# Introduction e live in an age of information, and spatial information is one of the most critical elements underpinning decision making for a range of disciplines (Rajabifard and Williamson, 2001). Health, wealth and population distributions are all examples of spatial information commonly attached to administrative polygons. In fact, there are few areas of the economy and environment that do not rely either directly or indirectly on the integration of data attached to administrative boundaries for planning, maintaining or rationalizing activities (Eagleson et al., 2001). GIS has the ability to examine the complex behavior of geographically referenced data. Today, with increased spatial and network analysis capabilities, GIS are designed to predict and understand the interactions between entities and phenomena throughout space and over time (Tomlinson, 2002). Since the 1980s the benefits of geographic data and GIS for analysis and modeling have been realized. As a result, virtually all people, property and infrastructure have the ability to be referenced by location (Openshaw, 2000). With geographic data readily collected and GIS technology now available at a relatively low cost, it should be possible to build a digital representation of virtually any phenomena of interest. Related technical issues such as data exchange, differences in geographic boundary design and data integration present serious limitations, however, and must be addressed if geographic data is to be used to its full potential. Health is vital for all of us and understanding the determinants of a disease, its spread from person to person and community to community has become increasingly global (ESRI, 1999). As expressed by (Scholten and De Lepper, 1991), ''health and ill-health are affected by a variety of life-style and environmental factors, including where people live". There are various factors such as climate, environment, water quality and management, education, air pollution, natural disasters, social and many others which are the reasons for the emergence of diseases. The characteristics of these locations (including socio-demographic and environmental exposure) offer a valuable source for epidemiological research studies on health and the environment. Epidemiological research ranges from outbreak investigation, data collection, design and analysis including the development of statistical models. Since health is a geographical phenomenon and various factors attributing to the health diagnostics and planning are geography dependent, as such, GIS (Geographic Information System) for health studies serves as an important tool. Therefore, this project focuses on mapping health facilities and to show their spatial distribution in Gulele sub-city and finds gaps on which woreda health service provision is needed. This project utilizes different methods of data collection, and the sources of information were both primary and secondary data. The primary data collected through Garmin 76 GPS were, the spatial distribution of health facilities (pharmacies, hospitals, clinics and health posts). Secondary data were population data from the Central Statistical Agency (CSA), these data were combined to form a database for this project. # a) Data Sources and Methodology of the Study # c) Data Collection During the present study, both spatial and non spatial data were gathered, from primary and secondary sources. Primary data were generated from the analysis of field visits. A field survey was undertaken to record the geographic coordinates of all health facilities (government hospitals, private hospitals, government clinics, private clinics, government pharmacies, health centers, private pharmacies and government health centers) were recorded. The addresses of these facilities were obtained from the Ministry of Health and their spatial locations were gathered using GPS. In addition, the major primary data required for the study has been extracted from various road category net works were also generated from 1:50,000 scale topographic maps of Addis Ababa obtained from Ethiopian Mapping Agency through digitizing and the sub-city layer was obtained from Ethio-GIS. ? Field Visit Presenting the results using thematic maps and table addresses of the health facilities were generalized to maintain the confidentiality, it was often difficult to record the precise location. Various approaches were used to identify the location as close as possible to the actual location. These included checking the addresses by using Google Earth software prior to undertaking the fieldwork, Field visits were the major data collection mechanisms. Preliminary field visit was carried out to get an overall overview of the study area, to identify health facilities and to collect using Garmin 76 GPS readings of the various health facilities readings of the study area. # d) Data Analysis The data that was collected using Garmin 76 GPS, after recording these locations using GPS and data obtained from GPS were loaded into the ArcView 3.2 GIS software and later converted into the ArcGIS 9.3 format. All data were in ArcGIS 9.3 format. The data entered with were provided in Microsoft Excel format, then export to SPSS version 20.0 to change DBF format, after changing to DBF format, ArcGIS capable with this format, then with this software changed to shapefile. Since the data handled is large and diverse i.e. from different sources, the database is organized in a series of tables so that it can be shared. Each table is called a relation and it consists of a number of rows and columns. The method of analysis employed in this project was subjected to different softwares depending on the objective of analysis. For analysis, GIS data was vertically organized into layers or themes. Using database query, these basically involve basic retrieval of what is already in the database. In addition, point pattern analysis was conducted to evaluate the distribution of health facilities in the sub-city. Nearest neighbor statistics was calculated for the evaluation of the pattern of the distribution of the cases. Nearest neighbor ratio (R) calculates the distance between one point to its nearest points. It was calculated using equation ( 1) below (Robinson, 1998). ?? = ???????? ???????? (1) Where, R = nearest neighbor ratio ???????? = observed average distance between nearest neighbor ???????? = expected average for a hypothetical random distribution This index gives a systematic measure of the pattern within a specific region. R values range from 0 to 2.1491, where an R ratio value less than 1 indicates that the point pattern distribution is more clustered than random, the value of R is greater than 1 indicates that the point pattern distribution is more dispersed than random (Robinson, 1998). This analysis was undertaken with the ArcGIS9.3 software. # II. # Results and Discussion # a) Physical and Socio-Economic Background of Addis Ababa Addis Ababa, the capital of Ethiopia, which is by far the largest in the country located in the central part of Ethiopia and belongs to the Western highlands. All sides of the capital city is bordered by Oromiya Regional state, and covers an area of 530 square kilometers (53,000 ha) and a total population of 3,041,002 million persons (CSA, 2011). Addis Ababa has a history of 125 years. Thus, one can say with certainty that it had its earliest beginnings in the mid 1880s and had evolved to be the capital of Ethiopia around 1886/87 City Council of Addis Ababa, 1986(Johnson, 1974). Then after that changed the name to Addis Ababa City Municipality and also in 2005 called as Addis Ababa City Administration. Astronomically, Addis Ababa is located at 09 0 02' N Latitude and 38 0 44' E Longitude. It is situated on the foot of escarpment of Mt. Entoto in the North which rises to 2900 meters to the south with an average altitude of 2400 meters. This varying topography of the city has affected its spatial expansion favoring the relatively flat landscape in the south as a major factor contributing to the "unsafe housing" condition in the city, as cited in Hadgu (1988). Thus, males comprise 48 % and females 52 % of the total population of the sub-city. The sex ratio for Gulele sub-city equals 94 %. It is calculated as the number of males over the number of females. When the ratio is greater than hundred it indicates a higher number of males than females. Ratio closer or equal to hundred shows balanced number of males and females in a given woreda. The female population exceeds the number of male population by 8832 persons on average at woreda level. Higher number of females are recorded at woreda 09. Sex ratios greater than hundred are observed in woreda 06. The highest sex ratio, that is, 118%, is recorded in woreda 06. This map on Figure 5 shows the spatial location of all health facilities in the sub-city. Based on the spatial distribution of health facilities illustrated on the map, woreda 09 have consisted all health facilities (i.e., pharmacies, hospital, clinics and health center). But, woreda 04 have consisted only one clinic. The rest woredas of the sub-city health facilities have less sparsely distributed. The North West, north and north east part have outskirts of the sub-city have no health facilities, because these parts of the sub-city have steep slope and covered with forests. ii. Point pattern analysis of the Health Facilities Figure 6 : Results of point pattern analysis Point pattern analysis is used to evaluate the physical distribution of point events and test whether there is a significant clustering of points in a particular area. Health facilities distribution in Gulele sub-city were clustered, because, the ratio obtained was 0.76 (which is <1). Z score = -3.64 standard deviations; whereas the P-Value is 0.0000274. Thus, in Gulele sub-city health facilities were concentrated only in some woredas, for example, in woreda 09 all facilities are found. But, there were many cases have seen no built up of health facilities in the sub-city. # Volume XVI Issue III Version I # ( B ) f) Gulele sub-city Pharmacies Figure 7 : Gulele sub-city pharmacies Gulele sub-city has a total population of 267,624; this map on Figure 7 represents the spatial location of pharmacies in the sub-city. According to the spatial location of pharmacies, there are 16 pharmacies in the Sub-City. The distribution of pharmacies is not equitable because all of them except one are privately owned pharmacies and the owners opened the pharmacies on downtown where more market is available. The pharmacies mostly concentrated on woreda 09 but no pharmacies opened on woreda 04, 05 and 06. The above table 1 shows that, for each woreda the ratio shows how many people share a pharmacy on average. It is calculated as the number of people divided by the number of pharmacies. There is still a need for constructing new pharmacies for showing zero population sharing for the pharmacy. The average population of the sub-city is 26762 whereas at sub-city level average number of pharmacies is 1.6 thus for on average for 16726 people needs nearly 2 pharmacies. But, the distribution of pharmacies in the sub-city is not on the above proportion. On woreda 09 there is more concentration of pharmacies, that is 7 pharmacies. # g) Gulele sub-city Hospitals Figure 8 : Gulele sub-city hospitals Gulele sub-city has a total population of 267,624; this map shows that the spatial location of hospitals in the sub-city. In this sub-city there are two specialized government hospitals and one general private hospital. The first specialized hospital (Kidus Petros TV Center) is found in woreda 01. In addition, the other specialized hospital (Poulos hospital) is found in woreda 09. The private general hospital (Cure hospital) is found in woreda 02. The hospitals found in Gulele sub-city serve not only to Gulele sub-city but also for the country as a whole. Out of six specialized hospitals found in Addis Ababa city Administration two of them are found in Gulele sub-city. The proportion of population expected to serve in specialized hospital is 1:5,000,000 population means that 1 specialized hospital expected to serve for 5,000,000 people. Whereas, for general hospital is 1:1,000,000 population means that 1 general hospital expected to serve for 1,000,000 people. Thus, it is enough general and hospitals found in The above table 2 shows distribution of clinics in the ten woredas and one clinic supposed gives services. Accordingly, the average population of the sub-city is 26762 and clinics are 4, thus, averagely 6690 people share 4 pharmacies. But the distribution is very sparse in woreda 02, 04, 06, 07 and 08. The above table 3 shows that, for each woreda how many people share a health center. According to Ministry of Health on Health Sector Development Program IV (2011-2015), the proportion of health center is 1:25,000 population means that 1 health center expected to serve 25,000 people. But, the reality seen on the sub-city is different. Based on the spatial distribution, woreda 03 is only 1 health center for 28,124 people, woreda 06 is 1 health center for 26,123 people and woreda 09 is also 1 health center for 32,704 people. On the rest woredas there is not a health center. From this we conclude that in woredas where no health center, there is a need for built additional health center. # III. # Conclusions and Recommendations GIS is designed to predict and understand the interactions between entities and phenomena Health Facilities Distribution Mapping in Addis Ababa, Ethiopia throughout space and over time. This project mapped the distribution of health facilities. A field survey was undertaken to record the geographic coordinates of all health facilities and their spatial locations were gathered using GPS. The results showed that health facilities are sparsely distributed in most of the woredas. Planners and concerned government officials should be built extra health facilities based on the population ratio in Gulele sub-city. The necessary extra pharmacies, clinics and health centers should be built timely in order to solve the health facility problem which prevails in the study area. Analyses provide essential information for health practitioners to understand the problem and help them plan for additional health facilities distributed equally in the sub-city. IV. 1![Figure 1 : Overview of the Methodology b) Source and Nature of DataThis project utilizes different methods of data collection, and the sources of information were both primary and secondary data. The primary data collected through Garmin 76 GPS were, the spatial distribution of health facilities (pharmacies, hospitals, clinics and health posts). Secondary data were population data from the Central Statistical Agency (CSA), these data were combined to form a database for this project.](image-2.png "Figure 1 :") ![field survey was undertaken to record the geographic coordinates of health facilities. As the Data identification and collection Field Surveying using GPS Getting spatial data into the data base Making spatial data usable Getting attribute data into the data base Linking the database Performing analysis Data output](image-3.png "A") 23![Figure 2 : Gulele sub-city total populationAccording to the 2011 statistical abstract, Gulele sub-city has a total population of 267,624. This map portrays the number of inhabitants of each woreda. According to cartographic standards, dot density symbol is used. Each dot represents 2,000 people. The average woreda population is 26762. Woreda 04 is the least populated with 13,264 inhabitants, whereas woreda 07 is the most populated with 37,066 inhabitants.](image-4.png "Figure 2 :Figure 3 :") 45![Figure 4 : Gulele sub-city population density This map on Figure 4 shows the population density of Gulele sub-city. The average density of all woredas is 125 persons per hectare (ha). Woreda 01 and Woreda 10 of the sub city are less densely populated, whereas woreda 03, 04, 05 and 09 are densely populated.](image-5.png "Figure 4 :Figure 5 :") 9![Figure 9 : Gulele sub-city clinics The above map on Figure 9 represents the spatial location of clinics in the sub-city. According to the map distribution of clinics in the woredas are not equitable. Most of the clinics distributed in woredas below average. The average numbers of clinics are 4. Woreda 01, 05, 09 and 10 are above average, but the rest woredas are below average. The probable reason for the distribution of clinics in most woredas are below average because all clinics are owned by privately, thus, owners opened clinics where better market is available and accessible to main road.](image-6.png "Figure 9 :") 10![Figure 10 : Gulele sub-city health centers This map shows the spatial allocation of health centers in the sub-city. Based on the spatial distribution of health centers illustrated on the map, in this sub-city there are three health centers namely shiro meda health center (woreda 03), woreda 6 Tena tabiya (woreda 06) and Selam Tena tabiya (woreda 09) all of them are governmental. The recent structure of Ministry of Health there is not private health centers all of them are owned by government. According to their spatial location, health centers are not equally distributed on the woredas of the sub-city.](image-7.png "Figure 10 :") ![Journals Inc. (US)](image-8.png "") ![](image-9.png "") ![](image-10.png "") ![](image-11.png "") ![](image-12.png "") ![](image-13.png "") ![](image-14.png "") ![](image-15.png "") ![](image-16.png "") ![](image-17.png "") ![](image-18.png "") 1Gulele sub-cityPopulationPharmacyPopulation sharing one pharmacyWoreda 0129235129235Woreda 0220932120932Woreda 0328124128124Woreda 0413264-0Woreda 0524495-0Woreda 0626123-0Woreda 0737066312355Woreda 0831954131954Woreda 093270474672Woreda 1023727211864 2Gulele SubcityPopulationClinicPopulation sharing one ClinicWoreda 012923555847Woreda 0220932120932Woreda 032812439375Woreda 0413264113264Woreda 052449583062Woreda 0626123213062Woreda 0737066218533Woreda 0831954215977Woreda 093270474672Woreda 102372763955 3Gulele Sub cityPopulationHealth CenterPopulation sharing one Health CenterWoreda 0129235-0Woreda 0220932-0Woreda 0328124128124Woreda 0413264-0Woreda 0524495-0Woreda 0626123126123Woreda 0737066-0Woreda 0831954-0Woreda 0932704132704Woreda 1023727-0 © 2016 Global Journals Inc. (US)Health Facilities Distribution Mapping in Addis Ababa, Ethiopia ## Acknowledgements This research work was sponsored by Addis Ababa University. The author thus likes to express their heartfelt gratitude to the University for the financial support. * The1994 Population and Housing Census of Ethiopia Central Statistical Agency II 2007 Report Result at country level * Statistical Abstract 2011 * Developing a theoretical framework for the delineation of administrative boundaries within a rural context SEagleson FEscobar IPWilliamson 2001. 2001 Melbourne, Australia * Enterprise GIS in health and social service agencies Esri White Paper 1999 * AIDS: predicting the next map interfaces PRGould JKabel WLGorr AGolub 1991 21 * Geo computation SOpenshaw Geocomputation SOpenshaw RIAbrahart Taylor Francis New York 2000 * ARajabifard IPWilliamson 2001a Tsukuba Japan the University of Melbourne Regional administrative boundaries pilot project * Methods and techniques in human geography GMRobinson 1998 John Wiley and Sons * The benefits of the application of Geographic information systems in public and environmental health HJScholten MjcDe Lepper WHO Stat Quart 3 44 1991 * An overview: The Future of GIS RFTomlinson 2002. 2012