Peripheral insulin resistance rather than beta cell dysfunction accounts for geographical differences in impaired fasting blood glucose among sub-Saharan African individuals: findings from the RODAM study.

Meeks, K.A. ; Stronks, K. ; Adeyemo, A. ; Addo, J. ; Bahendeka, S. ; Beune, E. ; Owusu-Dabo, E. ; Danquah, I. ; Galbete, C. ; Henneman, P. ; Klipstein-Grobusch, K. ; Mockenhaupt, F.P. ; Osei, K. ; Schulze, M.B. ; Spranger, J. ; Smeeth, L. ; Agyemang, C. ;
Peripheral insulin resistance rather than beta cell dysfunction accounts for geographical differences in impaired fasting blood glucose among sub-Saharan African individuals: findings from the RODAM study.
Diabetologia, 2017;

The aim of this study was to assess the extent to which insulin resistance and beta cell dysfunction account for differences in impaired fasting blood glucose (IFBG) levels in sub-Saharan African individuals living in different locations in Europe and Africa. We also aimed to identify determinants associated with insulin resistance and beta cell dysfunction among this population.

Data from the cross-sectional multicentre Research on Obesity and Diabetes among African Migrants (RODAM) study were analysed. Participants included Ghanaian individuals without diabetes, aged 18-96 years old, who were residing in Amsterdam (n = 1337), Berlin (n = 502), London (n = 961), urban Ghana (n = 1309) and rural Ghana (n = 970). Glucose and insulin were measured in fasting venous blood samples. Anthropometrics were assessed during a physical examination. Questionnaires were used to assess demographics, physical activity, smoking status, alcohol consumption and energy intake. Insulin resistance and beta cell function were determined using homeostatic modelling (HOMA-IR and HOMA-B, respectively). Logistic regression analysis was used to study the contribution of HOMA-IR and inverse HOMA-B (beta cell dysfunction) to geographical differences in IFBG (fasting glucose 5.6-6.9 mmol/l). Multivariate linear regression analysis was used to identify determinants associated with HOMA-IR and inverse HOMA-B.

IFBG was more common in individuals residing in urban Ghana (OR 1.41 [95% CI 1.08, 1.84]), Amsterdam (OR 3.44 [95% CI 2.69, 4.39]) and London (OR 1.58 [95% CI 1.20 2.08), but similar in individuals living in Berlin (OR 1.00 [95% CI 0.70, 1.45]), compared with those in rural Ghana (reference population). The attributable risk of IFBG per 1 SD increase in HOMA-IR was 69.3% and in inverse HOMA-B was 11.1%. After adjustment for HOMA-IR, the odds for IFBG reduced to 0.96 (95% CI 0.72, 1.27), 2.52 (95%CI 1.94, 3.26) and 1.02 (95% CI 0.78, 1.38) for individuals in Urban Ghana, Amsterdam and London compared with rural Ghana, respectively. In contrast, adjustment for inverse HOMA-B had very minor impact on the ORs of IFBG. In multivariate analyses, BMI (β = 0.17 [95% CI 0.11, 0.24]) and waist circumference (β = 0.29 [95%CI 0.22, 0.36]) were most strongly associated with higher HOMA-IR, whereas inverse HOMA-B was most strongly associated with age (β = 0.20 [95% CI 0.16, 0.23]) and excess alcohol consumption (β = 0.25 [95% CI 0.07, 0.43]).

Our findings suggest that insulin resistance, rather than beta cell dysfunction, is more important in accounting for the geographical differences in IFBG among sub-Saharan African individuals. We also show that BMI and waist circumference are important factors in insulin resistance in this population.