1887

Abstract

Objectives: Chronic diseases such as diabetes, obesity and cancer are caused by the complex interaction between environmental factors (such as diet, lifestyle, and the built environment) and genetic factors. To understand the ultimate role of environmental, behavioral, and genetic factors along with their interactions, large-scale population cohorts have been established, mainly in Europe, North America, China, Japan, and Korea. The Qatar Biobank is a Qatar national population based prospective cohort study which includes the collection of biological samples, with long-term storage of data and samples for future research. The ultimate goal is to allow physicians and researchers to use the data collected from the biobank to conduct a large-scale study of the combined effects of genes, environment, and lifestyle on these diseases, to educate people on risk factors for these common diseases and to study disease incidence patterns and develop new diagnostic and therapeutic approaches. Using this pilot data, we had access to 60 features measured on 1000 Qatari citizen. To the best of our knowledge, this is the first study that has been done on Qatari biobank few months after its release. The main objective of the study is to identify the associated risk factors in Qatari population compared to those previously found in other parts of the world. Methods: In this study, we apply a panorama of state-of-the-art statistical methods and machine learning algorithms to investigate risk factors for diabetes. The statistical methods rely on lasso and group-lasso based techniques that can even use mixed continuous and categorical variables. The machine learning methods rely on tree-based models that provide importance of variables in predictions. In contrast to relying solely on the widely used baseline statistics, which perform marginal analysis considering a single variable at a time, these methods are based on multivariate analysis of the medical conditions. Moreover, we have applied survival and risk analysis on the prognosis of diabetes in the Qatari population. In our analysis, we used survival analysis to estimate the distribution of time of diabetes development. We have used Cox proportional hazards model to investigate the effect of different variables on the risk of diabetes. Results: Our study strongly confirms known risk factors associated with diabetes in Qatari population as previously found in other population studies in different parts of the world. For diabetes, biomarkers in Qatari population (as identified by different methods) include magnesium, calcium, HDL-C, chloride, insulin, c-peptide of insulin which have been previously reported by to list a few. Our study has revealed interactions of hypomagnesemia with HDL-C, triglycerides, and free thyroxine. These findings need further investigations. The survival analysis reveals that at the age of 40, there are 15% chances of developing diabetes in Qatari population and the chances increase to 50% at the age of 63. Qatari females are slightly at more risk to diabetes than males before the age of 40 but later on males have more chances to develop diabetes. The risk analysis reveals that calcium, magnesium, hemoglobin, triglycerides, and free-triiodothrymine play a very significant role in determining risk of the disease in Qatari population. Conclusion: Our study strongly confirms known risk factors associated with diabetes and obesity in Qatari population as previously found in other population studies in different parts of the world. Moreover, interactions of hypomagnesemia with other risk factors merit further investigations.

Loading

Article metrics loading...

/content/papers/10.5339/qfarc.2018.HBPD821
2018-03-15
2019-12-12
Loading full text...

Full text loading...

http://instance.metastore.ingenta.com/content/papers/10.5339/qfarc.2018.HBPD821
Loading
This is a required field
Please enter a valid email address
Approval was a Success
Invalid data
An Error Occurred
Approval was partially successful, following selected items could not be processed due to error