Background The diabetes burden in sub-Saharan Africa is high, but data on the relative importance of insulin resistance and beta-cell dysfunction there is scarce. We investigated the association between dietary patterns with insulin resistance and beta-cell dysfunction among HIV-infected and HIV-uninfected adults in Mwanza, Tanzania.

Methods In a cross-sectional study, insulin resistance and beta-cell function were measured from plasma insulin and glucose during an oral glucose tolerance test. Diet data were collected using a validated food frequency questionnaire, and dietary patterns were derived by principal component analysis and reduced rank regression. Socio-demographics, smoking, alcohol taking, and physical inactivity data were collected using structured questionnaires. Logistic regression analysis was used to assess the association between insulin resistance, and beta-cell dysfunction with dietary patterns adjusting for potential confounders.

Results Of 462 participants, the mean age was 42 (±12) years, 58% were females, and 60% were HIV-infected. The proportion with insulin resistance was 43% and 35% by the Matsuda index and HOMA-IR, respectively. Beta-cell dysfunction was present in 37%, 43%, and 43.3% by the insulinogenic index, HOMA-β, and oral disposition index, respectively. Higher adherence to a carbohydrates-dense diet pattern was associated with more insulin resistance by HOMA-IR (aOR 3.7, 95% CI 2.2; 6.3) and Matsuda index (aOR 6.2 3.4; 11.2), and less beta-cell dysfunction by HOMA-β (aOR 0.4 0.2; 0.6) and insulinogenic index (aOR 0.5 0.3; 0.9). Higher adherence to the vegetable-rich pattern was associated with insulin resistance by the Matsuda index (aOR 2.2 1.3; 3.7). Conclusion Carbohydrate-dense pattern increases the risk of insulin resistance but decreases beta-cell dysfunction. Higher adherence to a vegetable-rich pattern increases the risk of insulin resistance. Further studies to look at glucose metabolism and why a vegetable-rich pattern has an odd effect in sub-Saharan Africa are warranted.

Funding: EDCTP-2 program, grant agreement number: TMA2017GSF-1965-REEHAD