Using geographical information systems in epidemiology : mapping & analyzing occurrence of diarrhea in urban–residential area of Islamabad, Pakistan

Abstract: Almost 2 million people are living in Islamabad, significant number of workforce commute from neighboring areas and other parts of the country. The city is facing restraint on natural resources due to increased urban migration coupled with rapid population growth. As a result, the city infrastructure facilities and living conditions are deteriorating. The problems associated with drinking water availability and quality is rising along with other challenges including poor hygiene conditions and low income areas. A significant portion of population living in Islamabad is vulnerable to diarrhea epidemic associated with contamination of water, which has become a major public health problem in recent years. Selected residential sectors were examined for the mapping of diarrhea incidences rising during a complete year (2013). Relevant data on district boundary, residential sectors, Streets and Highways, Water Filtration Plants (WFP), Water Drainage Streams (WDS), Slums and Non-slum was received and prepared according to research requirements using Geographical Information System (GIS). All the data were digitized using point shapefiles, polyline shapefiles and polygon shapefiles during data preparation. Global and focused test were conducted for cluster detection based on covariate factors illustrated using GIS. Hypothesis testing was conducted using global test (Average Nearest Neighbor) on diarrhea patients, the results indicated the spread of diarrhea had strong evidence of pattern. The study further explored, whether the disease spread is random or has some covariate factors influencing the spread. Henceforth, focused test was conducted using the location of WFP, WDS, Slums and Non-Slum to understand the role of those factors for the spatial dispersion of the disease using Average Nearest Neighbor. Diarrhea Incidence Rate (DIR) for each factor was found for summer and winter season to understand the effect of season/ temperature. In addition to that, KD maps were used for the detection of diarrhea hotspots visually. Finally, hotspot maps were produced for summer and winter through Getis-Ord GI' statistics method using DIR to show vulnerable areas in Islamabad with diarrhea Hotspots & Coldspots. The results clearly demonstrated: the increase in distance of a household from WDS decreases diarrhea risk in summer and winter; the increase in distance of a household from WFP increases diarrhea risk in summer but the same is not observed for winter; increased diarrhea risk is observed in Slum during summer and winter, and increased diarrhea risk is observed during summer but the same is not observed for winter in Non-Slum. The overall research indicated Geographical Information System’s ability to highlight the public health problems.