What is the purpose of measuring a patients weight as a part of patient assessment?
|
Journal Article Show
S. Darnis, 1From the 1Université Aix Marseille II / Faculté de Pharmacie la Timone, Marseille, France, 2Faculté de Pharmacie-UPS, Toulouse III, France, 3Department of Pharmacy, Alfred Hospital, Melbourne, Australia 3004, 4Faculty of Pharmacy and Pharmaceutical Sciences, Melbourne, Australia 3004, 5Centre for Medicine Use and Safety, Monash Institute of Pharmaceutical Sciences, Monash University, Melbourne, Australia 3004, 6Infectious Disease Unit, Alfred Hospital, Melbourne, Australia 3004 and 7Department of Epidemiology and Preventive Medicine, Monash University, Melbourne, Australia 3004 Search for other works by this author on: N. Fareau,1From the 1Université Aix Marseille II / Faculté de Pharmacie la Timone, Marseille, France, 2Faculté de Pharmacie-UPS, Toulouse III, France, 3Department of Pharmacy, Alfred Hospital, Melbourne, Australia 3004, 4Faculty of Pharmacy and Pharmaceutical Sciences, Melbourne, Australia 3004, 5Centre for Medicine Use and Safety, Monash Institute of Pharmaceutical Sciences, Monash University, Melbourne, Australia 3004, 6Infectious Disease Unit, Alfred Hospital, Melbourne, Australia 3004 and 7Department of Epidemiology and Preventive Medicine, Monash University, Melbourne, Australia 3004 Search for other works by this author on: C.E. Corallo,1From the 1Université Aix Marseille II / Faculté de Pharmacie la Timone, Marseille, France, 2Faculté de Pharmacie-UPS, Toulouse III, France, 3Department of Pharmacy, Alfred Hospital, Melbourne, Australia 3004, 4Faculty of Pharmacy and Pharmaceutical Sciences, Melbourne, Australia 3004, 5Centre for Medicine Use and Safety, Monash Institute of Pharmaceutical Sciences, Monash University, Melbourne, Australia 3004, 6Infectious Disease Unit, Alfred Hospital, Melbourne, Australia 3004 and 7Department of Epidemiology and Preventive Medicine, Monash University, Melbourne, Australia 3004 Search for other works by this author on: S. Poole,1From the 1Université Aix Marseille II / Faculté de Pharmacie la Timone, Marseille, France, 2Faculté de Pharmacie-UPS, Toulouse III, France, 3Department of Pharmacy, Alfred Hospital, Melbourne, Australia 3004, 4Faculty of Pharmacy and Pharmaceutical Sciences, Melbourne, Australia 3004, 5Centre for Medicine Use and Safety, Monash Institute of Pharmaceutical Sciences, Monash University, Melbourne, Australia 3004, 6Infectious Disease Unit, Alfred Hospital, Melbourne, Australia 3004 and 7Department of Epidemiology and Preventive Medicine, Monash University, Melbourne, Australia 3004 Search for other works by this author on: M.J. Dooley,1From the 1Université Aix Marseille II / Faculté de Pharmacie la Timone, Marseille, France, 2Faculté de Pharmacie-UPS, Toulouse III, France, 3Department of Pharmacy, Alfred Hospital, Melbourne, Australia 3004, 4Faculty of Pharmacy and Pharmaceutical Sciences, Melbourne, Australia 3004, 5Centre for Medicine Use and Safety, Monash Institute of Pharmaceutical Sciences, Monash University, Melbourne, Australia 3004, 6Infectious Disease Unit, Alfred Hospital, Melbourne, Australia 3004 and 7Department of Epidemiology and Preventive Medicine, Monash University, Melbourne, Australia 3004 Search for other works by this author on: A.C. Cheng1From the 1Université Aix Marseille II / Faculté de Pharmacie la Timone, Marseille, France, 2Faculté de Pharmacie-UPS, Toulouse III, France, 3Department of Pharmacy, Alfred Hospital, Melbourne, Australia 3004, 4Faculty of Pharmacy and Pharmaceutical Sciences, Melbourne, Australia 3004, 5Centre for Medicine Use and Safety, Monash Institute of Pharmaceutical Sciences, Monash University, Melbourne, Australia 3004, 6Infectious Disease Unit, Alfred Hospital, Melbourne, Australia 3004 and 7Department of Epidemiology and Preventive Medicine, Monash University, Melbourne, Australia 3004 Search for other works by this author on: Received: 09 January 2012 Revision received: 05 March 2012
Close Navbar Search Filter Microsite Search Term Search AbstractAim: To examine the bias and precision of different methods of estimating body mass and height in hospitalized adult patients. Methods: Patients were enrolled at the Alfred and Caulfield hospitals, Melbourne, Australia following verbal consent. Estimates were made using the Lorenz formula (that utilizes height, waist and hip circumference), the Crandell formula (that utilizes height and arm circumference) and visual estimation of weight based on the average results obtained by two pharmacy interns. Statistical error was calculated as the ratio of estimated to actual weight; bias was assessed as the mean error and precision as the proportion of estimates within 10 and 20% of measured weight and standard deviation of the error. Results: In a 5-week period July to August 2010, 198 patients were enrolled. The median age was 64 years (range 19–91) and 52% were female. Thirty-four (17%) patients were obese (BMI >30 kg/m2) and 8 (4%) were underweight (BMI <18 kg/m2). With the Lorenz formula an estimate within 10% was obtained for 56% of patients; with the Crandell formula prediction was poor. Documentation of body weight in notes and patient self-reporting were both accurate. Seventy-two patients (43%) were prescribed one or more drugs for which dosing potentially should be adjusted for body weight. Conclusions: In adult hospitalized patients, the estimation of body weight by anthropomorphic measures is not accurate. This supports the need for equipment to be made widely available to accurately weigh patients directly in hospital, including in unconscious and immobile patients. IntroductionAn assessment of body weight is required for calculations of drug doses, tidal volumes in ventilated patients, patient safety (such as equipment load limits), estimation of renal function and nutritional status. In some patients, direct measurement of body weight is difficult because of immobility, trauma or burns, the lack of availability of measuring scales or intravascular lines that may easily become dislodged. Estimation of body weights by medical and nursing staff have previously been shown to be inaccurate.1–3 Estimating formulae have been developed based on anthropomorphic data in Europe and the USA.4,5 An estimation table of height based on ulna length is also commonly used.6 In this study, we aimed to examine the accuracy, assessed as systemic bias and precision, of different methods of estimating body mass, height and body mass index in hospitalized adult patients. MethodsStudy designPatients were approached on hospital wards at the Alfred Hospital, a tertiary referral hospital, and Caulfield Hospital, a geriatric hospital with acute and subacute care services, in Melbourne, Australia. The study was approved by the Alfred Health Ethics Committee. Enrolment included inpatients in both surgical and medical wards, as well as outpatients in surgical pre-admission clinic. Patients were excluded if they were immobile or unable to transfer (precluding measurement of body weight), both upper arms were inaccessible (e.g. burns, dressings) or if they were unable to co-operate or give verbal consent. A drug history from chart review was undertaken to identify if any currently prescribed drugs had dosing that could potentially be adjusted for body weight (Table 1).7,8 Table 1 Common drugs where weight-based dosing adjustments may be required in adults
Table 1 Common drugs where weight-based dosing adjustments may be required in adults
Estimates of body height and weight were made using the following methods (Figure 1) and compared to direct measurement using calibrated scales:
Figure 1. Estimators of body height and weight assessed in this study. Scales used in measurements were assessed by comparing a single measurement from each of the two pharmacy students and comparing the results with a calibrated scale provided by the Nutrition Department. Data analysisAccuracy was determined by assessment of bias and precision that both reflect the difference between the estimated and measured body weight or height (statistical error). The bias represents the systematic error between the mean estimated and mean measured body weight or height. Precision is the magnitude of differences between estimated and measured body weight or height for individuals. Statistical error was calculated as the ratio of estimated to actual weight or height; bias was assessed as the mean error minus 1, and precision as the standard deviation of the error. The Bland Altman plot reflects agreement between two methods of measurement using a scatterplot of the ratio of difference against the mean of the two measures. This shows both bias (the degree to which points are distributed around the zero difference line) and precision (the degree of variability around the zero difference line) along the full range of estimated body weights and heights. A priori subgroups of interest included medical patients, young patients and the elderly, patients by gender and in those with oedema. ResultsDuring the period July to August 2010, a total of 198 patients were enrolled from inpatients at the Alfred Hospital (n = 75), the Alfred Hospital pre-admission clinic (n = 45) and Caulfield Hospital (n = 78). The median age was 64 years (range 19–97) and 52% were female. Thirty-five (18%) patients were obese (BMI >30 kg/m2) and 8 patients (4%) were underweight (BMI <18 kg/m2). The mean (SD) height of male and female patients was 175 (7.9) cm and 164 (6.8) cm, respectively, and the mean (SD) body weight was 80 (14.9) kg and 68 (15.4) kg, respectively. Precision was poor in all methods to estimate body weight, and the Crandall formula also had a bias to over-estimate actual weight (Table 2 and Figure 2). The Crandall formula estimated body weight within 10% of the measured weight in only 67 (34%) patients; the error was between 10% and 20% in 47 (24%) patients and was >20% in 84 (42%) patients. The Lorenz formula over-estimated the weight in 32 and 34% of males and females, respectively and it under-estimated the weight in 11 and 10% of males and females, respectively. The Lorenz formula estimated body weight within 10% of the measured weight in 110 (56%) patients; the error was between 10% and 20% in 58 (29%) patients and was >20% in 30 (15%) patients. The precision of both methods was poor in all subgroups; the Crandall method systematically over-estimated body weight in all subgroups except those with a BMI >30 kg/m2. Figure 2. Bland Altman plot of weight estimates using the Lorenz formula against measured body weight. Table 2 Bias and precision of estimates of height and weight compared with direct measurement
Bias calculated as mean relative error minus 100%; precision calculated as standard deviation of relative error. BMI: body mass index. Table 2 Bias and precision of estimates of height and weight compared with direct measurement
Bias calculated as mean relative error minus 100%; precision calculated as standard deviation of relative error. BMI: body mass index. Body weight was documented in the clinical notes in 131 patients (66%) and this was found to be accurate (mean relative error <1%, SD 0.9%) The patient was able to provide an estimate of their weight in 175 cases (88%) and this self-reported weight was also accurate (mean relative error 1.0%, SD 3.9%). Estimation of body weight by pharmacy interns improved with experience. The bias and precision (standard deviation) for the first 60 patients was −3% and 9.5%, for the second 60 patients was −2% and 7.5% and for the final 78 patients was −3% and 6.3%. Ulna length as an estimator for height was associated with minimal bias in all subgroups, and the precision was <5% in most subgroups (Table 2 and Figure 3). Of the 16 patients with a MUAC of <23.5 cm, 10 patients (63%) had a BMI <20 kg/m2. Of the 41 patients with a MUAC >32 cm, 24 patients (59%) had a BMI >30 kg/m2. Figure 3. Bland Altman plot of height estimates based on ulnar length against measured body height. Of the 198 patients, 98 patients (49%) were on at least one medication for which dose potentially should be adjusted for body weight (Table 1). The most common drug was enoxaparin; in the 79 patients on this anticoagulant, 21 patients (27%) were dosed incorrectly for weight. DiscussionThe accurate measurement of height and weight is important in clinical practice, particularly as the ‘standard’ 70 kg ideal type is becoming less representative of the general population; in this study, only 9% of patients were within 10% of 70 kg body weight. Inaccurate measures of body weight may have potentially adverse consequences for dosing of many drug classes, and a number of studies have reported substantial estimation errors for height and weight in paediatric patients, operating room patients, emergency departments and intensive care unit.1–3 Where direct measurement is not possible, a number of estimation methods have been described.4,5,9 Our results suggest that the estimation of body weight from anthropomorphic measures is associated with significant error in hospitalized patients. The Crandall method was derived from 1471 individuals weighing >100 kg in a US population survey.4 Although we found that bias associated with this method was minimal in obese patients, estimates obtained using this method were imprecise in this group. The Lorenz method was derived from a German general practice sample and validated in two cohorts of hospital inpatients admitted with acute stroke.2,5 In the Lorenz method validation studies, only 93 and 80% of patients, respectively, had body weight estimates >10% from the measured weight, in contrast to our finding of only 56% of patients within 10% of measured body weight. This suggests that this method may not be generalizable outside of the European stroke patient populations where it has been validated. Of the two formulae, the Lorenz method is less practical in an emergency situation (requiring measurement of hip and waist circumference) but was associated with less bias and higher precision than the Crandall method. We found self-report to be a reliable measure in this group of conscious hospitalized patients as we found patients were often informed about weight measurements. Previous literature has supported self-reported weight in hospitalized patients2,9,10 but not outpatient populations.11 Mid-upper arm circumference was a poor estimator of BMI in this population. Ulna length as an estimator of height appeared to perform reasonably with minimal bias and moderate precision, but the limits of agreement correspond roughly to an error of ±15–20 cm, which is probably insufficiently precise for clinical use. Lin et al.9 obtained anthropometric estimations utilizing knee height and mid-arm circumference using a weight estimation tool. These estimations were found to be slightly more accurate (69%) than visual estimations by medical and nursing staff (54%) but less than patient reported weight (86%). We found an improvement of the pharmacy interns to estimate body weight with experience, with the overall accuracy comparable with published studies.2 While we were not able to assess the ability of trained staff formally, this suggests that the estimates of experienced clinicians (such as nutritionists) who regularly estimate body weight may be reasonably accurate. However, these estimates were made under ideal circumstances (in a well-lit environment, non-emergency situation, with a mobile, co-operative patient) and only used two interns; previous studies suggest that the ability of other clinical staff in estimating body weight is poor.1,12–15 Goutelle et al.16 conducted a study in elderly hospitalized patients to assess the accuracy of estimated by three pharmacy students who had little experience in clinical rounds. They conclude that in this patient group a single visual estimate should not be used because of large inter-observer variability. However, using the mean of several (at least 3) visual estimates may provide a reasonable accurate estimation for most patients. We found that a significant proportion of patients were on drugs that required dose adjustment based on body weight. This further reinforces the need to obtain this data in hospitalized patients. The results of our study indicate that in adult hospitalized patients, the estimation of body weight by anthropomorphic measures is not accurate. Ulna length as an estimate of height was moderately accurate. This reinforces the need for obtaining and documenting this information early in the admission process. It also highlights the need to make available equipment to accurately weigh patients directly, including the provision of specialized equipment in emergency departments and for immobile or unconscious patients. FundingAC is supported by a National Health and Medical Research Council Career Development Fellowship. Conflict of interest: None declared. References1 , , , . Accuracy of weight and height estimation in an intensive care unit: implications for clinical practice and research , Crit Care Med , 2006 , vol. 34 (pg. 2153 - 7 ) 2 , , , , , , et al. Weight approximation in stroke before thrombolysis: the WAIST-Study: a prospective observational ‘dose-finding’ study , Stroke , 2010 , vol. 41 (pg. 2867 - 71 ) 3 , , , , , , et al. Accuracy of weight and height estimation in an intensive care unit , Anaesth Intensive Care , 2010 , vol. 38 (pg. 930 - 4 ) 4 , , . Estimation of total body weight in obese patients , Air Med J , 2009 , vol. 28 (pg. 139 - 45 ) 5 , , , , , , et al. Anthropometric approximation of body weight in unresponsive stroke patients , J Neurol Neurosurg Psychiatry , 2007 , vol. 78 (pg. 1331 - 6 ) 6 , , , , . , The ‘MUST’ Explanatory Booklet , 2003 Redditch British Association for Parenteral and Enteral Nutrition 7 , . Adjustment of dosing of antimicrobial agents for bodyweight in adults , Lancet , 2010 , vol. 375 (pg. 248 - 51 ) 8 , , . Effect of obesity on the pharmacokinetics of drugs in humans , Clin Pharmacokinet , 2010 , vol. 49 (pg. 71 - 87 ) 9 , , , . A better way to estimate adult patients’ weights , Am J Emerg Med , 2009 , vol. 27 (pg. 1060 - 4 ) 10 , , , , . Weight assessment in cardiac patients: implications for prescription of low molecular weight heparin , Postgrad Med J , 2009 , vol. 85 (pg. 124 - 7 ) 11 , , , . A comparison of direct vs self-report measures for assessing height, weight and body mass index: a systematic review , Obes Rev , 2007 , vol. 8 (pg. 307 - 26 ) 12 , , , . The accuracy of visual estimation of body weight in the ED , Am J Emerg Med , 2004 , vol. 22 (pg. 526 - 9 ) 13 , , , . Challenges in the estimation of tidal volume settings in critical care units , Intensive Care Med , 2006 , vol. 32 (pg. 670 - 1 ) 14 , , , , , . Estimated height, weight, and body mass index: implications for research and patient safety , J Am Coll Surg , 2006 , vol. 203 (pg. 887 - 93 ) 15 , , . The accuracy of the estimation of body weight and height in the intensive care unit , Eur J Anaesthesiol , 2000 , vol. 17 (pg. 698 - 703 ) 16 , , , , . Visual estimation of patients’ body weight in hospital: the more observers, the better? , Pharm World Sci , 2009 , vol. 31 (pg. 422 - 5 ) © The Author 2012. Published by Oxford University Press on behalf of the Association of Physicians. All rights reserved. For Permissions, please email: © The Author 2012. Published by Oxford University Press on behalf of the Association of Physicians. All rights reserved. For Permissions, please email: Topic:
CitationsViewsAltmetricEmail alertsRelated articles in PubMedCiting articles via
More from Oxford Academic Why is it important to measure a patient's weight?Recording patient weight is a standard practice for all hospital admissions, with this measurement influencing other daily practices that rely on the delivery of safe and effective patient care. Patient weight is important in the areas of medication prescribing, fluid balance and assessment of nutrition.
Why is it important to know the residents height and weight?Most patients have their height and weight routinely measured during admission to a health care facility. An accurate record of the patient's height and weight is essential for calculating dosages of drugs, anesthetics, and contrast agents; assessing nutritional status; and determining the height-weight ratio.
Is it really important to measure and record the patient's height and weight on check up or on admission Why?Taking the patient's height and weight may indicate that the patient is overweight, underweight, or is retaining fluids (edema). The health care team can observe changes in weight caused by specific disease processes and determine the effectiveness of nutrition supplements prescribed to maintain weight.
When should we measure a Residents weight?Every resident should be weighed at least on a monthly basis. In some cases it is more suitable to weigh residents daily or weekly. One of the major causes of illness in care homes is malnutrition, affect- ing one third of care home residents. Major symptoms of malnutrition include a low weight reading and weight loss.
|
