Association between arsenic exposure and diabetes in Ronpiboon District Nakhon Si Thammarat Thailand : case control research

Abstract

Background: Diabetes mellitus (DM) is a chronic metabolic disease resulting from diminished insulin production by pancreas (type 1) or the ineffective use of insulin by the body (type 2). Known risk factors for DM, especially DM type2, include older age, obesity; unbalanced diet, physical inactivity, stress, family history, and genetic polymorphisms. Chronic arsenic exposure at high level was considered as additional DM risk factors, but inconclusive epidemiological results still exist. Contamination of arsenic in environment had been found since 1987 in Ronphiboon district, Nakhon Si Thammarat Province, Thailand. The increased rates of DM patients in that area also led us to the investigation of DM risk factors, to add more information for DM risk mitigation. Thus, this study focused on investigation of determinants of DM type2 risk among residents of 3 Moo Ban (villages) of Ronphiboon sub-district, Nakhon Si Thammarat Province, Thailand. Methods: This unmatched and matched case-control studies aimed to compare the socioeconomic as well as low dose arsenic exposure patterns between villagers with DM Type 2 (Cases, N=185) and those who had not been diagnosed with DM (controls, N=200 for unmatched; N=185 for matched). The data used were based on previous community-based studies in 2000 and 2008. The technique of Multiple Imputation (MI), with the Predictive Mean Matching (PMM, an imputation method used to prevent negative value after MI) was used to impute missing values for independent variables. The stepwise modelling was constructed to investigate the influence of socio-economic background and arsenic-related independent variables on DM risk. For fully imputed two data set of cases-unmatched controls and cases-matched controls, multiple logistic regression and conditional logistic (cox model) were respectively used to assess associations. Results: BMI (p=<0.001, 0.007), age (p=0.003, unmatched), and history of sibling illness (p=0.021, 0.031), drinking (p=0.002, matched) were statistically significantly associated with increased risk of DM type 2, whereas having motorcar (representing better economic status, p= 0.020, 0.010), exercise (p=0.051, 0.027) were associated with lower DM type2 risk in the unmatched and matched case-control, respectively. We did not observe convincing association of water arsenic concentration with diabetes risk in both unmatched and matched controls studies. Conclusions: Our findings on sociodemographic information of both unmatched and matched case control studies, confirm that older age, BMI, having history of illness in siblings were the determinants for increased DM type2, whereas having better economic status, exercise were associated with lower DM type2 risk in this area. Our analysis suggested no association between water arsenic concentration and DM type2 risk, though a limit inconsistent association was identified for the use evidence of rain water year 2008. Further research is needed on this topic.