The aim of this study was to evaluate the precision of estimated height from ulna length (UL) using the Malnutrition Universal Screening Tool (MUST) equation and compare it with predicted height from knee height (KH), and self-reported height in estimating actual body height in inpatients.This cross-sectional study was carried out with patients admitted to an emergency service of a tertiary public hospital. Data were collected, at the patients' bedsides, from a specific questionnaire and anthropometric measurements. Height was estimated from UL and KH, and compared with self-reported and actual height. The Bland-Altman methods were used to evaluate agreement between measures.This study included 427 inpatients (52.6% women, ages 54.30 ± 15.39 y). A significant difference was found when actual height (161.07 ± 8.77 cm) was compared with estimated height from KH (163.64 ± 8.61 cm) and self-reported height (164.54 ± 8.95 cm). A not significant difference of 1.07 cm was observed between actual height and estimated height from UL (160.74 ± 7.48 cm); however, the limits of agreement between measures were large (from 13.69 to 14.39 cm). The difference observed between actual and self-reported height was higher in men (-8.50 [-17.00; -2.00] cm) than in women (1.00 [-6.00; 8.00] cm), whereas the difference between actual height and estimated height from KH and UL did not differ statistically between men and women.Estimated height from UL (MUST equation), self-reported height and estimated height from KH led to errors in predicting actual height of inpatients.