Ulna length predicts height measured by stadiometer  among adults attending the University of Cape Coast  Hospital, Ghana

Kasim Abdulai; Safianu Osman Aleboko; Emil Kafui Amenu; Awal Mohammed Seidu; Gifty Korsah; Fatimatu Salifu; Olivia Ohui Otchie; Ivan Addae-Mensah; Sylvester Ofosu-Gyeabour; Abdul-Malik Bawah

doi:10.46829/hsijournal.2026.4.8.2.1508-1515

Kasim Abdulai Department of Nutrition and Dietetics, School of Allied Health Sciences, University of Cape Coast, Cape Coast, Ghana
Safianu Osman Aleboko Department of Nutritional Sciences, Oklahoma State University, United States
Emil Kafui Amenu Department of Nutrition and Dietetics, School of Allied Health Sciences, University of Cape Coast, Cape Coast, Ghana
Awal Mohammed Seidu Department of Nutrition and Dietetics, School of Allied Health Sciences, University of Cape Coast, Cape Coast, Ghana
Gifty Korsah Department of Nutrition and Dietetics, School of Allied Health Sciences, University of Cape Coast, Cape Coast, Ghana
Fatimatu Salifu Department of Nutrition and Dietetics, School of Allied Health Sciences, University of Cape Coast, Cape Coast, Ghana
Olivia Ohui Otchie Department of Nutrition and Dietetics, School of Allied Health Sciences, University of Cape Coast, Cape Coast, Ghana
Ivan Addae-Mensah Department of Nutrition and Dietetics, School of Allied Health Sciences, University of Cape Coast, Cape Coast, Ghana
Sylvester Ofosu-Gyeabour Faculty of Health and Allied Sciences, KAAF University College, Ghana
Abdul-Malik Bawah Department of Nutrition and Dietetics, Faculty of Allied Health and Pharmaceutical Sciences, Tamale Technical University, Ghana

DOI: https://doi.org/10.46829/hsijournal.2026.4.8.2.1508-1515

Keywords: Height, Ulna length, Agreement, Prediction

Abstract

Background: The Malnutrition Universal Screening Tool (MUST) suggests the use of ulna length as an alternative method for determining patients’ height when measured standing height is not possible to obtain. Ulna length has been studied as a potential surrogate measure for height estimation in various populations.
Objective: This study evaluated the agreement between height predicted by ulna length using the Elia (2003) predictive equation and height measured by stadiometer among adults in the University of Cape Coast Hospital (UCC-H) Outpatient Department (OPD) in Ghana.
Methods: This cross-sectional study sampled 402 adults from the UCC-H OPD in Ghana. Data on anthropometric measurements, including height and ulna length, were collected. R version 4.3.2 was used for statistical computing and graphics. Measurement error, error range, limits of agreement, and the Bland-Altman plot were used to assess the agreement between standing height and the height predicted by ulna length.
Sex-stratified analyses and internal validation were also performed.
Results: The mean difference (bias) between predicted and measured height was an overestimation of +14.6 cm (95% CI: 13.6 cm,15.5 cm). The 95% Limits of Agreement (LOA) were wide, ranging from −4.9 cm to +34.1 cm. Linear regression showed a strong correlation (R2 =0.86), but a Bland-Altman analysis revealed the lack of clinical interchangeability. The analysis for proportional bias was non-significant (p= 0.18). Sex-specific equations materially reduced the bias and LOA, although the agreement remained clinically unacceptable.
Conclusion: The Elia (2003) equation significantly and systematically overestimates the height of this sampled adult OPD population. The wide LOA indicates that the predicted height is not clinically interchangeable with measured height, which could lead to substantial misclassification of Body Mass Index (BMI). We recommend the development and external validation of population-specific equations before any routine clinical use in this setting.