The burden of type 2 diabetes mellitus, dysglycaemia and their co-determinants in the adult population of Malta

Abstract

The main aim was to determine the burden of diabetes mellitus, dysglycaemia and their co-determinants within the adult population of Malta. Furthermore, specific objectives included an exploration of the Maltese co-determinants of T2DM including links between different anthropometric, biochemical, and socio-demographic factors as well as between ten specific genetic SNPs and T2DM. This was aimed to provide the required evidence to empower public health efforts to target prevention as well as to develop nation-wide policies and strategies. Methodology The cross-sectional study’s target population was adults residing in Malta for at least 6 months aged between 18 and 70 years. The study population was selected from a national registry. A randomized stratified single stage sampling method was conducted to establish the study population. The strata for selection were age, gender and locality. Considering a possible 50% response rate and an expected pre-diabetes prevalence rate of 25% (based on published literature), the PiFace software® was used to estimate the sample size for this study. A sample of 4,000 adults was required. Permissions to conduct this study were granted from the University of Malta research ethics committee, the information and data protection commissioner, the Ministry for health, the chairman of the pathology department, the chief executive officer of Mater Dei Hospital and the laboratory of molecular genetics. A validated questionnaire and validated tools for health examination measures were utilized based on the European Health Examination Survey guidelines. A health examination hub was set up every weekend at governmental peripheral health clinics across all of the Maltese towns. In order to reduce information bias, a limited number of fieldworkers were enrolled, and trained regularly. Invitations to the randomly selected participants were sent offering a free health examination, two weeks prior to the examination appointment. Participants gave their informed consent and answered the socio-demographic questionnaire. This was followed by measurements for blood pressure, weight, height, waist and hip circumference. Blood samples for fasting plasma glucose (FPG), lipid profile and a whole blood sample for genetic studies were drawn as the last stage of the examination. An oral glucose tolerance test was offered to those obtaining an impaired fasting plasma glucose (IFG) result. All the data gathered during the fieldwork was inputted by a single fieldworker to avoid bias. Secure inputting software was used that was programmed to perform data validation while inputting data. In order to compensate for non-respondents and maintain strata representation, a weighting factor was applied to each individual in the sample using the IBM SPSS software. The weighting data was only used when national representative population analysis was performed. Prevalence rates (T2DM, IFG, overweight-obesity; hypertension and the metabolic syndrome) were established for each category of age and gender. Socio-demographic, anthropometric and biochemical parameters were analysed (descriptive and analytic) and associated links were investigated with T2DM and IFG by using the IBM SPSS software. Non-parametric statistical testing using the Mann-Whitney U test and the Kruskal Wallis test were performed since the data did not follow a normal distribution. Dunn’s test was used as a post-hoc test following Kruskal Wallis testing. The Chi-squared test was used to identify significance between categorical variables. The Spearman’s correlation testing was performed to test for associations between variables. Binary logistic regressions and multiple regression analysis were performed to identify the independent associated risk factors for T2DM and IFG. Using regression analysis and receiver operating curves, a Maltese specific diabetes risk score was established. The cost burden for T2DM and obesity was calculated based on cost per case rates obtained from the scientific literature, after adjusting for gross domestic product (GDP) per capita and for deflation. A 2% compound interest per annum was added on the cost burden obtained for obesity from local data. A sub-population of the participating study population was randomly selected from within each different metabolic profile category (dysglycaemic, metabolically abnormal and metabolically normal) to undergo case-control genetic analysis. DNA extraction from whole blood samples gathered during the fieldwork, followed by real time PCR genotyping for ten identified literature based single nucleotide polymorphism (SNPs) was performed. Descriptive and analytic analyses were performed using IBM SPSS software. A case-control design was followed to evaluate this sub-population’s biochemical and anthropometric phenotype in relation to the 10 SNPs under study. Multiple regression analysis was performed to identify any associated links between the 10 SNPs and a diagnosis of T2DM.