V N Mbanya1, A P Kengne2, J C Mbanya3, H Akhtar4. 1. Section of International Health, Department of Community Medicine, University of Oslo, Norway; Health of Populations in Transition (HoPiT) Research Group, Faculty of Medicine and Biomedical Sciences, The University of Yaoundé 1, Yaoundé, Cameroon. Electronic address: vivimbanya@yahoo.com. 2. Health of Populations in Transition (HoPiT) Research Group, Faculty of Medicine and Biomedical Sciences, The University of Yaoundé 1, Yaoundé, Cameroon; South African Medical Research Council Cape Town, South Africa; University of Cape Town, Cape Town, South Africa. Electronic address: akengne@georgeinstitute.org.au. 3. Health of Populations in Transition (HoPiT) Research Group, Faculty of Medicine and Biomedical Sciences, The University of Yaoundé 1, Yaoundé, Cameroon. Electronic address: jcmbanya@yahoo.co.uk. 4. Section of International Health, Department of Community Medicine, University of Oslo, Norway. Electronic address: hussain.akhtar@medisin.uio.no.
Abstract
BACKGROUND: The link between measures of adiposity and prevalent screen-detected diabetes (SDM) in Africa has been less well investigated. We assessed and compared the strength of association and discriminatory capability of measures of adiposity including body mass index (BMI), waist circumference (WC), hip circumference (HC), waist-hip-ratio (WHR) and waist-height-ratio (WHtR) for prevalent SDM risk in a sub-Saharan African population. METHODS: Participants were 8663 adults free of diagnosed type 2 diabetes, who took part in the nationally representative Cameroon Burden of Diabetes (CAMBoD) 2006 survey. Logistic regression models were used to compute the odd ratio (OR) and 95% confidence interval (95%CI) for a standard deviation (SD) higher level of BMI (7.3), WC (12.5), HC (11.7), WHR (0.19) and WHtR (0.08) with prevalent SDM risk. Assessment and comparison of discrimination used C-statistic and relative integrated discrimination improvement (RIDI, %). RESULTS: The adjusted OR and 95%CI for prevalent SDM with each SD higher adipometric variable were: 1.05 (0.98-1.13) for BMI, 1.30 (1.16-1.46) for WC, 1.18 (1.05-1.34) for HC, 1.05 (1.00-1.16) for WHR and 1.26 (1.11-1.39) for WHtR. C-statistic comparisons and RIDI analyses showed a trend toward a significant superiority of WC over other adipometric variables in multivariable models. Combining adiposity variables did not improve discrimination beyond multivariable models with WC alone. CONCLUSION: WC was the best predictors and to some extent WHtR of prevalent SDM in this population, while BMI and WHR were less effective.
BACKGROUND: The link between measures of adiposity and prevalent screen-detected diabetes (SDM) in Africa has been less well investigated. We assessed and compared the strength of association and discriminatory capability of measures of adiposity including body mass index (BMI), waist circumference (WC), hip circumference (HC), waist-hip-ratio (WHR) and waist-height-ratio (WHtR) for prevalent SDM risk in a sub-Saharan African population. METHODS:Participants were 8663 adults free of diagnosed type 2 diabetes, who took part in the nationally representative Cameroon Burden of Diabetes (CAMBoD) 2006 survey. Logistic regression models were used to compute the odd ratio (OR) and 95% confidence interval (95%CI) for a standard deviation (SD) higher level of BMI (7.3), WC (12.5), HC (11.7), WHR (0.19) and WHtR (0.08) with prevalent SDM risk. Assessment and comparison of discrimination used C-statistic and relative integrated discrimination improvement (RIDI, %). RESULTS: The adjusted OR and 95%CI for prevalent SDM with each SD higher adipometric variable were: 1.05 (0.98-1.13) for BMI, 1.30 (1.16-1.46) for WC, 1.18 (1.05-1.34) for HC, 1.05 (1.00-1.16) for WHR and 1.26 (1.11-1.39) for WHtR. C-statistic comparisons and RIDI analyses showed a trend toward a significant superiority of WC over other adipometric variables in multivariable models. Combining adiposity variables did not improve discrimination beyond multivariable models with WC alone. CONCLUSION: WC was the best predictors and to some extent WHtR of prevalent SDM in this population, while BMI and WHR were less effective.
Authors: Dale S Hardy; Devita T Stallings; Jane T Garvin; Francine C Gachupin; Hongyan Xu; Susan B Racette Journal: J Diabetes Date: 2016-07-12 Impact factor: 4.006
Authors: Yuchun Tao; Jianxing Yu; Yuhui Tao; Hui Pang; Yang Yu; Yaqin Yu; Lina Jin Journal: Int J Environ Res Public Health Date: 2016-08-09 Impact factor: 3.390
Authors: Li-Xin Tao; Kun Yang; Fang-Fang Huang; Xiang-Tong Liu; Xia Li; Yan-Xia Luo; Li-Juan Wu; Xiu-Hua Guo Journal: Int J Environ Res Public Health Date: 2017-10-10 Impact factor: 3.390