The accuracy of adult patient weight estimation in tertiary level hospital in South Africa

Abstract

Introduction The weight of a patient is an important variable in the emergency department (ED) as it determines various drug doses and certain interventions. Patients, however, may be too sick or injured to allow for an actual weight to be measured. ED healthcare providers commonly rely on patient estimates, “guesstimation” and formulae. The aim of this study was to assess the accuracy of patient self-estimation as well as the Buckley, Broca and Mercy methods for adult weight estimation. Methods This was a prospective, cross-sectional study conducted at a tertiary, academic hospital in a metropolitan area of Johannesburg, South Africa. Anthropometric variables of height, abdominal circumference, thigh circumference, mid-arm circumference and humeral length were measured. These variables were then applied to the various weight estimation methods and compared to the patient’s actual weight. Results There were 188 adult patients included in the study. The Mercy method was more accurate than the other methods with 60.2% of estimations within 10% of measured weight (PW10) and 90.9% of weight estimation within 20% of measured weight (PW20). Patient self-estimate was also an accurate way to obtain patient weight (PW10 67.6% and PW20 86.1%). Conclusion The Mercy method using the mid-arm circumference showed similar benefit in adults as a weight estimation method as has been found in children. It performed functionally and statistically better than other weight estimation methods. When available, patient weight self-estimate is reliable when actual weight cannot be measured.