sf} ‘1 3e <1“ co, «8’ k. A_._ «.3’ Wfl= 25¢ per day per item RETURNIMS LIBRARY MATERIALS: Phce in book return to remove charge from circulation recon AN APPROACH TO PLANNING A PRIMARY HEALTH CARE DELIVERY SYSTEM IN JEDDAH, SAUDI ARABIA by Abdulaziz Sagr Al-Ghamdi A DISSERTATION Submitted to Michigan State University in partial fulfillment of the requirements for the degree of DOCTOR OF PHILOSOPHY Department of Geography 1981 NW .fi.««lul..lli NN N: O ABSTRACT AN APPROACH TO PLANNING A PRIMARY HEALTH CARE DELIVERY SYSTEM IN JEDDAH, SAUDI ARABIA by Abdulaziz Sagr Al-Ghamdi As urbanization in less developed countries acceler- ates, the absence of planning and appropriate social ser- vices becomes increasingly problematic. One important so- cial service component is a health care delivery system. The urban areas in Saudi Arabia have grown rapidly in the last decade because of the high immigration but experience a lack of vital public services, including health care de- livery. The purpose of this study is to study the inade- quacy of primary health care services and to present a quan- titative model that may be used to plan for an equitable de- livery system in Saudi Arabian cities. The City of Jeddah was selected for detailed analysis. The objective of the study is to develop a program for locating governmental health care service dispensaries that will be within access, not exceeding ten minutes travel time, for every inhabitant of Jeddah. Three major steps Abdulaziz Sagr Al-Ghamdi were used in the study: (1) study of the users' percep- tions regarding existing facilities; (2) a determination of the total number of potential service centers needed for the city based on varying trip time thresholds and the dis- trict areas of the city assuming the existing traffic flow system; and (3) assignment of potential service centers in the city (for selected areas). To achieve these goals, a field investigation was conducted. A sample of 400 users were interviewed, most of whom preferred using the public dispensaries to other facilities. The majority wanted prim- ary services, located within ten minutes travel time and more than half were found to use public dispensaries for emergency cases. Within the city 42 selected potential service cen- ters (with regard to trip time) were identified. The final goal was to assign the potential service centers for various trip thresholds. The enumeration technique of integer linear programming was employed to obtain the minimum number of service centers for various time constraints. Results in- dicated that within five minutes, 39 centers can be con- sidered. Within 10 minutes 27 centers are possible and within 15 minutes 23 centers can be assigned in the city. Recommendations for implementation and further study are stated. To the memory of my father-- I dedicate this work ACKNOWLEDGEMENTS I wish to express my gratitude and appreciation to Professor Ian M. Matley, committee chairman, for his guid- ance and comments. I am also indebted to Professor Assefa Mehretu who devoted several hours to directing me with his helpful suggestions. My thanks to Professor David Campbell and Professor Sanford Farness for their willingness to serve on the committee and their helpful comments. Sincere appre- ciation and gratitude are extended to Professor Stanley Brunn, Chairman of the Geography Department atijmaUniversity of Kentucky for his suggestions, comments and encouragement. To those in governmental offices who provided helpful in- formation my deep thanks. I also want to thank the Depart- ment of Geography at King Abdulaziz University, Jeddah, and express my appreciation for the financial and moral support provided by the University of Umm Al-Qura, Mekkah. Finally, my deep love and gratitude to my family, relatives and friends for their efforts, patience and sup- port throughout my years of study and work that have culmin- ated in this dissertation. ii TABLE OF DEDICATION . . . . . ACKNOWLEDGEMENTS . . LIST OF TABLES . . LIST OF FIGURES . . Chapter I. INTRODUCTION Statement The Study Area CONTENTS of the Problem . The Objective of the Study Organization of the Study . II. LITERATURE REVIEW OF LOCATION ANALYSIS OF HEALTH CARE DELIVERY IN CITIES Contributions by Geographers to the Study of Public Services . Socioeconomic and Demographic Variations The Hierarchy of Health Services Attempts to Optimize the Location of Health Facilities Utilization of the Emergency Room . Relevance of Literature to the Exist- ing Situation in Jeddah . . III. METHODS OF THE STUDY The Survey Method iii Page ii vi vii |._a \lkwN 11 12 14 17 19 19 Chapter IV. v. The Travel Speed Movement Rates The Application of Linear Programming . . . . . . . . CURRENT PATTERNS OF DISPENSARIES' USE AND CONDITIONS OF URBAN ACCESSIBILITY . . . . . . . . . The Socioeconomic Status of the Users . . . . . . . . . . . . Type of Transportation Used to Seek Health Care and Travel Time Required 0 O O O I O O O O O O C Type of Medical Care Assistance Sought by Respondents . . . . . . Accessibility of Dispensaries and Their Demand . . . . . . . . . . Reasons for Choosing Dispensaries and How Users Rate Health Care Services . . . . . . . . . . . Preference for Dispensaries for Emergency Use . . . . . . . . . . The Existing Conditions of Dispensaries . . . . . . . . . . The Travel Speed on Current Roads ASSESSMENT OF POTENTIAL SERVICE CENTERS FOR HEALTH DISPENSARIES . . Selection of Potential Service Centers (PSC's) . . . . . . . . . PSC's within the Five Minute Threshold Constraint . . . . . . PSC's within the Ten Minute Threshold Constraint . . . . . . PSC's within the Fifteen Minute Threshold Constraint . . . . . . iv Page 22 23 28 28 31 35 38 43 49 53 54 61 61 65 68 70 Chapter Page Relocating the Existing Fac- ilities O O O O O O O O O 0 O O 0 O O 74 PSC's which Appear in all Three TTT'S O O O O O O O I O I O O O O O 76 VI. CONCLUSIONS AND RECOMMENDATIONS . . . . 80 Conclusions . . . . . . . . . . . . . 80 Recommendations . . . . . . . . . . . 83 Strategy for Implementation . . . . . 85 Need for Further Study on Health Care . . . . . . . . . . . . . . . . 86 BIBLIOGRAPHY . . . . . . . . . . . . . . . . . . 87 APPENDIX . . . . . . . . . . . . . . . . . . . 94 IO. 11. 12. LIST OF TABLES Occupations of Survey Respondents . . Type of Housing in which the Respon— dents Live . . . . . . . . . . . . . . The Respondents Travel Time Prefer- ence to Seek Health Care . . . . . . . Types of Medical Facilities Patronized by Respondents . . . . . . . . . . . . Respondents' Perceptions of Dispen- sary Accessibility . . . . . . . . . . Respondents' Reasons for Choosing a Dispensary . . . . . . . . . . . . Types of Medical Facilities Pre- ferred for Emergency Cases . . . . . . Purposes for which Respondents Use Dispensary Services . . . . . . . . . Assigned PSC's within Five Minutes TTT O O O O O O C O O O O O O O O O O Assigned PSC's within Ten Minutes TTT O O O O O O O O O O O I O O O O O Assigned PSC's within Fifteen Minutes TTT O O O I O O O O O O C O O O I O O PSC's which Appear in all three TTT's vi Page 29 30 34 36 43 45 50 51 66 69 72 77 Figure 1. 2. 10. 11. 12. LIST OF FIGURES Map of the Study Area-—Jeddah City . . The Location of Existing Public Health Dispensaries . . . . . . . . . District Boundaries within Jeddah . . The Respondents' Monthly Income Levels 0 O O O O O O O O O O O O O O 0 Types of Transportation Respondents Use to Travel to the Dispensaries . . The Percentage of Respondents Using the Four Types of Medical Facilities . Distances Travelled to the Dispen- saries, by Percentage . . . . . . . . Distances Travelled by Respondents to Reach each Dispensary, by Percentage . Maximum Distances Respondents are Willing to Travel to a Dispensary . . Respondents' Attitudes toward Health Services at Dispensaries . . . . . . . Respondent Ratings of Services Avail- able at Existing Dispensaries . . . . Purposes for which Respondents Use each Dispensary . . . . . . . . . . . vii Page 21 24 32 33 37 40 41 44 47 48 52 Figure 13. 14. 15. l6. 17. 18. 19. 20. 21. 22. Speed Travel Time Measurements at 8:00 aOm. O O O O O O O O I O O O 0 Speed Travel Time Measurements at 4:00 p.m. . . . . . . . . . . . . . Speed Travel Time Measurements at Midnight . . . . . . . . . . . . . The 42 Selected PSC's for Jeddah . . . Population Gravitational Points in Jeddah O O O O O O O O O O O O O 0 Selected PSC's within Five Minutes TTT . . . . . . . . . . . . . . . . . Selected PSC's within Ten Minutes TTT O O O O O O O O O O I O O O O O 0 Selected PSC's within Fifteen Minutes TTT . . . . . . . . . . . . . Relocating Existing Health Facilities . The PSC's that Appear in all Three TTT's . . . . . . . . . . . . . . . . viii Page 56 57 59 63 64 67 71 73 75 78 CHAPTER I INTRODUCTION Urban and social geographers are aware of the polit- ical, economic and social problems facing urban areas, of which health care delivery is one component (Cox, 1973;Brunn, 1974; Berry, 1968; Johnson, 1976; Haggett, 1977; Herbert, et al. [eds], 1979). During the last two decades most studies of health care have emphasized the importance of distance, time, socioeconomic conditions and accessibility of medical care services in cities. Morrill, et a1. (1970) focused on the factors influencing distance travelled to hospitals. With the modern life saving technologies available, response time has become increasingly significant in the survival of patients who could not have been saved even a few years ago (Mayer, 1980). The characteristics of location have thus become worthy of study as planners seek the optimum location for a city's medical services (Rushton, 1979). Health care delivery planning is one of the areas in which geographers have made significant contributions to solving people's prob- lems especially in the deve10ped countries. The spatial behavior and welfare of the patients ultimately are the most important considerations in health care delivery studies (Earickson, 1970, 1972;de Vise, 1973). The main purpose of geographical and non-geographical studies regarding health care delivery has been to increase availability, accessibility, quality and distribution of health centers. An attempt has been made to locate these centers to improve the accessibility of each and to insure that patients can easily reach a facility within a short time. Unlike the advanced health care delivery available in the urban and industrialized nations, most of the less developed countries face a lack of availability, inaccessi- bility and low quality of health care facilities and delivery planning that can offer basic services in urban. and rural areas simultaneously (Good, et al., 1979; Kleozkowski, 1976). Statement of the Problem As urbanization in less developed countries continues to increase at unprecedented rates, the issues of planning and implementing appropriate social services have become in- creasingly complex. One of the major urban problems in these countries is a lack of efficient and equitable systems of health care delivery. Since everyone has a basic need for, and the right to, medical care; services need to be available at reasonable cost within reasonable proximity for all in- habitants. The urban areas in Saudi Arabia have grown rap- idly in the last decade but they experience a lack of vital public services, including health care. Health care de- livery systems in urban areas currently do not insure a reasonable level of service. The purpose of this research is to measure the seriousness of the inadequacy and to pre— sent a quantitative model that may be used to plan for an equitable delivery system in Saudi Arabian cities. The City of Jeddah is selected for detailed analysis because it is one of the major cities in the country and because it pre- sents useful research possibilities for investigating the spatial dimensions of health care delivery in some detail. It is hoped that the results of the research will have use- ful applications in other cities of Saudi Arabia, or else- where in urban areas of the developing world. The Objective of the Study The main objective of this study is to develop a pro- gram for locating governmental dispensaries of health care services that will be within access, not exceeding ten min- utes travel time, for every inhabitant in Jeddah City. To achieve this objective, three steps will be followed: (1) study of the users' perceptions regarding existing dispen- saries and their conditions; (2) a determination of the total number of potential service centers needed for the city based on the trip time threshold and the areas of the city with the existing traffic flow systems; and (3) assignment of poten- tial service centers for the selected facilities. In order to achieve the previous objectives, a field investigation has been conducted, the design which is ex- plained in Chapter III and the results summarized in Chap- ters IV and V. The Study Area Jeddah is situated in western Saudi Arabia on a coastal plain east of the Red Sea at 21° 30' north and 39° 12' east (Figure 1). Jeddah's location enables it to dominate the western region, since it has the major international airport, seaport, major wholesaling and retailing facilities for the region. Jeddah is also the main transit point for most trav- elers making their pilgrimages to the Holy City of Mekkah. It, thus, is called upon to provide comprehensive services, not just for the western region but for the whole country. The city is increasingly expanding outward to the north and east, its direction of growth influenced by main transport routes north to the new airport and Medina City, and east toward the City of Mekkah. Implementation of the government's five year plan has contributed loans to busi- nesses and the public for project and residence construction. new airport “~. ~_______..__-_____ ~‘ ~ \ \ \ Souroo' Kingdom 0' Saudi Ambi- Ministry of Municipal and Rural All-us Existing Conditions 0! m Mumpoliun Aron-Jeddah. 1960 Map of the Study Area - Jeddah City Figure 1. Major commercial and bank activities are concentrated in the city's central business district while retail services are distributed throughout the city's districts. Expansion of the city usually involves a mixture of residential and com- mercial development (Al—Ghamdi, 1979). In the residential areas the socioeconomic groups are mixed. In every district there are both high income people and low income people liv- ing side-by-side. No segregated districts for specific classes exist, although there are areas for those groups that like to live close to each other. Within the last decade, Jeddah has experienced a growth that has more than doubled the population. In 1971 the pOpulation was estimated at 381,000 (Ministry of In- terior, 1972), while it may now exceed 800,000. Of the present population, about 47 percent are Saudis and 53 percent are non-Saudis (Ministry of Municipality and Rural Affairs, 1980). Jeddah's growth is related to the variety of its job Opportunities, that draw immigrants from the rural areas to the city, and a large influx of laborers from other Arab countries, such as Africa, India, Pakistan and southeast Asia. These laborers work in both the private and public sectors in industries such as constructions and manufactur- ing. European and North American immigrants are generally employed in the technical areas. Many of the problems the city currently faces are the result of rapid growth and the lack of a carefully planned system of social services. There is currently no coordination or policy for formal c00peration between public service agencies to provide comprehensive services and avoid future service gaps and duplication. Among Jeddah's major problems is the lack of an adequate health care delivery system. Organization of the Study This study is divided into six chapters. In Chapter II the literature on the location analysis of health care delivery in cities will be reviewed. The survey design and methodology are discussed in Chapter III while the cur— rent patterns in use of dispensaries and conditions of urban accessibility are analyzed in Chapter IV. In Chapter V the selection and assessment of potential service centers for health dispensaries are presented. The final chapter will present conclusions and recommendations. CHAPTER II LITERATURE REVIEW OF LOCATION ANALYSIS OF HEALTH CARE DELIVERY IN CITIES A number of research contributions have been made by geographers and non-geographers in the field of health care delivery in the last two decades. These include a study by Godlund (1961) of planning locations for regional hospitals in Sweden; Gould and Leinbach's (1966) presentation of an approach for the geographic assignment of hospial services in Guatemala; and the evaluation by Shannon, et a1. (1975) of the geographic accessibility of health services in the Cleveland metropolitan area. These studies have provided a useful conceptual framework for this study as well as suggesting techniques which may be used for planning systems of health care delivery in urban areas of less developed countries. Contributions by Geographers to the Study of Public Services Most of the attention in public service location re- search has been devoted to developing locations for shopping centers, while social services, such as for health care, have been neglected until recently (Thomas, 1976). In determining the optimum location of such public services as clinics and hospitals the pertinent literature indicates that many fac- tors need to be considered. These include: (1) centrality (to population served); (2) accessibility (to ensure that service locations approximate the density and distribution of population for any metropolitan area); and (3) optimiza- tion that satisfies the demand for services and thereby avoids under-utilization of facilities (Jackle, et a1. (1976). Brunn (1977) has also stated that social geographers can make con- tributions to public policy by investigating the accessibility of the public to physicians, hospitals and other social ser- vices. Providing an efficient and equitable delivery system is closely tied to the location of such public facilities. Health care delivery is considered an essential service for the public and geographers have increasingly been contribut- ing to the study of such policies and problems (Bashshur, 1970; Dear, 1978; Pyle, 1979). Socioeconomic and Demographic Variations Variations in socioeconomic and demographic charac- teristics, such as income, occupation and religion, have an important influence on hospital care and the utilization and types of services people receive. For instance, high income peOple are able to purchase a greater variety of 10 health services and thereby have improved health, partly as a result of better housing and diet. Their level of edu- cation and ability to purchase certain goods and services, which increases with income, permits them to create the foundation for a better level of health (Pauly, 1974; Marden, 1966). As Earickson (1972) stated, "poverty is the crucial variable for the poor who have the greatest need for health services" (p. 1). In contrast, families with greater re- sources to spend for medical care can patronize private physicians, rather than wait for available public services. Anderson and Anderson (1967) also indicate that high income families are the greatest users of private physicians. These economic factors have become important in determining physicians' office locations. Physicians take into consid- eration those factors which affect the inflow of money for their services, including population size, proximity of ex- isting medical facilities, and the propensity for health services consumption (Lankford, 1972; 1974). Planners con- cerned about reducing inequities should understand and be able to measure the impact of these various economic and demographic factors in providing for more effective utili- zation of health care services (Reinke, et a1. 1967). 11 The Hierarchy of Health Services Health services are characterized by a hierarchy of functions. A good example can be demonstrated with hos- pitals. Hospitals vary in size by number of beds and other services provided. Only a few large hospitals within any major metrOpolitan area can usually provide a highly spec- ialized treatment. Small hospitals are unable to provide such services, so consumers desiring them may have to travel long distances. Schneider (1967) classified Cincinnati hos- pitals into three hierarchial levels: (1) a few very large hospitals; (2) several of large and medium sizes; and (3) a number of small hospitals. Morrill and Earickson (1968 classified Chicago hospitals into the following categories: (1) teaching and research hospitals; (2) large regional or district hospitals; (3) regional or district hospitals; (4) community hospitals; and (5) very small hospitals. Hospitals with highly specialized services are usually located close to city centers, transportation modes, and centers of popu- lation (Morrill and Earickson, 1968; Shannon, et a1. 1975; Bashshur et a1. 1970). All hospitals, whether in Cincin- nati, Chicago or Cleveland may be classified according to: (1) their size and number of the beds; (2) range of facil- ities: (3) size of the medical staff; (4) population distribution in the cities; and (5) types of programs offered. Morrill and Earickson (1968) used a principal component an- alysis for 99 variables to describe and classify each hospital within the Chicago area, As a result, nine major dimensions were identified as being important: (1) service volume; (2) the character of the service area; (3) quality of care and length of stay; (4) the emphasis upon obstetrics and ped- iatrics; (5) recent service capacity surveys; (6) competition; (7) services for non—white patients; (8) the range of per— sonnel, expense per bed and the proportion of public aid pa- tients; and (9) the emphasis on elderly patients. Morrill (1970) also noted that the number of patients using the hos- pitals declines with intervening opportunities and distance as all hospitals cannot provide needed specialized care. Attempts to Optimize the Location of Health Facilities The location of health care facilities is a major aspect of any health care services operation. A hospital should, ideally, be centralized to be Optimally accessible for the greatest number of the population, so that people can reach it in the minimum time and for the least cost. The locations of health services are important to people of different classes, levels of income, and ages. Hunter 13 (1974) stated that locational planning must take into account the scope of services to be provided, the dimensions of the catchment area, the geographical distribution of the popula- tion, and the rate of increase in the number of residents within the planned area. Various humanistic, behavioral and transportation models have been adopted to study the location and distribu- tion of health facilities. Schneider (1967) and Abernathy et a1. (1972) indicated that locational efficiency is a function of the cost of operating a hospital in a given lo- cation and the users' costs in terms of time and money to get to the hospital. Godlund (1961) determined several al- ternative sites for regional health centers and hospitals in Sweden through'Uuause of various geographic techniques. Demographics, economics, transportation and travel time to various centers were taken into consideration. Gould and Leinbach (1966) in western Guatemala used a linear pro- gramming model to assign hospital locations which yielded the lowest travel costs. A study of the location of primary health services in Guatemala City used a location-alloca- tion model to assign patients to a set of existing dis- pensaries. Minor relocations and capacity changes were permitted for specific dispensaries (Mulvihill, 1979). Morrill, et a1. (1969) explained the Optimum allocation of 14 health services, using central place theory in a heuristic approach. A simulation model was developed in which the probabilities of patients from different areas visiting var- ious hospitals are estimated from the modified construct for an initial estimate of hospital use (Morrill and Earickson, 1960; Earickson, 1969). Bennett (1979) indicated in his location-allocation study of primary health care centers in the Lansing area, where five new centers need to be located. He devised two location-allocation algorithms to determine facility locations and associated user allocations based on the distribution of undoctored households reported in an area mail survey. These studies and approaches to planning and allocating medical care facilities have been adapted for more efficient treatment in health care delivery planning based on the specific needs of people (Pyle, 1974; Dear, 1978). Utilization of the Emergency Room Another area in which contributions have been made in health care delivery is concerned with emergency health care where accessibility is of prime importance. Emergency departments of hospitals in the United States were, at one time, used almost exclusively for the treatment of accidents and injuries. Since World War II the emergency rooms have grown rapidly in importance with increasing utilization. 15 Emergency departments have become out-patient clinics and places where patients can receive regular treatment and where most private physicians can practice after office hours and on weekends. Because of the large increase in the number of uses of emergency room services during the 1960's, studies to ex- amine the extent of the problem this creates have been done by analyzing hospital emergency room data in different cities (Weinerman, 1965, 1966; Vaughn,l965; Coleman, 1967). Perkoff (1970) explained the results of two studies of the Barnes hospital emergency room in St. Louis, Washington. Charts from 6,688 visits were reviewed to find interlationships be- tween social and demographic factors and the use of the emerg- ency room. The study revealed that primarily two groups used the room: poor white and poor black ward patients who lived in the area close to the hospital. It was clear that non- urgent illness predominated in the two groups. Emergency floor records at Boston City Hospital were investigated to classify patient visits in terms of accident/non-accident and emergency/non-emergency use. About two-thirds of the patient visits were non-accident, non-emergency cases (Kirk- patrick, 1967). The increasing tendency of patients to use emergency rooms for non-emergency reasons can be attributed 16 to the difficulties poor pe0p1e have in obtaining primary health care. Some specific factors influencing non-urgent use of the emergency room are: lack of a relationship with a personal physician, age, residential mobility, minority group status, andtfingeographical location of health centers in the city (Weinerman, 1966). In regard to valid emergency cases, the ABC-TV pro- gram "20/20" on Thursday, April 24, 1980 indicated in its report on emergency services that between 18,000 and 20,000 accident victims out of the 115,000 who die annually in the United States might be saved if immediate care were rendered. Mayer (1980) found, in a geographical study of Seattle emerg— ency medical services, that the response time was dependent on three major objectives: (1) to minimize response time; (2) to minimize cost; and (3) to insure that most of the emergency calls will be answered within specified periods of time. Roghmann and Zastowny (1979) and Mayer (1979a, 1979b) concluded that quick reponse is the main objective of most emergency medical services. In separate studies they found that an average difference of three minutes in responding to an emergency can save many lives. The Los Angeles county- wide coordinating council studied emergency medical services in 1975 and classified the victims' conditions as mildly, moderately, or severely critical. The bulk of lives lost 17 were within the first ten minutes of when the emergency services were required (Navin and Stevens, 1979). Delivery and location of emergency medical services and ambulance services are still under investigation by plan- ners, economists, operations researchers and geographers. Planning is still considered the key issue in solving the problem of less than optimum location of such services (Achabal, 1978; Toregas, et al., 1972; Hamilton, 1974). Relevance of Literature to the Existing Situation in Jeddah In the City of Jeddah two types of health services are available--public and private. The cost of private health services is very high, as only the rich can afford them, whereas public health services are provided by the govern- ment at no cost to the users. Jackle, et a1. (1976) state that many factors need to be considered in prOposing plans dealing with such aspects as centrality, accessibility and optimization to satisfy the demand for services. During the current rapid urban develop- ment in Jeddah, these factors and others have not been given priority in planning. Location of health service outlets within Jeddah is 18 not based on population density, accessibility or centrality. Nor is the allocation of services based on classified hierar- chial levels, such as teaching and research hospitals, large regional hospitals, small community hospitals, dispensaries and clinics. Instead, seven government hospitals and seven dispensaries, under the supervision of the Ministry of Health, have been built at random to serve the entire popu- lation of the city. As a result, the emergency services are extremely poor. There is only one major emergency room func- tioning in the Central Hospital of Jeddah. There is a great demand for emergency services as there are more accidents and injuries occurring and recorded every day. In case of an emergency in Jeddah, while the rich can afford quick ser- vices at private health facilities, it is often difficult for the poor to get to the emergency room at a public hos- pital within less than an average of thirty minutes. This may be too long for the most serious cases. As Mayer (1979b) points out, quick response time is very critical and one of the main targets in saving lives in emergency situations. The data collection procedures and techniques used to analyze the field data collected are outlined in Chapter III. CHAPTER III METHODS OF THE STUDY In order to achieve the goals of this study, three approaches were used in the field: (1) a survey method was developed to collect and analyze data from a selected sample of users of public dispensaries within the city; (2) the travel speed movement rates at various times of the day were collected to help understand the existing transportation con- straint in the city; and (3) application of the method of linear programming was used to analyze the data at various levels of trip time thresholds (TTT's) to determine the need for future additional health facilities in the City of Jeddah. The Survey Method Seven major assumptions were made regarding the exist- ing situation of the prime health facilities in Jeddah. 1. The majority of user respondents use their own cars for seeking health assistance. 2. The majority of user respondents prefer a trip time of less than ten minutes. 3. The majority of health facility users pre- fer to use public dispensaries. l9 20 4. The proximity of dispensaries is a major factor in their use. 5. The user respondents do not see significant differences in services between the public dispensaries. 6. Most of the respondents' emergency cases are taken to the public dispensaries. 7. Most of the emergency cases for which people seek help are minor complaints for which dispensaries offer effective assistance. Answers to these and other questions provided a general profile of the residents' perceptions about the health care system in Jeddah. Also, they provided some idea of what is needed to improve health care delivery and where specific future facilities might be located (Appendix A). There are seven free public dispensaries in the city. These are under the supervision of the City's Directorate General of Health Affairs (Figure 2). A sample of 400 users from these dispensaries was selected to be interviewed re- garding their perceptions about the services available at these dispensaries. The sample for each dispensary was chosen in the following way. For each dispensary a percentage was computed between the number of users of that dispensary and the total number of users of all seven dispensaries during the month prior to the field survey. The number of users rict m D k m m m mm.“ M um m dhum I MGNH B 1.2.3.4. 7. fl 0 .m. n a w m O H n a m 0 21 The Location of Existing Public Figure 2. ispensaries Health D 22 to be interviewed from each dispensary was based on the per- centage for that dispensary in the total sample (Lansing, 1971). Then, systematic random sampling based on the random table was used to select the actual interviewees (Blalock, 1979). Frequency analysis was used to analyze the data and the Michigan State University Computer was employed. Four senior students experienced in conducting field surveys, from the Department of Geography at King Abdulaziz University in Saudi Arabia, helped in conducting the inter- views. The interviews were scheduled in the evenings to en- sure that the sampled population included those who worked in different sectors during the day. The survey was carried out from November 1980 through January 1981. The Travel Speed Movement Rates Measurements of road travel speed movement for various time periods were conducted in Jeddah. A field measurement was carried out around locations of the seven existing public dispensaries. The threefold purpose behind conducting these measurements was: (1) to get an idea of the travel speed in the city; (2) to determine the length of trip time it would take an emergency vehicle to travel to the closest dispensary from an accident site or a patient's residence; and (3) to 23 determine the threshold time experienced by the inhabitants in getting to the existing health facilities at various times. Usually the public dispensaries are open from 7:30 a.m. to 12:30 p.m., and from 4:30 p.m. to 7:30 p.m. every day, except on Thursday afternoons and all day Friday. Accord- ingly, the travel speed measurements were taken at three times of the day: 8:00 a.m., 4:00 p.m. and at midnight. At each of these hours, measures of distance travelled in every dir- ection from each of the seven public dispensaries were re- corded every five minutes up to a maximum of twenty minutes travel time. All of the recorded measures were then trans- ferred to the city map to create a contour map for the var- . ( ious periods of measurements. The Application of Linear Programming In order to assign the appropriate number of potential service centers (PSC's) for future health dispensaries in Jeddah, a linear programming technique was employed. Data came from the field survey, especially the travel speed move- ment rates of urban traffic, and the following pieces of in- formation were used to Operationalize the model: 1. Jeddah city map and its subdivisons, referred to as districts in this study (Figure 3). There are no official boundaries between the districts. The district boundaries used for the purposes 24 21 . . Karantina Ghulail . Gurayat and Thalbaa Nuzla Yemenia Bukharia . Hindawia Shati . Sabeel . Harat Sarah 10. Sahaita 11. Nuzla Sharkia 12. Mekkah Rd. Kilo. 6-10 13. Mekkah Rd. Kilo. 1-5 ' 14. Harat Yemen Ascham V 15. Baghdadia and Amaria 16. Kandara 17. Sharafiyah 18 Bani Malik . ' - 3' ' ~ 19: Ruwais. Hamra. Medina Rd. - .f~ 5 1' 20.nnusharnan . 13 21. Northern parts Red Sea wmsmwewwd fits... r0 d LN 3 Km. Source: Kingdom of Saudi Arabia Ministry 01 Municipal and Rural Aflarrs Existing Conditions 01 the Metropolitan Area-Jeddah. 1%0 Figure 3. District Boundaries within Jeddah 25 of this study are statistical areas which were used in the 1978 socioeconomic survey based on a previous 1971 social survey (Ministry of Municipal and Rural Affairs, 1980). 2. size of each district in kmz; 3. the population density in each district based on a 1978 survey. 4. average effective traffic speed limit in each district. 5. the location of 133 selected population gravitational points (PGP's) in the city in- cluding 42 potential service centers (PSC's) of the public health dispensaries, based on the field observation and criteria outlined in Chapter V; and 6. binary 0/1 matrix between the 42 PSC's and the 133 PGP's on the basis of accessibility, using the three trip time thresholds (TTT's). In this study the enumeration techniques of integer programming were used in order to minimize the PSC's for a given TTT constraint (Cohen and Stein, 1978). The output of the program is MPOS input file defining a 0/1 integer programming problem to minimize the number of facilities selected, given the constraints. The objective function will be satisfied if the sum of the assigned sites gives the minimum number of locations that would be accessible to all of the 133 PGP's within each of the specified TTT's. The constraints of the problem were set up to force each PGP to at least one or more PSC for a given TTT constraint. In summary: l. 26 Objective function is to minimize the number of PSC's from each of the TTT's. 42 . . . . (= 1 if selected M = . . . . inimize 5:? x a 3 x3 (= 0 1f else a . j 1 j = l, ..., 42 Where x = number of services for TTT = K = 5, 10, 15 minutes aij j = l, .. , 42 i = 1, ...,133 aij = 1 if TTTéK = 0 if TTT7K 2. Subject to constraints, a. Each PGP should be assigned to at least one PSC. b. PSC services itself. c. Maximum time TTT which equals 5, 10 and 15 minutes. Additional information was collected through several meetings with officials in the Ministry of Health in Riyadh, 27 officials of the Directorate General of Health Affairs in Jeddah, physicians who are working in the existing public dispensaries, the officials of Jeddah municipality. These meetings were useful since they provided overview about the existing situation and the future plans for health care de- livery in the city. In the following chapter, current patterns of dispen— sary use and conditions of urban accessibility will be dis- cussed. CHAPTER IV CURRENT PATTERNS OF DISPENSARIES' USE AND CONDITIONS OF URBAN ACCESSIBILITY It is useful to describe existing travel patterns to the seven public dispensaries in the city. A general dis- cussion of the users' perceptions regarding existing travel patterns to the dispensaries and their conditions is divided into eight sections. They are: (l) the socioeconomic status of-uuausers; (2) type of transportation used for seeking health care and travel time required; (3) type of medical care assistance sought by users; (4) accessibility of dis- pensaries and their demand; (5) reasons for choosing dispen- saries and how users rate health care services; (6) prefer- ence for the dispensaries for emergency use; (7) the exist- ing conditions of dispensaries; and (8) the travel speed on current roads. The Socioeconomic Status of the Users In this section the demographic data relating to age, size of family, occupation, housing, and monthly income are discussed. About 29 percent of the respondents were over 34 years old, 15 percent were 30 to 34 years, 28 percent 28 29 were 25 to 29 years, 20 percent were 20 to 24 years, and about 8 percent were less than 19 years of age. Almost half of the respondents were between 20 to 29 years. The survey results indicated an average family size of six persons. The highest percentage of large households were among the residents of Hindawia, Ruwais and Bani Malik districts. A classification of the respondents' occupations is shown in Table 1. Table l. Occupations of Survey Respondents Type of Occupation Respondents % of Respondents l. Goverment Official 146 36.5 2. Businessman 36 9.0 3. Laborer 165 41.3 4. Student 22 5.5 5. Unemployed 31 7.7 6. Other -- -- Total 400 100.0 Source: Field Survey by author, 1980 A high percentage of the respondents were laborers and a majority of those laborers were non-Saudi. The com— 30 panies which employ them do not provide health insurance benefits for them. Another 37 percent of the respondents were government officials. Less than ten percent were busi- nessmen . In regard to type of housing, survey results revealed that about 36 percent of the respondents owned their homes while 64 percent did not. The majority of the respondents were found to live in traditional houses, which are less expen- sive than apartments and villas (Table 2). Table 2. Type of Housing in which the Respon- dents live. Type of Occupation Respondents % of Respondents 1. Villa 15 3.7 2. Apartment 121 30.2 3. Traditional House 247 61.9 4. Shack 16 4.0 5. Other 01 0.2 Total 400 100.0% Source: Field Survey by Author, 1980. 31 The average monthly income of the respondents is shown in Figure 4. About 34 percent receive from $600 to $799 monthly. Only about 17 percent receive an income higher than $1,000 per month. These figures show that most of the respondents do not have a high income relative to the cost of private hospitalization. For example, the cost of three days of in-patient medical care at a private hos- pital in Jeddah may exceed $1,000. Type of Transportation Used to Seek Health Care and Travel Time Required Almost half of the respondents indicated that they have their own cars. The survey revealed that about 48 per- cent use their own car for seeking health care,while about 17 percent use a taxi, and another 17 percent ride buses. Only about 14 percent of the total respondents walk to the dispen- sary. It seems that there are variations in types of trans— portation used by respondents from one dispensary to another (Figure 5). However, access to automobiles for transport is very easy,especia11y in cases of illness. Respondents were also asked how long it would take them to drive to the nearest dispensary. About 28 percent indicated that it would take them less than ten minutes,whi1e about 18 percent within 15 to 19 minutes, about 12 percent within 20 to 24 minutes, and only 5 percent would be delayed 32 [:1 less than $399 90_ $400 to $599 - $600 to $799 so— - $800 to $999 - more than $1000 70‘ 60- Percent 0'! O l 40- Monthly Income Note: One US. dollar = 3.5 Saudi Riyal Figure 4. The Respondents' Monthly Income Levels 33 2383: v .215 D «:0 § .5 I .3 29:5 m ass§>-_ 31:22 23 a m_NDZ m moAHMmcmmmHo on» cu Ho>mufi ou mm: mnemocommmm coauousommcmufi mo momwe Smocmx m is“ .5350 N rov [on .m madman coautoomcmfi .0 09¢ ma flaiamm 6590 3.6%... F 13 r8 rah iueoied waxed 34 more than 25 minutes in reaching the nearest dispensary. The respondents were also asked their opinions regarding travel time to reach the nearest dispensary. About 88 percent wished to have the dispensary located within less than 10 minutes travel time from their homes. Only 12 percent indicated they would not mind travelling more than 10 minutes from their homes to visit a health facility (Table 3). Table 3. The Respondents Travel Time Prefer- ence to Seek Health Care Time Respondents of Respondents 1. Less than 10 minutes 350 87.5 2. 10 to 14 minutes 49 12.3 3. 15 to 19 minutes 1 .2 Total 400 100.00 Source: Field Survey by Author, 1980 The fact that most of the respondents would like to be able to reach a dispensary in less than 10 minutes is an important factor to be considered in planning for health care 35 delivery systems in the future. Type of Medical Care Assistance Sought by Respondents Patients usually seek more than one source of medical care assistance. Previous studies have shown that the patienth income plays an important role in the source of health care assistance sought. PeOple with high incomes usually frequent the private hospitals for many obvious reasons. These include faster services, prolonged individual attention by medical per- sonnel, and the feeling that private hospitals offer a better quality of service than the public hospitals. About 72 per- cent of the respondents indicate use of the dispensaries, 15 percent frequent public hospitals, 12 percent patronize private hospitals, and only one percent visit family doctors (Table 4). The percentages of respondents utilizing each of the four types of medical care facilities in each of the seven districts is shown in Figure 6. The results reveal that.family doctors are not sought by most of the respondents, a finding which may be due to socioeconomic reasons. Only 4 of 400 re- spondents, three in Nuzla and one respondent in Bani Malik call at a family doctor's clinic. Regarding the use of private hospitals, a rather low percentage (8 to 13 percent) of respondents who use public 36 owma .uozua< an >w>uam oaowm "condom o.ooH cm o.ooH mm o.ooH we o.ooH me o.ooH «a o.ooH mm o.ooH ow o.oo~ ooq A4909 whom «.oN mm m.Hw mm o.om me «.mn mm c.mo so m.wc no ox 0H «.mm arm Icmdmwn w.qa w q.m N H.@ m m.mH o o.oH ma m.m~ mm OH N N.mH Ho Hmufiamo: ofiabsm o.na N H.HH m m.~H w m.mH c m.NH NH ~.w m CH N m.HH cw kufldmom muw>fium w.H H ii ii an ii ii ii ~.m m ii ii an ii o.H q souuoo NHNEmm .dmmm .qmmm .amwm .Qmom .dmmm .amom .amom .ammm .Qmmm .Qmom .dmom .ammm .dwom .dmmm mucmocoam mucmocoam Suwawomm N .oz N .oz N .02 N .oz N .oz N .oz N .02 log mo N 10m mo .oz mo waxy xaamz «com mumvcmx mamasm mfismocwm mausz Hfimaaso uuapumfio Howuu ImsocH Hmumcou mmfiumwcoamwa ecu mo :oNumooA mucmpcoammm ha wouficonumm mofiufiawumm HmUNomZ mo moazfi .s warms 0% Industrial District Ghulail hwfla Hindawia Ruwais Kandara Bani Malik Figure 6. 37 l Family Doctor Private Hospital Public Hospital Dispensary The Percentage of Respondents Using the Four Types of Medical Facilities 38 dispensaries also use private hospitals. This limited use of private hospitals may be attributed to their high cost for services and the free public health facilities available elsewhere. Between 7 to 23 percent of dispensary users also seek treatment at the general public hospitals. This low percentage may be due to the location of the hospitals,types of selected specialized services these hospitals offer,traf- fic congestion and the difficulty in finding parking facil- ities there. It also takes longer and costs too much to reach most of the public hospitals. The main hospital is the general hospital in town which provides all types of ser- vices. The situation at the dispensaries is altogether dif- ferent and most of the respondents often use these facilities. About 80 percent of the users seek health care at the closest dispensary. The free service, the relative proximity of the location, and the primary nature of the medical treatment offered attract these users. It is understandable that, pro- viding easily accessible facilities will be helpful for this group of users. Accessibility of Dispensaries and Their Demand The respondents were asked to indicate how accessible 39 the dispensaries were to their residences. About 22 percent travel less than 2 kilometers, 40 percent from 2 to 3 kil- ometers, 19 percent from 3 to 5 kilometers, 13 percent from 5 to 10 kilometers, and 6 percent travel more than 10 kilometers (Figure 7). Nearly 60 percent travel less than 5 kilometers. The data were further analyzed to determine the travel be- havior patterns of respondents in each of the seven districts (Figure 8). Respondents from the industrial district, Ghulail, and Nuzla dispensaries shared similar travel patterns. While al- most 50 percent of the Hindawia dispensary users travel 2 to 3 kilometers, none travel more than 10 kilometers. Travel patterns of users of the Ruwais dispensary are somewhat equally distributed: 21 percent of respondents travel less than two kilometers, about 23 percent from 2 to 3 kilometers, and 21 percent travel from points 4 to 10 kilometers distant. The largest percentage (14.5 percent) of respondents travel- ing more than 10 kilometers was found among the users of the Ruwais dispensary; this pattern may be attributed to the lack of dispensaries in the north districts of the city. About 11 percent of the respondents travel less than 2 kilometers to the Kandara dispensary, 41 percent between 2 and 3 kil- ometers, 37 percent between 3 and 5 kilometers and only 11 Percent 90- 70- 60" 50‘ 30- 20~ Figure 7. Distance 40 {3 less than 2 kilometers 2 to 3 kilometers - 3 to 5 kilometers - 5 to 10 kilometers - more than 10 kilometers Distances Travelled to the Dispensaries, by Percentage 4 Hindawia 70~ 3 Nuzla 70- 2 Ghulail l T i 1 l T E 53 8 53 8 53 S «on .2 b a .2 D I! L- 4.- U) D U S v- iiiiiiii 2 ES 8 weaned Distance 41 2 kilometer C OM19! % 2103ki \- o a o E N a 3 00-0.! ~01 °’ 2 S 8:25 :3: xcfi “'5'.” b 2.90 00:05 ...7 WE] .4.‘ - E“ (///// .E 7,, (U [D h 7 I I j T fi 53 8 E3 8 $3 8 52 ° m l] 3- m U C (U N % co EM s. l T I l I fl i3 8 5% 8 53 8 53 ° .9 (U 3 3 E LO r T I T I E? 8 5% 8 $3 8 53 ° iuemed Distance Figure 8. Distances Travelled by Respondents to Reach each Dispensary, by Percentage 42 percent travel beyond 5 kilometers because the dispensary is within a very crowded residential area. Thirty-seven per- cent of the respondents travel less than 2 kilometers to Bani Malik dispensary; another 35 percent from 2 to 3 kilometers while less than 10 percent travel from points 4 to 10 kil— ometers distant. Another 13 percent travel more than 10 kil- ometers. Accessibility to Bani Malik dispensary is similar to the situation in Ruwais district. Most of the north districts in Jeddah are without health facilities. When questioned about the location and accessibility of the seven dispensaries, in a four stage response pattern, 17 percent of the respondents found the existing locations of the dispensaries very accessible. Another 50 percent found them accessible. More than one quarter of the respon— dents felt the locations of the dispensaries were inacces- sible; about 6 percent perceived them to be very inacces- sible (Table 5). In order to determine users' perceptions regarding the accessibility of dispensary locations, the respondents were asked about the distance they were willing to travel to obtain health care. Almost 77 percent of the respondents would prefer to seek health care within a traveling distance of less than 2 kilometers; the remaining 23 percent are will- 43 Table 5. Respondents' Perceptions of Dispensary Accessibility Perceived Level of Accessibility Respondents % of Respondents 1. Very Acces- sible 69 17.3 2. Accessible 198 49.6 3. Inaccessible 109 27.2 4. Very inacces- sible 22 5.5 5. Do not know 02 .4 Total 400 100.0 Source: Field Survey by the Author, 1980 ing to travel 2 to 3 kilometers (Figure 9). No one surveyed would prefer to travel further. These results show that users are unwilling to travel long distances, a factor which needs to be considered in planning future health care facilities. Reasons for ChoosinggDispensaries and How Users Rate Health Care Services There are both medical and non—medical reasons for choosing a particular dispensary. The non-medical reasons 44 [3 less than 2 kilometers rt. 2 to 3 kilometers 90- 70- 60- Percent $ 30- Distance Figure 9. Maximum Distances Respondents Are Willing to Travel to a Dispensary 45 may be related to accessibility, a particular physician who treats his patients well, or faster services in one dispen- sary than another. The survey results identify three major categories of responses that need to be considered (Table 6). Roughly 77 percent of the respondents indicated that they chose dispensaries because they are close to their resi- dences while 13 percent believed that the dispensary they frequent has good physicians. Seven percent of the respon- dents felt that services at their dispensaries are faster than Table 6. Respondents' Reasons for Choos- ing a Dispensary Categories of Reasons Respondents % of Respondents 1. Because it is close to my residence 307 76.8 2. Because it has a good physician 53 13.2 3. Because it is faster than others 29 7.2 4. Do not know 11 2.8 Total 400 100.0 46 at the other medical facilities in the city. More than 60 percent of the respondents believed the dispensaries to be crowded. Distance, that is close proximity, is the primary reason given by the majority of the respondents choosing a local dispensary. An attempt was also made to examine the users' atti- tudes toward the quality of the health care services offered by the dispensaries. Of the total sample about 29 percent of the respondents rated the services as excellent, 37 per- cent above average, 25 percent average, and less than 5 per- cent below average (Figure 10). These ratings, however, vary slightly for each of the seven dispensaries. About 30 percent of the respondents who use the Industrial District dispensary, or Ghulail, Nuzla, Hindawia and Ruwais dispensaries rated the services as excellent, while about 60 percent of the re- spondents at Kandara dispensary rated the services as excel- lent. Users of the Bani Malik dispensary,have a lower opin- ion of the services provided there. Only 15 percent of the respondents rated the services as excellent (Figure ll).This could be related to the type of physicians in the dispensary, the kquuality of services providing medical treatment.Some— times, language is a barrier to communication with the phys- icians there since most of them are Pakistanis. Percent 90‘ 80- 70- 60- 40- 30- 20- 47 [3 Excellent Above average - Average - Below average - Do not know Level of Rate Figure 10. Respondents' Attitudes toward Health Services at Dispensaries .9 3 (U '0 .E I V l r r E 53 8 2 N 3 Z ('0 r m 1 8 £3 8 E 3 .C (D N F 1 T 2 £3 8 E2 .5. .Q 0 .72 5 fl (0 3 'O E v- r r T r l n 1 2 23 8 58 8 53 9 ‘3 waxed Level of Rate 48 7BaMIMflm - Excellent 3:? é a a fig; <