The association between anthropometric measures and glycated haemoglobin (HbA1c) is different in Russian, Somali and Kurdish origin migrants compared with the general population in Finland: a cross-sectional population-based study

Abstract

Abstract Background Persons of African and Middle-Eastern origin living in European countries have a high prevalence of type 2 diabetes, accompanied by high prevalence of obesity among women but not always among men. The aim of this study was to examine whether there are differences in the association between anthropometric measures and glucose levels measured with glycated haemoglobin and fasting blood glucose among persons of migrant origin in Finland. Methods Cross-sectional population-based data of the 30–64 year-old participants in the health examination of the Migrant Health and Wellbeing Study was used, selecting persons without diabetes (Russian origin n = 293, Somali origin n = 184, Kurdish origin n = 275). The reference group were non-diabetic participants in the Health 2011 Survey (n = 653), representative of the general Finnish population. Anthropometric measures included body mass index (BMI), waist circumference (WC), waist-to-height ratio (WHtR) and waist-to-hip ratio (WHR, available for Maamu Study participants only). Results Depending on whether continuous or categorical anthropometric measures were used, age, sex and anthropometrics explained 13–18% of variation in HbA1c among persons of Russian origin, 5–10% among persons of Somali origin, 1–3% among persons of Kurdish origin and 11–13% among the general population. Also depending on whether continuous or categorical anthropometric measures were used, age, sex and anthropometrics explained 13–19% of variation in fasting blood glucose among persons of Russian origin, 15–20% among persons of Somali origin, 13–17% among persons of Kurdish origin and 16–17% among the general population. With exception for BMI, strength of the association between continuous anthropometric measures and HbA1c was significantly lower among persons of Kurdish origin compared with the general Finnish population (p = 0.044 for WC and p = 0.040 for WHtR). Conclusions A low degree of association between anthropometric measures and HbA1c was observed among persons of Kurdish origin. Findings of this study suggest caution is warranted when using HbA1c as a screening tool for glucose impairment among persons without diabetes in populations of diverse origin.

Keywords

• HbA1c
• Glucose
• Ethnicity
• Migrant
• BMI
• Waist circumference