Background and Aim: Anthropometric equations are widely used to estimate fat mass (FM). The primary aim of this study was to develop and validate a new simplified anthropometric equation for estimating FM, using dual-energy X-ray absorptiometry (DXA) as the reference method in apparently healthy adults. A secondary aim was to compare its accuracy with previously published two-component models. Methods and Results: A total of 515 adults (60% men, 40% women; age 24.2 ± 6.0 years, range 18–45; BMI 23.6 ± 3.1 kg·m⁻², range 16.2–36.0) underwent standardized anthropometric assessment according to the International Society for the Advancement of Kinanthropometry (ISAK) guidelines, including skinfold thickness measurements, and whole-body DXA scanning. Participants were randomly allocated to a development sample (n = 344) and an independent validation sample (n = 171). A simplified multiple regression model based on a four-skinfold protocol (triceps, subscapular, iliac crest, and calf) was developed. Model performance was evaluated using R², standard error of the estimate (SEE), mean absolute error (MAE), root mean square error (RMSE), and Bland–Altman analysis. In the validation sample, the four-skinfold equation explained 80% of the variance in DXA-derived FM, with low prediction error (MAE = 2.37 kg; RMSE = 3.03 kg) and no significant bias compared with DXA (p = 0.25). In contrast, all previously published traditional equations (Durnin & Womersley, Faulkner, Yuhasz, and Wilmore) showed significant bias (p < 0.001), consistently underestimating FM. Conclusions: The simplified four-skinfold equation, referenced against DXA and derived from standardized ISAK anthropometry, demonstrated excellent agreement and lower systematic bias than traditional two-component models, confirming its value as a robust and time-efficient tool for field assessment of fat mass.
Contesto e obiettivo: Le equazioni antropometriche sono ampiamente utilizzate per stimare la massa grassa (FM). Questo studio mirava in primo luogo a sviluppare e validare una nuova equazione antropometrica semplificata per la stima della FM, utilizzando l'assorbimento a raggi X a doppia energia (DXA) come metodo di riferimento in adulti apparentemente sani. Successivamente, ci siamo proposti di confrontare la sua accuratezza con modelli basati su due componenti precedentemente pubblicati. Metodi e risultati: Cinquecentoquindici adulti (60% uomini, 40% donne; età 24.2\pm6.0 anni, intervallo 18-45; IMC 23.6\pm3.1 kg\cdot m^{-2}, intervallo 16.2-36.0) sono stati sottoposti a valutazione antropometrica standardizzata secondo le linee guida della International Society for the Advancement of Kinanthropometry (ISAK), inclusa la misurazione delle pliche cutanee, e a DXA total-body. I partecipanti sono stati suddivisi casualmente in un campione di sviluppo (n=344) e un campione di validazione indipendente (n=171). È stato sviluppato un modello di regressione multipla semplificato basato su un protocollo a quattro pliche cutanee (tricipitale, sottoscapulare, cresta iliaca e polpaccio). Le prestazioni del modello sono state valutate utilizzando R^{2}, l'errore standard della stima (SEE), l'errore assoluto medio (MAE), la radice dell'errore quadratico medio (RMSE) e l'analis di Bland-Altman. Nella validazione, l'equazione a 4 pliche ha spiegato l'80% della varianza della FM derivata da DXA, con un basso errore di predizione (MAE 2.37 kg; RMSE 3.03 kg) e nessun bias significativo rispetto alla DXA (p=0.25). Al contrario, tutte le equazioni tradizionali precedentemente pubblicate (Durnin & Womersley, Faulkner, Yuhasz, Wilmore) hanno mostrato un bias significativo (p<0.001), sottostimando costantemente la FM. Conclusioni: L'equazione semplificata a 4 pliche, referenziata con DXA e derivata dall'antropometria standardizzata ISAK, ha dimostrato un'ottima concordanza e un minore bias sistematico rispetto ai tradizionali modelli basati su due componenti, confermandosi una soluzione robusta e a basso impiego di tempo per la valutazione sul campo.
Sviluppo e validazione di una nuova formula per stimare il grasso corporeo su base antropometrica
CECCHINATO, ALBERTO
2025/2026
Abstract
Background and Aim: Anthropometric equations are widely used to estimate fat mass (FM). The primary aim of this study was to develop and validate a new simplified anthropometric equation for estimating FM, using dual-energy X-ray absorptiometry (DXA) as the reference method in apparently healthy adults. A secondary aim was to compare its accuracy with previously published two-component models. Methods and Results: A total of 515 adults (60% men, 40% women; age 24.2 ± 6.0 years, range 18–45; BMI 23.6 ± 3.1 kg·m⁻², range 16.2–36.0) underwent standardized anthropometric assessment according to the International Society for the Advancement of Kinanthropometry (ISAK) guidelines, including skinfold thickness measurements, and whole-body DXA scanning. Participants were randomly allocated to a development sample (n = 344) and an independent validation sample (n = 171). A simplified multiple regression model based on a four-skinfold protocol (triceps, subscapular, iliac crest, and calf) was developed. Model performance was evaluated using R², standard error of the estimate (SEE), mean absolute error (MAE), root mean square error (RMSE), and Bland–Altman analysis. In the validation sample, the four-skinfold equation explained 80% of the variance in DXA-derived FM, with low prediction error (MAE = 2.37 kg; RMSE = 3.03 kg) and no significant bias compared with DXA (p = 0.25). In contrast, all previously published traditional equations (Durnin & Womersley, Faulkner, Yuhasz, and Wilmore) showed significant bias (p < 0.001), consistently underestimating FM. Conclusions: The simplified four-skinfold equation, referenced against DXA and derived from standardized ISAK anthropometry, demonstrated excellent agreement and lower systematic bias than traditional two-component models, confirming its value as a robust and time-efficient tool for field assessment of fat mass.| File | Dimensione | Formato | |
|---|---|---|---|
|
Cecchinato_Alberto.pdf.pdf
accesso aperto
Dimensione
1.19 MB
Formato
Adobe PDF
|
1.19 MB | Adobe PDF | Visualizza/Apri |
The text of this website © Università degli studi di Padova. Full Text are published under a non-exclusive license. Metadata are under a CC0 License
https://hdl.handle.net/20.500.12608/115457