Alterations of arsenic levels in arsenicosis residents and awareness of its risk factors: A population-based 20-year follow-up study in a unique coal-borne arsenicosis County in Guizhou, China

Currently, most arsenic (As) studies in populations are concerned with water-borne arsenicosis. However, residents in Xingren County of Guizhou Province, Southwest of China, represent a unique case of arsenicosis which is related to indoor combustion of high As-containing coal. This study aimed to assess the alterations of As levels and its risk factors in coal-borne arsenicosis residents during the past 20 years.Four follow-up investigations in Xingren County were selected from the year 1998 to 2017, a total of 245, 272, 584, and 309 residents were involved in the four investigations, respectively. Local external environmental medium (coal, soil, water, air, rice, corn and chili peppers) and biological samples (urine, hair) were collected at each time of investigation for total As analysis. Sociodemographics and lifestyles variables were extracted from the questionnaire investigation. Both univariate and multivariate unconditional logistic regression models were performed to analyze the variation of risk factors for coal-borne arsenicosis.A substantial reduction of total As levels was observed both in external environmental medium and biological samples in the unique coal-borne arsenicosis region, especially since the year 2006. In addition, age, duration of consuming high As-containing coal and smoking status were found to be the most significant risk factors for coal-borne arsenicosis during the past 20 years by both two different logistic regression models. Room ventilation and grain drying modes were no longer to be risk factors since 1998 survey. Annual household income had always been an important protective factor for coal-borne arsenicosis in recent 20 years by both two different logistic regression models. Grain storage modes had become significant protective factor in 2014 and 2017 survey. A certain correlation between sex, education and coal-borne arsenicosis was observed by univariate logistic regression model but no clear links were found by multivariate logistic regression model.Considerable efforts to blocking As exposure from burning coal and As contaminated foods in this region are observed over the study period. Further practical health education programs may need to target individuals with long-term of As exposure, lower socioeconomic status and smoking in order to better prevent and control the occurrence and development of coal-borne arsenicosis.