not actually used. |
Wells 2013 [5] | 453 | South Africa | P | 0 to 12 years | Broselow tape, PAWPER tape | < 10%† | 1, 2 | Low | Findings: PT performed better than Broselow tape in every category analysed and better than any previously published system. Comments: Broselow tape 2007B. Population with mixed under- and overweight. Multi-centre study of habitus-modified length-based weight estimation. Limitations: Assessment of body habitus based on visual estimation. |
Young 2013 [44] | 207 | USA | P | 1 to 9 years | EPLS, Broselow tape, parental estimates, Luscombe, finger counting | < 10%* | 1, 2 | Low | Findings: Finger counting system as accurate as Broselow tape and more accurate than other formulas. Conceptually a simple system. To increase the accuracy weight-estimation systems may cause increased complexity and stress during resuscitations. Comments: The finger counting method is equivalent to formula Wt = 2.5 × age(years) + 7.5. Median BMI of study population was 17.2. Limitations: Incomplete presentation of data. Broselow tape not actually used and version not reported. |
Abdel-Rahman 2014 [4] | – | – | – | – | – | None | 1 | N/A | Findings: Some experts have suggested that weight estimation cannot be accurate. This is likely to be to the disadvantage of children as dual length-and habitus-based can achieve acceptable accuracy. Comments: Brief review of weight estimation systems. Limitations: No target of weight estimation accuracy suggested; Mercy method recommended for environments where no scale available—no mention of emergency use. |
Ackwerh 2014 [144] | 10,488 | USA | R | 2 to 12 years | EPLS, Luscombe, Theron | < 10%* | 1 | High | Findings: Unreliable. Comments: Unclear statistical analysis. The reported results for age-based formulas were the best ever reported, but are completely incorrect. The graphically presented data contradict the other findings. Limitations: Incomplete presentation of data. |
Allison 2014 [54] | 2102 | Australia | R | 0 to 5 years | EPLS, BG, Luscombe, Argall, Nelson, Broselow tape, Sandell tape | None | 1, 2 | Low | Findings: Broselow tape was most accurate in this study. Comments: Broselow tape 2007 edition B. Aboriginal and Torres Strait Island children included (low weight-for-length). Only published study of Sandell tape. Limitations: Broselow tape not actually used. Incomplete presentation of data. Narrow age range evaluated. Very fat and very thin children excluded from study. |
Batmanabane 2014 [49] | 374 | India | P | 0 to 16 years | EPLS, ARC, Argall, BG, Broselow tape, Cattermole, Leffler, Luscombe, Nelson, Shann, TJ, TK, Mercy method | None | 1, 2 | Low | Findings: Mercy method performed well in Indian children, similar to that shown in Western populations. Comments: Good statistics. Overestimation of weight by all methods except Mercy. Limitations: Broselow tape not actually used and Broselow tape version not reported. |
Chiengkriwate 2014 [145] | 3869 | Thailand | R | 0 to 15 years | Broselow tape | < 10%* | 1, 2 | Low | Findings: Broselow tape underestimated weight in Thai children, more so in older children, similar to findings in Western populations. Comments: Broselow tape 2007 edition A. Broselow tape performance consistently at a PW10 of just below 60%. Limitations: Broselow tape not actually used. |
Dicko 2014 [50] | 473 | Mali | P | 0 to 16 years | Mercy, EPLS, ARC, Broselow tape, Nelson | None | 1, 2 | Low | Findings: Mercy method performed extremely well in this population in Mali, similar to its performance elsewhere in the world. Other methods overestimated weight. Comments: BMI of study population was 15.6, with 22% underweight and 1.7% overweight or obese. Good inter-rater reliability. Limitations: Broselow tape not actually used and version not reported. |
Eke 2014 [146] | 370 | Nigeria | P | 1 to 12 years | APLS | None | 1 | High | Findings: The APLS formula underestimated weight in these Nigerian children. Comments: The findings are unreliable in view of the major methodological flaws. Limitations: Limited and incomplete data presentation. |
Erker 2014 [147] | N/A | WHO/CDC reference centiles | R | 1 to 12 years | EPLS, APLS, Luscombe, BG, Park, Shann, Nelson, novel habitus-modified formulas | None | 1 | High | Findings: Potentially enhanced accuracy of age-based equations by using a habitus-specific formula, if length-based methods not available. Comments: Three formulas for “thick”, “normal” and “thin” children. Theoretical accuracy in derivation study. Limitations: No original data or validation of formulas. Incomplete presentation of data. |
Flannigan 2014 [46] | 10,081 | UK | R | 0 to 15 years | New APLS, Luscombe, novel formula | None | 1, 2 | Low | Findings: New APLS formula very inaccurate and should not be used. Weight should be adjusted according to a data table with the 5th and 95th weight-for-height values. Comments: Study performed in ICU patients. Limitations: Incomplete presentation of data. New regression formula untested, even with an internal validation sub-sample. Adjustment of weight estimation based on a “guess”. |
Omisanjo 2014 [148] | 2754 | Nigeria | P | 1 month to 11 years | Best Guess, Nelson | MPE < ±5% | 1, 2 | Low | Findings: Neither formula was accurate in Nigerian children with a substantial overestimation of weight. Comments: This was one of the largest prospective studies of age-based formulas in the developing world. Limitations: Some limitations from incomplete statistics. |
Abdel-Rahman 2015 [64] | 400** | USA | P | 2 month to 16 years | EPLS, Luscombe, DWEM, HCP guess, Broselow tape, 2DMT, 3DMT | 1 | High | Main results: “Real world” performance of weight estimation less accurate than when performed by experts. Calculation errors were common with age-based formulas. Weight and habitus were often underestimated by visual inspection. Usage errors with Broselow tape and Mercy tapes were common. Limitations: Not all modern weight estimation systems evaluated. | |
Asskaryar 2015 [57] | 1185 | India | P | 1 month to 12 years | Broselow tape | < 10%† | 1, 2 | Low | Findings: Broselow tape significantly overestimated weight in Indian children. An 8% modification of the tape improved its accuracy. Comments: Broselow tape 2007 edition B. The improved performance was not substantially better than the original and was still below acceptable performance. Recalibrating bias alone is not enough when precision is low. Limitations: Incomplete presentation of data. |
Badeli 2015 [149] | 216 | Iran | P | 1 to 10 years | DWEM, Oakley, TJ, TK, MAC, Theron, Leffler, EPLS, healthcare provider guess, parental estimate | None | 1 | High | Findings: The authors reported that healthcare provider guesses and EPLS formula were more accurate than other methods. Comments: The findings are completely unreliable and the conclusions consequently unreasonable. Limitations: Incomplete presentation of data. Analysis of bias confused with accuracy. |
Britnell 2015 [47] | 376 | New Zealand | P | 5 to 10 years | EPLS, Shann, Theron, Broselow tape | < 10%† | 1, 2 | Low | Findings: Broselow tape more accurate than age-based formulas in children < 143 cm. Current acceptance of formulas needs to change. Large differences in accuracy of weight estimation in different ethnic groups. Comments: Broselow tape 2011 edition A. Large proportion of Pacific Island children in sample. Broselow tape could not be used in one fifth of children, but best ever performance of the Broselow tape in a study. Habitus was assessed but data not presented. Limitations: Incomplete presentation of data. |
Chavez 2015 [13] | 324 | USA | P | 1 month to 12 years | Broselow tape, PAWPER, APLS, MAC | None | 1, 2 | Low | Findings: Age-based formulas and the MAC formulas performed badly. The Broselow tape performed better and the PAWPER was most accurate overall, although not as accurate as in previous studies. Comments: Underestimation of obesity (and habitus score) caused underestimation of weight. High level of obesity in study population. Limitations: Broselow tape version not reported. Incomplete presentation of data. PW5 mistaken for PW10 by authors—confusing for readers. |
Garcia 2015 [65] | 1698 | USA | P | – | PAWPER | None | 1, 2 | Low | Findings: The PAWPER did not perform as well as the original study. Comments: Most assignment of habitus score by nurses. Very high proportion of obesity. Limitations: Many incorrect assignments of habitus score. Children not fully undressed to assess habitus. |
Khouli 2015 [150] | 815 | Mexico | P | 0 to 12 years | Broselow tape | None | 1, 2 | Low | Findings: Broselow tape not accurate in Mexican population. Comments: Reasonable statistics. Both under- and overestimation found to be a problem. Nearly half of children had at least one colour zone error. Limitations: Broselow tape not actually used and Broselow tape version not reported. |
Skrobo 2015 [151] | 3155 | Ireland | R | 1 to 15 years | EPLS, Luscombe | < 15% | 1, 2 | Low | Main results: The Luscombe formula underestimated weight less than the EPLS formula. Comments: Increase in children’s weight in “modern” populations related to increase in both lean body weight and adipose. Limitations: Only data on bias presented. Limited and incomplete data presentation. |
Talib 2015 [152] | 318 | USA | P | 0 to 18 years | APLS, Broselow tape, MAC formula, Mercy method | None | 1 | High | Main results: Mercy method performed better than Broselow tape and age-based formulas in children with Down syndrome. Comments: No system performed well in this population. Limitations: Limited and incomplete data presentation. |
Young 2015 [153] | 207 | Philippines | P | 1 to 9 years | EPLS, APLS, Luscombe, BG, finger counting, Broselow tape | None | 1 | High | Findings: Broselow tape performed best in this population, updated APLS formula performed worst. Comments: Broselow tape 2011 edition A. Limitations: Broselow tape not actually used. Incomplete presentation of data. |
AlHarbi 2016 [154] | 3537 | Saudi Arabia | P | 1 month to 12 years | Broselow tape 2007B and Broselow tape 2011A | None | 1 | High | Findings: Broselow tape 2011A performed better than Broselow tape 2007B in this population. The authors suggested that the tapes were accurate. Comments: The method of statistical analysis does not support the conclusions drawn in this paper. Limitations: Unclear if the Broselow tape was used or if derived from length measurements. Limited and incomplete data presentation. |
Aliyu 2016 [155] | 300 | Nigeria | P | 0 to 5 years | Broselow tape, APLS | None | 1, 2 | Low | Findings: Broselow tape and APLS formulas performed well in this population. Comments: Contrary to the authors’ conclusions, although the Broselow tape and APLS formula performed similarly, they were both inaccurate. Limitations: Not clear if Broselow tape used or if derived from length measurements. Broselow tape edition not reported. Limited and incomplete data presentation. |
Britnell 2016 [51] | 376 | New Zealand | P | 5 to 10 years | Novel weight lookup table | < 10% | 1 | Low | Main results: Length- and habitus-based model better than age- and habitus-based model. Model outcome similar to Broselow tape. Comments: Useful additional information on value of habitus-based methods of adjusting weight estimations. Limitations: Limited and incomplete data presentation. |
Clark 2016 [62] | 583 | Sudan | R | 6 months to 5 years | Broselow tape | < 10%** | 1,2 | Low | Findings: The Broselow tape performed very poorly. Comments: Abstract. Study in South Sudan, the “hungriest place on earth” where 61% of study population was malnourished. There was up to a two colour-zone overestimation in severely malnourished children with only 26% agreement in normally nourished children. Very poor performance of the Broselow tape. Dangerous overestimation of weight in undernourished children. Limitations: Broselow tape not actually used and Broselow tape version not reported. |
Carasco 2016 [156] | – | – | – | – | – | None | 1 | N/A | Main results: EPLS formula may be better suited to identifying ideal body weight rather than total body weight. Comments: Systematic review. Ideal body weight should only be used in obese children. Limitations: No weight estimation data. |
Chassee 2016 [157] | 197 | USA | R | 0 to 12 years | 911 caller weight estimation | < 20% | 1 | Low | Main results: Family members could frequently provide a weight estimate to the 911 operator. Comments: More than 20% of cases had a weight estimation error > 20%. Limitations: Limited and incomplete data presentation. |
Georgoulas 2016 [21] | 300 | South Africa | P | 1 month to 12 years | Broselow tape, PAWPER, Wozniak, Mercy | < 10%* | 1, 2 | Low | Findings: PAWPER tape performed best, but good performances from Wozniak and Mercy methods. Broselow tape was worst performer. Comments: Unpublished data. Broselow tape 2011 edition A. Poor population with high proportion of underweight children. First comparative study of these methodologies. Relatively weaker performance by all methods in infants, but Wozniak especially was very weak. Limitations: Assessment of body habitus done by single researcher. |
Jung 2016 [16] | 906 | Korea | P | 0 to 17 years | Broselow tape, novel device | None | 1, 2 | Low | Main results: The novel device performed better than the Broselow tape in all outcome measures and was quicker to use. Comments: Both devices underestimated weights. Overweight and underweight children were often misclassified into wrong habitus category. Broselow tape 2011 edition A. Limitations: Device is not commercially available. Incomplete statistical analysis |
Jung 2016 [158] | – | – | – | – | – | None | 1 | N/A | Non-systematic review. Main results: Weight frequently underestimated in older children. Differences in body habitus account for this error, which needs further research. Comments: Article in Korean. Limitations: Non-systematic review. No quantitative assessment of methodologies. |
Lowe 2016 [55] | 3018 | USA | R | 0 to 13 years | Broselow tape, Handtevy tape | None | 1, 2 | Low | Main results: The Broselow tape performed better than the Handtevy tape in all outcome measures. Both tapes underestimated weights. Comments: Although the authors recognised the increased accuracy when obese children were excluded, they still erroneously advocated “recalibration” of the tapes. Broselow tape 2011 edition A. Limitations: Tapes not actually used. Limited and incomplete data presentation. |
Mishra 2016 [159] | 603 | India | P | 0 to 10 years | Broselow tape | None | 1 | Low | Findings: Broselow tape performed best in smallest children. Comments: Broselow tape 2007 edition B. Only colour zone accuracy was assessed, which was poor in children > 18 kg. Limitations: Limited and incomplete data presentation. |
Nosaka 2016 [160] | 237 | Japan | P | 0 to 10 years | EPLS, Park, Broselow tape, parental estimates | None | 1, 2 | High | Main results: Parental estimates were most accurate, then Broselow tape then APLS formula. Mothers’ weights, as well as Broselow tape and EPLS estimations were extremely accurate. Comments: The BMI for this population suggests that very few obese children were included and the sample was skewed towards very young children. Broselow tape 2007 edition B. Limitations: Limited and incomplete data presentation. |
Ralston 2016 [29] | 453,990 | Multicentre International | R | 6 months to 5 years | Broselow tape, MAC, height + MAC model | None | 1, 2 | Low | Findings: A novel model incorporating height and MAC was the most accurate. Broselow tape 2011 edition A less accurate than 2007 edition B in this population. Comments: Broselow tape 2007 edition B and 2011 edition A. A very large multinational database study. Limitations: Broselow tape not actually used. |
Sahar 2016 [161] | 1163 | Malaysia | P | 0 to 12 years | Broselow tape | None | 1, 2 | Low | Findings: Broselow tape underestimated weight in small children and overestimated weight in older children. It was not accurate. Comments: As with studies elsewhere there was a large variation in accuracy. Limitations: Broselow tape version not reported. Limited and incomplete data presentation. |
So 2016 [162] | 4600 | Hong Kong | R | 1 to 12 years | APLS, EPLS, BG, Luscombe, CAWR-1, CAWR-3 | None | 1, 2 | Low | Main results: A new CAWR formula was more accurate than other formulas in this study. Comments: No formula was very accurate and the differences between formulas were generally small. Limitations: Data very skewed towards younger children. No inclusion of length-based methods. |
So 2016 [163] | 4178 | Hong Kong | R | 1 to 9 years | Finger counting | None | 1, 2 | Low | Main results: The finger-counting formula was more accurate than other formulas. Comments: Same data was used for So 2016, but with a narrower age-restriction and the new finger-counting formula. Limitations: Duplicate data. Only the data on finger counting was added. |
Tanner 2016 [17] | 178 | USA | P | 2 to 18 years | Broselow tape | None | 1 | High | Main results: The Broselow tape was very inaccurate in overweight and obese children, who made up more than half the sample. Habitus assessment was poor by parents and nurses, but better by the principal investigator. Comments: Using a corrective formula based on waist circumference to adjust the Broselow tape weight improved accuracy in obese children. Limitations: Broselow tape version not reported. Limited and incomplete statistical analysis. Only included children who fell within the length limitations of the tape. |
Young 2016 [9] | – | – | – | – | – | – | 1 | N/A | Systematic review. Main results: Parental estimates were the most accurate technique followed by length- and habitus-based methods. Broselow tape more accurate than age formulas. Limitations: No quantitative data analysis. |
Bowen 2017 [164] | 1381 | Zambia | P | 0 to 14 years | Broselow tape, APLS, EPLS, ARC, Argall, BG, CAWR, Garwood, Leffler, Luscombe, Michigan, Nelson, Park, Shann, Theron | < 10%* | 1, 2 | Low | Findings: Broselow tape performed better than every formula in this population, BG and Michigan formulas performed worst. Comments: None of the methods were accurate, and all methods overestimated weight. Limitations: Broselow tape not actually used. Broselow tape version not reported. |
O’Leary 2017 [22] | 199 | Australia | P | 0 to 14 years | PAWPER, APLS, Luscombe, BG, Broselow tape, Mercy | < 10%* | 1, 2 | Low | Findings: Age-based formulas performed badly. The Broselow tape and Mercy method performed significantly better and the PAWPER was most accurate overall. Comments: All systems performed worst in infants. Reasonably good performance of Broselow tape, possibly because of newer edition used. Limitations: Broselow tape version not reported. |
Reilly 2017 [165] | None | 1 | N/A | Commentary. Main results: MAC is useful as an adjunct to estimate weight. Limitations: No original data. | |||||
Samerchua 2017 [68] | 430 | Thailand | P | 0 to 12 years | Parental estimations, Mercy method, Broselow tape, EPLS, APLS | None | 1, 2 | Low | Main results: Parental estimates were most accurate, followed by the Mercy method. Comments: Parents could only provide an estimate in 80% of cases—the Mercy method was the only method that could be used in all children. Limitations: It was not clear what training the raters had received. |
Trainarongsakul 2017 [166] | 345 | Thailand | P | 0 to 15 years | RAMA Ped card | None | 1 | Low | Main results: RAMA Ped card had a fair correlation with weight. Comments: The use of only correlation data made the findings hard to interpret. Limitations: Limited and incomplete data presentation. |
Waseem 2017 [167] | 538 | USA | R | 0 to 8 years | Broselow tape | None | 1 | N/A | Main results: Broselow tape underestimated weight in older children. It was not accurate in nearly half the population. Broselow tape 2011 edition A. Comments: Many underweight and obese children included which cause the poor performance. Limitations: Broselow tape not actually used. Limited and incomplete data presentation. Only included children who fell within the length limitations of the tape. |
Wells 2017 [20] | 328 | South Africa | P | 0 to 16 years | Broselow tape, PAWPER, Wozniak, Mercy | < 10%† | 1, 2 | Low | Findings: The Broselow tape performed poorly in this study, the Wozniak method and the Mercy method showed intermediate accuracy and the PAWPER was most accurate overall. Comments: This was a population with many older children and children with deviations from “average” weight-for-length. The PAWPER XL tape worked well in this population. Limitations: The Mercy method was used in a simulated resuscitation setting (supine children), which may have affected its accuracy. |
Wells 2017 [11] | 1085 | South Africa | P | 0 to 16 years | 20 age-based formulas, Traub-Johnson and Traub-Kichen formulas including habitus-modified age- and length-based formulas | < 10% | 1, 2 | Low | Findings: No age- or length-based formula performed acceptably well. Age-based habitus-modified formulas also performed poorly, but length-based habitus modified formulas were, surprisingly, extremely accurate. Comments: Length-based formulas can predict both TBW and IBW, but are mathematically complex. Limitations: Habitus assessment by visual estimation is subjective. |
Wells 2017 [31] | 13,134 | USA | R | 0 to 18 years | PAWPER XL-MAC method | < 10% | 1, 2 | Low | Findings: The PAWPER XL-MAC method was very accurate in data from both the USA and South Africa. Comments: This sytem was completely objective with no visual assessment of habitus. Limitations: Retrospective study |
Whitfield 2017 [14] | – | – | – | – | – | – | – | – | As for Wozniak 2012 |