Self-reported diabetes mellitus and education in NOWAC, a cross-sectional study

Abstract

Abstract Introduction: Diabetes mellitus is an increasingly public health concern, causing increased mortality, morbidity and healthcare costs globally. Global increase in overweight and obesity, high calorie diets, sedentary lifestyle, smoking habits and aging population are the main drivers for this epidemic. Social inequalities reveal systematic differences in health and disease. Social inequalities are increasing in Norway, especially among women. Persons with lower socio-economic position are more often exposed to negative lifestyle behavior. Education are one of several measures for SEP used in epidemiological studies. This study aims to investigate the association between educational levels and diabetes among Norwegian female population. Material and methods: This study used data from the Norwegian Women and Cancer cohort that included participants between 30-70 years of age, recruited in waves in 1996, 1997, 1998, 2003 and 2004 (N=100 347). Questionnaires collected at enrolment was used. Complete-case analysis was performed, where 21.1% were excluded due to missing information, resulting in N=79 221 eligible participants. Diabetes was self-reported, and education level grouped (“primary” <9 years, “secondary” 10-12 years of education, and “tertiary” >13 years of education). Binary logistic regression analysis was performed with self-reported diabetes as the dependent variable, and education as the main independent variable with “primary as reference category. The logistic regression analyses were performed with three models: A=univariable, B= age-adjusted and C=multivariable (adjusted for: age, BMI, physical activity, smoking status, alcohol consumption, CVD and diet (total intake of fiber, vegetables, fruit and total sugar)). Results: In the unadjusted model A, the results showed that participants with tertiary education had lower odds of being diagnosed with diabetes compared to those with primary education: OR:0.49 (95% CI: 0.43-0.57), similar is seen for secondary compared to primary education: OR:0.69 (95% CI: 0.6-0.8). After age-adjustments, those with tertiary education had lower odds of being diagnosed with diabetes compared to those with primary education OR:0.65 (95% CI: 0.56-0.76), secondary compared to primary gave an OR of 0.86 (95% CI: 0.74-1.0). After adjusting for covariates, none of the results were significant; tertiary compared to primary OR: 0.91(95% CI:0.78-1.08), and secondary compared to primary OR: 1.02 (95% CI: 0.87-1.19). Thus, the results showed a strong interference of covariates. Conclusion: This study found an educational gradient in self-reported diabetes within the Norwegian female population, but the association was strongly confounded by different factors, suggesting that educational level does not predict diabetes. Educational is assumed to be a proxy for lifestyle factors, which predict diabetes. The inconsistency regarding diabetes prevalence induces the urgent need for a diabetes register in Norway. Keywords: Diabetes, education, socio-economic position