RELATIVE LEG LENGTH IS INVERSELY ASSOCIATED WITH DIABETES IN BRAZILIAN ADULTS: THE BRAZILIAN LONGITUDINAL STUDY OF ADULT HEALTH
Wednesday, 20 August 2014: 5:30 PM
Ballroom D (Dena'ina Center)
Noel T Mueller, PhD
,
Universidade Federal do Rio Grande do Sul, Porto Alegre, Brazil
Bruce B Duncan, MD
,
Federal University of Rio Grande do Sul, Porto Alegre, Brazil
Sandhi M Barreto, PhD
,
Universidade Federal de Minas Gerais, Belo Horizonte, Brazil
Dora Chor, PhD
,
Fundação Oswaldo Cruz, Rio de Janeiro, Brazil
Alvaro Vigo, PhD
,
Universidade Federal do Rio Grande do Sul, Porto Alegre, Brazil
Estela M Aquino, PhD
,
Universidade Federal da Bahia, Bahia, Brazil
Ellen W Demerath, PhD
,
University of Minnesota, Minneapolis, MN
Maria Inês A Schmidt, PhD
,
Federal University of Rio Grande do Sul, Porto Alegre, Brazil
INTRODUCTION: Low relative leg length (RLL) - the ratio of leg length to sitting height - has been used as a marker of early-life undernutrition and earlier puberty. Studies from higher income countries have shown greater relative leg length predicted lower risk of type 2 diabetes mellitus. Whether RLL is associated with adult diabetes in a Latin American population, born and raised at a time when undernutrition was more common than overnutrition, and if this association is modified by age at menarche and adult BMI, is yet unknown. Our objective is thus to investigate in a large contemporary Brazilian cohort the hypothesis that relative leg length is inversely associated with diabetes risk, and evaluate whether this association is stronger among women with earlier menarche and adults who are overweight or obese.
METHODS: The Brazilian Longitudinal Study of Adult Health is an occupational cohort of 15,105 participants aged 35-74 years. Diabetes was defined based on a self-reported physician diagnosis, medication use, fasting glucose, 2-hour post 75g glucose, and glycated hemoglobin.
RESULTS: RLL was inversely associated with diabetes in Brazilian adults (PR for 1 SD increment = 0.89; 95% CI: 0.86-0.92), after adjustment (through Poisson regression) for age, sex, race, maternal education, maternal and paternal diabetes, birth weight, and BMI at age 20y. The association persisted after additional adjustment for menarche age (women), and hip and waist circumference at baseline, and it was stronger among women with earlier menarche (p for multiplicative interaction=0.06) and overweight or obese adults (p = 0.09).
CONCLUSIONS: Relative leg length is inversely associated with diabetes in Brazilian adults. This association is not explained by age at menarche, but it is modestly stronger among women with earlier menarche and overweight or obese adults. Future research is needed to elucidate the early-life social, behavioral, and biological underpinnings of this association.
