Type 2 diabetes mellitus (DM) is one of the most common chronic diseases globally, with worldwide prevalence of 8.3%. Due to its long-lasting nature and high risk of complications, the burden of type 2 DM is expected to rise. Patients with type 2 DM have an estimated two-to-six fold higher risk of developing cardiovascular disease (CVD) compared to the general population. Moreover, CVD is considered the leading cause of morbidity and premature mortality in type 2 diabetic patients. CVD risk assessment tools in general are mathematical models or charts used to estimate the risk of a CVD event in an individual. CVD risk estimation is important to plan the initiation of preventive and therapeutic measures for CVD prevention including anti-lipid, anti-hypertensive and anti-platelet therapies, as well as to plan appropriate health education. Various professional guidelines for the management of type 2 DM have advocated the use of CVD risk assessment tools to estimate CVD risk among type 2 diabetic patients using traditional CVD risk factors such as hypertension (HTN), dyslipidemia, high glycosylated hemoglobin (HbA1c), albuminuria, obesity, smoking status, and family history of CVD. However, most of the existing CVD risk assessment tools were derived from Western populations, with very few developed for East Asian populations. In Oman, as in other Arabian Gulf countries, type 2 DM represents a high public health burden. Related data in Oman have shown a gradual increase in the prevalence of DM from 10% in 1991 to 12.3% in 2008. As for CVD, local data collected from the Omani general population indicate that CVD accounted for 29.8% of total causes of death in 2013 according to the Ministry of Health. However, there is very limited literature available related to CVD occurrence and the distribution of its risk factors among Omani type 2 diabetic patients. As for CVD risk assessment models used in prevention and management of CVD among diabetics, no CVD risk assessment tool has yet been developed for any Arabian population, including Omanis. Despite the availability of international risk assessment tools, these are not considered optimal for Omani diabetics. As populations differ in many ways such as differences in lifestyle patterns, socio-demographic characteristics and trends in the incidence of diabetes and CVD risk factors, the existing international CVD risk assessment models are not suitable for Omani or even Arabian diabetic populations. Hence, the overall aim of this project was to develop a risk assessment tool that is suitable to estimate the 5-year CVD risk among Omanis with type 2 diabetes. To achieve the main aim of this research project, three subsequent studies were conducted. The objective of the first study was to assess the incidence of CVD and the patterns of traditional CVD risk factors among Omani type 2 diabetics. A retrospective cohort study was undertaken on a sample of 2,039 patients with type 2 DM selected from four primary healthcare institutions in Aldakhiliyah Governorate (Province), all of whom were free of CVD at baseline in 2009 – 2010. Socio-demographic and baseline data regarding traditional risk factors were retrieved from the patients' medical records. The CVD outcome was defined as the first confirmed diagnoses of coronary heart disease (CHD), stroke or peripheral arterial disease (PAD) during the study period, up until December 2015, with a mean and median follow-up period of 5.3 and 5.6 years respectively. This study revealed an overall cumulative CVD incidence of 9.4% among Omani diabetic patients, with an incidence density of 17.6 cases per 1,000 person-years. A high prevalence of most CVD risk factors was observed among the study sample. For example, the prevalence of poor glycemic control (as indicated by a high level of HbA1c), HTN, obesity, dyslipidemia, and albuminuria was 40%, 56.3%, 39%, 77.3% and 18.7% respectively. A univariate survival analysis showed a significant association between CVD and the following factors: age; DM duration; body mass index (BMI); glycemic control; HTN; total serum cholesterol and albuminuria (P value < 0.05 each). In addition, compared to similar global studies, important differences in the prevalence of some risk factors and their patterns in the univariate associations with CVD outcome were observed. The second study aimed to develop a suitable CVD risk prediction tool for Omanis with type 2 diabetes, in consideration of the specific patterns of CVD risk factors in this population. This study was conducted based on the same study sample used in the first study. However, patients with incomplete data related to any key risk factor were excluded in the derivation of the model using a Cox regression analysis. As such, a total sample of 1,314 patients with complete data was used to develop the model. All included patients were free of CVD at baseline (2009-2010) and were followed up until any of the following end-point events occurred: their first CVD event (either CHD, stroke, or PAD), death, or the end of the follow-up period in December 2015. All data were retrieved from the diabetes registry and the patients’ computerised files at the primary care settings. Among the study sample, 192 CVD events were recorded within a mean follow-up period of 5.3 years. This study modelled the 5-year probability of CVD as: 1 - 0.9991ExpΣXiBi, where Exp ΣXiBi (which represents the hazard ratio) = Exp (0.038 age [years] + 0.052 DM duration [years] + 0.102 HbA1c [%] + 0.201 total cholesterol [mmol/l] + 0.912 albuminuria [coded 1 if present] + 0.166 HTN [coded 1 if present] + 0.005 BMI [kg/m2]) The aim of the third study was to validate the model developed in the second study. The performance of the model was assessed in two samples: the derivation sample used to develop the model, which consisted of the 1,314 diabetics described previously, and another separate validation sample selected from two institutions in the same region. This validation sample included 405 type 2 diabetics which were not included in the model derivation. All patients were free of CVD at baseline (2009-2010). All of the end-point events for the validation sample were defined as for the derivation sample. All data were retrieved from the patients’ medical records. This study showed adequate model discrimination in both the derivation and validation samples, with an area under the receiver operating curve of 0.73 (95% confidence interval [CI]; 0.69 – 0.77) and 0.70 (95% CI: 0.59 – 0.75) respectively. The calibration of the model also showed acceptable results, with insignificant differences between the mean predicted risks (estimated by the model) and the actual mean risks, with differences ranging from 0.7% - 3.1% and 0.1% - 4.2% (P value > 0.05 each) in the derivation and validation samples respectively. In addition, the recommended optimal CVD risk cut-off point was 10.0%, yielding good sensitivity (73.0%) and reasonable specificity (60.3%). This research project revealed the limited applicability of existing international CVD risk assessment tools in Oman, and the need to develop a specific tool suitable to estimate CVD risk among Omani type 2 diabetic patients. Subsequently, a CVD risk assessment model for people with type 2 diabetes in Oman was developed in view of the specific risk factor profile of this population. The model was validated in both the study sample and an external sample, and was shown to be suitable for the Omani type 2 diabetic population. Therefore, the present model is considered a suitable tool to estimate CVD risk at least for type 2 diabetic patients in Aldakhilyah Province, in order to plan their clinical management strategies and CVD prevention measures. In addition, health planners may use this model to monitor CVD risk and estimate the future burden of CVD among Omani diabetics. However, the wider generalisability of this model requires further validation studies in different provinces of Oman, as well in neighboring Gulf countries. Nevertheless, the use of the present model in clinical settings would allow further validation of the model over time and enable researchers to assess the cost-effectiveness of utilising this model among Omanis with type 2 diabetes.

Type 2 diabetes mellitus (DM) is one of the most common chronic diseases globally, with worldwide prevalence of 8.3%. Due to its long-lasting nature and high risk of complications, the burden of type 2 DM is expected to rise. Patients with type 2 DM have an estimated two-to-six fold higher risk o...