Identifying young adults at high risk of prediabetes and diabetes in Malta : a cross-sectional survey

Abstract

The aim of this study was to quantify prediabetes risk and associated risk factors amongst young adults. We also sought to assess the validity and reliability of the risk assessment tool used. Fasting plasma glucose was also collected to further assess risk. A cross-sectional, correlational, study was conducted. Convenience sampling was used with a target sample size of 374. Out of a total of 14,483 eligible participants, 176 (response rate = 1.22%) young adults aged 18-35 attending two higher education institutions had their data collected from the 22nd of December, 2020 up until the 30th of April, 2021. An online questionnaire was used, including fasting plasma glucose tests for 57 participants. Analysis was carried out using the IBM® SPSS® Statistics Version 27.

The Kolmogorov-Smirnov and Shapiro-Wilk test, Mann- Whitney U test, Pearson and Spearman correlation,

Fisher’s exact test, Univariate General Linear Model and the receiver operating characteristic analysis were all used to analyse the data. 5.3% (n=3) of participants, who

were all male, obtained an abnormal FPG value (≥5.6 mmol/L), signifying prediabetes and one case of diabetes (8.57 mmol/L). Significant risk factors for prediabetes were sibling history of diabetes, high blood pressure, waist circumference and smoking. The area under the ROC curve of the German Diabetes Risk Score resulted in 0.787, with a sensitivity of 66.6% and a specificity of 78.0% when scoring 30 points. The German Diabetes Risk Score, while showing good performance, was not statistically significant in this population and not useful in identifying prediabetes/diabetes in local young adults. This may have been due to low response rate and small sample size, leading to a lack of representativeness of the larger population, which may affect the generalisability of the findings. The utilisation of fasting plasma glucose to identify such subjects appears to be superior and further diverse studies are needed to validate the findings and investigate the feasibility of wide-scale screening with refined prediabetes risk assessment tools.