Geographical variations of goitre prevalence in Sri Lanka: visualisation with geographic information systems

Abstract

INTRODUCTION: Widely held beliefs about the geographical distribution of goiter prevalence in Sri Lanka are not based on reliable data. Geographic Information Systems provide an excellent means to visualize and present epidemiological data. AIMS: To map and explore the geographical variation of goiter prevalence in Sri Lanka METHODS: An island-wide representative sample of residents (n=5200) from 95 Gramasewa Niladhari (GN) divisions in 87 Divisional Secretariats (DS) were screened for goiter. A subsample was investigated for thyroiditis. Digital maps were developed to visualize the goiter and thyroiditis prevalence across the country. Maps were developed for the GN and DS level representations. Based on the data from 95 GN divisions, goiter prevalence was estimated for the un-sampled areas of the country using the interpolation techniques. ArcGIS 10.0 software was used for analysis. RESULTS: A total of 426 goiters and 177 thyroiditis cases were identified. The male and female goiter prevalence showed similar patterns of geographical variation. The DS_.divisions with goiter prevalence more than 15% were seen distributed across the country without any discernible pattern. Distribution of thyroiditis exhibited a similar pattern. There was some evidence of clustering together of areas/ GN divisions with a similar prevalence of goiter but it was not possible to demarcate large contiguous areas ('zones'] with a similar prevalence of goiter. CONCLUSIONS: Geographical distribution of goiter prevalence did not show any large regional or zonal variations. Areas of high and low prevalence were distributed randomly across the country.