As Africa is experiencing the most rapid rise in the world of type 2 diabetes (T2D), identifying feasible, accurate ways to measure glucose is imperative. While venous glucose (ven-glucose) is recommended, cost and lack of access to pre-analytical processing often rules out this option. Fingerstick measurement of capillary glucose (fs-glucose) is a potential alternative. But data on the correlation between fs-glucose and ven-glucose are scant. Hence OGTT were performed in 512 Ugandans (Male: 46% (234/512), age: 43±17y (mean±SD) enrolled in the Diabetes in Low Resource Populations Study with fs-glucose (Contour Plus Glucometer) and ven-glucose (Glucokinase Method, Roche Cobas 6000) obtained at 0h and 2h. To determine correlation, bias and levels of agreement, Pearson’s correlations and Bland Altman analyses were performed. T2D, preT2D and normal glucose tolerance occurred in 2% (9/512), 18% (92/512) and 80% (411/512), resp. At 0h, fs-glucose and ven-glucose were modestly correlated (r= 0.56) (Fig) with a modest bias for lower glucose by fs-glucose 0.28 mmol/L (95% CI: 0.23, -0.34). At 2h, glucose, fs-glucose and ven-glucose were highly correlated (r=0.92) (Fig) with modest bias for higher glucose by fs-glucose 0.60 mmol/L (95% CI: 0.53, 0.66) mmol/L. Overall, fs-glucose may be an alternative to ven-glucose in Africans where the prevalence of undetected T2D is high and resources are limited.


A.J.Niwaha: None. A.E.Sumner: None. H.Nkabura: None. R.J.Mukasa: None. N.Mubiru: None. P.A.Balungi: None. I.Sekitoleko: None. B.Mpairwe: None.

Readers may use this article as long as the work is properly cited, the use is educational and not for profit, and the work is not altered. More information is available at