Visualization and Analysis of Epidemical Data of Tumkur District Hospital using GIS.
BG Premasudha1, Shivakumarswamy2 and BS Adiga3
1Research Scholar, Dr. MGR University, Chennai.
2 Mallige College of Pharmacy, Bangalore
3 TCS, Bangalore
*Corresponding Author E-mail: bgpremasudha@gmail.com
ABSTRACT:
Disease data analysis and sharing is important for the collaborative preparation, response, and recovery stages of disease control. Disease phenomena are strongly associated with spatial factors. Web-based Geographical Information Systems (WGIS) provide a real-time and dynamic way to represent disease information on maps. WGIS provides excellent means for visualizing and analyzing epidemiological data, revealing trends, dependencies and inter-relationships. GIS serves as a common platform for convergence of multi-disease surveillance activities. This paper focuses on presenting a WGIS application created for studying distribution of malaria patients in Tumkur city, India. The application covers two main epidemiological issues: (i) defining the spatial distribution of malaria patients; and (ii) modelling spatial variation of malaria in the city. This study is useful to find out patient distribution, patient data classifications and accessibility to hospital. The spatial data infrastructure can enhance the efficiency and effectiveness of public health surveillance for India in district levels.
KEYWORDS: Web-GIS, Thematic mapping, surveillance, epidemics.
INTRODUCTION:
Currently, the factors such as increasing population, environmental pollution, rapid Urbanization and global warming all influence the conditions for disease outbreaks in India. Disease studies have revealed strong spatial aspects, including disease case location and disease diffusion. Thus, mapping spatial aspects of diseases could help people understand some puzzles of disease outbreak (Gupta R, Shriram R, 2004). Unlike the raw disease data, disease maps offer a visual means of identifying cause and effect relationships existing between humans and their environment. Disease maps can enable decision makers to make their work easier.
Geographical Information Systems (GIS) provide an effective way of managing, storing, analyzing, and mapping disease information. GIS has strong capabilities in mapping and analyzing spatial and non-spatial data, GIS could represent disease information rapidly and analyze the disease's spread dynamically.
Meanwhile, the rapid development of the Internet influences the popularity of Web based GIS, which itself shows great potential for the sharing of disease information through distributed networks. Distributing and sharing disease maps via the Web could help decision makers across health jurisdictions and authorities collaborate in preventing, controlling and responding to a specific disease outbreak. Spatial epidemiology is considered one of the main GIS uses in healthcare research (Lang, Laura, 1999). Several worldwide cases are presented for the purpose of identifying the potentials and benefits of using GIS in epidemiology (Boulos MN, 2004). The relevant GIS functions and tools are identified and divided into data-exploratory and data-modelling functions. This paper covers two main epidemiological issues: (i) defining the spatial distribution of malaria patients and (ii) modelling spatial variation of malaria in the city. In addition, a WGIS geostatistical analysis technique is used to model the spread of malaria patients at Tumkur city. WGIS is an application for map representation of the spatial data via the Internet (Benneyan JC, et.al, 2000). The WGIS provides us the interactive mapping analysis which enables the user to visualize predefined maps of selected indicators in order to study the spatial distribution of a phenomenon for
Figure 1: Web Display of Tumkur City Map.
information purposes. This helps the user to view and control the map features directly through the internet. The Figure 1 shows the Web representation of Tumkur city map.
Tumkur is one of 57 cities in Karnataka. Tumkur is the headquarters of the district of the same name. It is located to the North West of Bangalore at a distance of 70km.City is in close proximity to Bangalore and has a decadal growth of about 40% over the last few decades. More than 15 Districts should pass through Tumkur to reach Capital city Bangalore. The presented study has selected government District hospital. This hospital has a capacity of 300 beds and 70 doctors working at different departments including family medicine, gynecology and pediatric department. It is located almost at the centre of the city but expected to serve most parts of city districts.
District hospital collects data of patients from all the PHC’s once per week, organizes them and prepares the final patient’s data for analysis. In this study, we are representing the outbreak of Malaria at this hospital. We are concentrating only on malaria patients data of Six months from January/2008 to June/2008. The data for infectious disease mapping used in this study includes the above disease data and the geometric boundary data of Tumkur city. The infectious disease data for Tumkur are represented by the hospital discharge data records at Tumkur District Hospital.
In present scenario finding out the outbreaks of disease in hospitals needs more manpower and consumes more time. For decision makers, they need to collect all patients data from all the hospitals, sort them out to find out factors such as, which disease is out breaking currently?, what may be the reason for the outbreak?, what action should be taken to overcome that problem?. This process is very much time consuming. Hence the WGIS helps us to predict the factors
such as catchments of patient’s location towards the infectious diseases. We can also categorize patients
according to their area, sex, age etc; it also provides thematic shading of maps which helps us in analysis. Web-GIS make the online analysis more interactive in right time.
GIS has several techniques and functions that can be used for health service planning. Each one of these functions can be applied on different health related issues. This study has selected two major hospital planning issues and uses GIS for analyzing these issues. The first issue is related to defining the spatial distribution of malaria patients by using on-screen digitization. This is used by the presented study to capture and define health demand location at Tumkur city. The geometric boundary data of Tumkur city has been collected from the Municipal Corporation of Tumkur city. The collected data included the ward wise population data, road maps and the hospital information. The road map and the ward map was geo-referenced using the survey of India Toposheet. The maps were then digitized using Mapinfo. Then a point feature layer was created for the hospitals and patients. All these data are linked to the demand coverage and used for the second issue of this application which is related to classification of spatial variations of malaria locations by knowing distribution of patients in city by using overlay analysis. Overlay analysis manipulates the spatial data organized in different layers to create combined spatial features according to logical condition specified in Boolean Algebra.(Roovali, R. Kiivet,1997).The logical conditions are specified with operands (data elements) and operators (relationships among data elements). The most well used overall functions are called union, intersect and identity. This overlay function creates a new output coverage with the display of thematic maps, that has city wards that falls inside hospital service area and patient data. The patient’s data are classified into different groups according to their age and sex to present it on thematic mapping and graphs to find out disease outbreak of malaria in Tumkur city and this data is visualized on internet web pages by using MapXtreme java(6).
Health facilities in general and hospitals in particular are faced with different challenges related to their locations, their market service areas and their demand status (L. Roovali, R. Kiivet. 1997). This part of the paper presents a GIS application that is created for District Hospital of Tumkur. The application is designed to be as a spatial decision support system for defining the spatial distribution of malaria patients. Using GIS in health care planning studies is well acknowledged by the western European researchers and it is used for various health care issues at the developed countries. However, in India this technology is still not very well explored by health authorities and researchers. Therefore, the created application provides a good example for explaining how to use GIS by health planners and officers in Tumkur and/or in any other developing cities.
The Tumkur district hospital has a database about its existing patient and saves such data in different Management Information Systems (MIS). These systems are used for finding needed information about patient number or recording file and for reviewing the medical history of every patient. One of the main issues related to spatial distribution is regarding defining patients location within the city. A GIS function called Geocoding can be used to create points features on a map from a table having x, y coordinates of any addresses. The presented study has used on-Screen digitizing method for the purpose of identifying hospital and patient location. Based on the collected data, hospital and patient data GIS coverage is created with their geographical locations and then the attributes data of health details are entered as records in the coverage table. After building the database of health details, the next step was to use GIS for identifying spatial distribution of malaria patients. This step is achieved by using the graduated color function that subdivides numerical data into a set of classes. The presented study has used the data classification in natural breaks method that minimizes the variance within class and maximizes the variance between classes (Roovali, R. Kiivet,1997). Figure 1 shows the resulted patient distribution for Tumkur District hospital.
One of the main results of this distribution is most of the malaria patients come from the ward number 12, where the city slum is situated. The less number of malaria patients are seen from high income residential areas and rural areas of the city.
All malaria patients data is divided into four groups. Figure 3 shows the thematic representation of male patients of age less than 5 years in which we can find that, there are more patients of this category in ward number 12 which is shaded in blue colour. Figure 4 shows the thematic representation of male patients of age greater than 5 years in which we can find that, there are more patients of this category in same ward. Figure 5 shows the thematic representation of female patients of age less than 5 years in which we can find that the outbreak of this category is also occurred from the same ward. Figure 6 shows the thematic representation of female patients of age greater than 5 years which also represents the outbreak in same ward.
Figure 2: Spatial distribution of the patients from Government District hospital of Tumkur city.
Figure 3: Ward wise Thematic map representing male patients of age less than 5 years.
Figure 4: Ward wise Thematic map representing male patients of age greater than 5 years.
Figure 5: Ward wise Thematic map representing female patients of age less than 5 years.
Figure 6: Ward wise Thematic map representing female patients of age greater than 5 years.
Figure 7: Graph representing ward wise patients data.
Figure 7 is representing the graph of malaria patients of all the above four groups. From this graph we can predict that there are 6 male patients of age less than 5 years, 11 male patients of age greater than 5 years, 4 female patients of age less than 5 years and 10 female patients of age greater than 5 years in ward number 12 which is considered as an outbreak in their particular categories respectively. Combining all the groups 31 patients were belong to ward number 12 which is considered as an outbreak area. The cause for the disease outbreak in that area may be evaluated further as future enhancement.
This paper discusses a GIS application for hospital facility planning in Tumkur city. The application covers two main hospital issues that are distribution of patients in city and spatial variation of malaria in the city. Each one of these issues has a direct spatial dimension. Therefore, the use of WGIS for analyzing and manipulating them was of greater value and benefit. WGIS is used to define all patients’ location and produces an output showing Malaria outbreak with spatial variation around the city. This output can be used by health planners to define the real catchments of health facilities. By studying the Figure 6 and 7 we can conclude that, the blue coloured area which is located at the centre of the city is the area where the malaria outbreak has occurred. This output is used for further studies to define the causes of the occurrences of malarial.
1. Benneyan JC, Satz D, Flowers, SH: Development of a Web-based multifacility healthcare surveillance information system. J.Healthc.Inf.Manag, 2000.
2. Boulos MN: Towards evidence-based, GIS-driven national spatial health information infrastructure and surveillance services in United Kingdom. International Journal of Health Geographics, 2004.
3. Gupta R, Shriram R: Disease surveillance and monitoring using GIS. 7th Annual international conference Map India, 2004.
4. H. Jordan, P. Roderick, D. Martin, S. Barnett, Distance, rurality, and the need for care: access to health services in South West England, Int. J. Health Geography. 2004.
5. http://reference.mapinfo.com/common/docs/mapxtreme-java_edition-3.1-readme-pdf-none-eng/readme.pdf
6. Lang, Laura, “GIS for health organization”, Environmental System Research Institute, Inc. 1999.
7. L. Roovali, R. Kiivet, Geographical variations in hospital use in Estonia, Health and Place 2006. Y. Chou, Exploring Spatial Analysis in Geographic Information Systems, On Word Press, Santa Fe, NM, 1997.
Received on 19.01.2010 Accepted on 11.02.2010
©A&V Publications all right reserved
Research J. Engineering and Tech. 1(1): Jan.-Mar. 2010 page 42-46