Predictive formulas for estimation of height in sub-Saharan African older people: A new formula (EPIDEMCA study).
Jésus P., Marin B., Pilleron S., Guerchet M., Mbelesso P., Ndamba-Bandzouzi B., Preux P-M., Fayemendy P., Desport J-C.
ObjectivesChumlea's formulas are a way of predicting height from knee height (KHt), including among individuals >60 y of age who cannot stand upright. However, the formulas were developed with white and African American people and have not yet been validated in native Africans >60 y of age. The aims of the study were to assess Chumlea's formulas in older people in central Africa and to propose a new validated formula for the same population.MethodsHeight (MHt) and KHt were measured in a population of individuals >65 y of age from the Republic of Congo and the central African Republic. Predicted height (PHt) was calculated using Chumlea's formulas for the American black population (CBP) and for whites (CC). The percentages of accurate predictions (±5 cm) were compared between MHt and PHt. A new formula was estimated after randomization in a derivation sample (n = 877) and assessed for accuracy in a validation sample (n = 877).ResultsThe study included 1754 individuals. Prediction was accurate (±5 cm) in 66.8% and 63.2% of CBP and CC, respectively. Overestimation was as high as 24.1% and 29%, respectively. Because an overestimation is unacceptable in clinical practice and population surveys, a new formula was proposed: height (cm) = 72.75 + (1.86 × KHt [cm]) - (0.13 × age [y]) + 3.41 × sex (0: women; 1: men). This new formula significantly increases accuracy (71.3%) and decreases overestimation (14.7%). The nutritional status based on body mass index did not differ with the MHt and the PHt by the new formula.ConclusionChumlea's formulas provided a poor estimate of height in this population sample. The proposed formula more accurately estimates height in older native Africans. This formula should be tested in other sub-Saharan African countries to extend its use in clinical practice and in future studies.