PLACE IN RETURN BOX to remove this checkout from your record. TO AVOID FINES return on or before date due. #— DATE DUE DATE DUE DATE DUE MY 2 0 I995; at: 2am ‘ -17"? "» MSU lo An Affirmative Action/Equal Opportunity Institution “domains-9.! ABSTRACT Cholera made a sudden, unexpected appearance at Conakry, Guinea in August, l970 after a seventy-five year absence. Cholera has since spread throughout West Africa and to North Africa, East Africa, Angola and South-Eastern Africa. The patterns of inter- regional diffusion of cholera in Africa have been mapped. The study focused on West Africa, where four distinct forms of cholera diffusion were found. These are the coastal, riverine, urban hierarchical and radial contact diffusion types. Diffusion barriers and channels, plus distance and central place attraction were found to be important in determining the paths of inter-divisional diffusion of cholera. ,A summarizing model was made for each diffusion type. Local cholera survival and diffusion was examined through a case study of local diffusion in Lagos and brief summaries of published case studies. General systems modelling techniques were used to focus on key elements and relationships involved in inter-divisional cholera diffusion and local survival and transmission. DIFFUSION OF CHOLERA IN NEST AFRICA, 1970-1974 by Robert Frederick Stock A Thesis submitted to Michigan State University in partial fulfillment of the requirements for the degree of MASTER OF ARTS Department of Geography 1974 ACKNOWLEDGEMENTS I wish to express my gratitude to a number of pe0ple without whose help this thesis would not have been possible. I would eSpecially like to thank my advisor, Dr. John M. Hunter for his encouragement, interest and guidance during my stay at Michigan State, in particular during the preparation of the thesis. I also wish to thank Dr. Stanley Brunn for consenting to be on my examining committee, and Dr. B.A.A. Dada, Chief Statistician, Lagos State Ministry of Health for providing information on cholera in Lagos which formed the basis for much of Chapter 6. Thanks are extended to my brother, George Stock for providing invaluable help in a dozen different ways. Finally, I would like to gratefully acknowledge the invaluable encouragement and assistance of my parents. ABSTRACT TABLE OF CONTENTS ACKNOWLEDGEMENTS LIST OF TABLES LIST OF FIGURES CHAPTER I - INTRODUCTION CHAPTER 2 .. The Study of Infectious Diseases by Medical Geographers Disease Diffusion Study Objectives Data Sources and Data Reliability Interpretation of Diffusion Arrows CHOLERA ,.Causative Agent CHAPTER 3 - Clinical Symptoms Immunity Transmission Cycle Control Measures HISTORY OF CHOLERA Occurrence of Cholera before l8l7 Nineteenth Century Cholera Pandemics Cholera in Africa During the Nineteenty Century Worldwide Incidence of Cholera from l923-l960 Cholera El Tor and the Seventh Pandemic iv 14 l6 l6 18 20 22 27 29 29 29 33 39 41 TABLE OF CONTENTS (cont'd) CHAPTER 4 - CHOLERA EPIDEMIC ZONES IN AFRICA ; West Africa '/ CHAPTER 5 - CHAPTE R 6 - North Africa East Africa Angola South-Eastern Africa INTER-REGIONAL DIFFUSION OF CHOLERA IN WEST AFRICA Origin of Infection Coastal Diffusion Riverine Diffusion Urban Hierarchical Diffusion Radial Contact Diffusion CASE STUDIES ILLUSTRATING THE LOCAL DISSEMINATION OF CHOLERA Lagos, Nigeria Ibadan, Nigeria Zaria, Nigeria Goudoume, Ivory Coast Louga, Senegal Akodessewa-Plage, Togo Goulfey, Cameroun Generalizations Page 46 46 53 57 62 66 72 72 73 90 lll 126 137 138 158 160 161 162 162 164 165 TABLE OF CONTENTS (cont'd) Page CHAPTER 7 - SYSTEMS MODELS OF CHOLERA DIFFUSION 167 External Diffusion 171 Local/Internal Diffusion - 180 Decline of an Epidemic 194 CHAPTER 8 - SUMMARY AND CONCLUSIONS 199 Spatial and Temporal Patterns of Cholera Diffusion 199 Spatial Diffusion Principles 202 Model of Cholera Survival and Diffusion 204 Endemicity 207 APPENDIX 1 209 APPENDIX 2 213 APPENDIX 3 215 BIBLIOGRAPHY 237 vi Table LIST OF TABLES Extent of the first six cholera pandemics Mali: Correlation of the week of cholera onset with selected variables Strength of relationship between the occurrence of cholera and selected variables in Nigeria Nigeria: Correlation coefficients between the week of cholera onset and selected variables Cholera morbidity in Lagos State by district or ward during the first sixteen weeks of the 1970-71 outbreak Age-sex distribution of the first 922 cholera cases in Lagos State vii Page 31 105 120 120 146 157 Figure —J <3 no co \4 Cl 01 .p g» k, 18[\ 18E; LIST OF FIGURES Three types of spatial diffusion Cholera transmission cycle Cholera in Eastern Africa 1864-1871 Trend of the global spread of cholera 1960-74 Cholera-affected areas in Africa 1970-74 Diffusion of cholera in West Africa Diffusion of cholera in North Africa Diffusion of cholera in East Africa Diffusion of cholera in Angola Diffusion of cholera in South-Eastern Africa Diffusion of cholera along the West African Coast Fanti fishermen, Cape Coast, Ghana The coastal diffusion system Schematic map of coastal diffusion First ten weeks of the riverine cholera epidemic Cholera diffusion in the Middle Niger watershed Diffusion of cholera along the Senegal Valley Corridor Distance, urban hierarchy, valley location and cholera diffusion in Mali and Western Niger The riverine diffusion system Schematic map of riverine diffusion viii Page 24 337 43 48 50 55 59 64 68 75 76 88 88 93 97 100 107 110 110 Figure 19 2C) 21l\ 2113 22 23l\ 23E3 24 25 26 27 28 29 3o 31 32 33 34 35 136 37 LIST OF FIGURES (cont'd) Urban hierarchical diffusion of cholera in Nigeria and South-Central Niger Distance, urban hierarchy and cholera diffusion in Nigeria and South-Central Niger Urban hierarchical diffusion system Schematic map of urban hierarchical diffusion Radial contact diffusion of cholera in the vicinity of Lake Chad Radial contact diffusion system Schematic map of radial contact diffusion Village at mouth of Ogun River, mile 15 on Lagos- Shagamu highway Housing density in Central Lagos Sequence of cholera onset in Lagos Epidemic curves of cholera in the four regions of the Lagos Metropolitan Area, December 26, l970—March 19, 1972 Explosive and implosive cholera diffusion during the first seven weeks of cholera on Lagos Island. Local and external cholera diffusion. EXternal (inter-divisional) cholera diffusion model. Market scene in Mokwa (Nigeria) Local survival and transmission of cholera Urban slum housing in Cape Coast (Ghana) SiZe of the cholera-susceptible population A village well, Dogon Doutche (Niger) Roadside foodsellers, Po (Upper Volta) Fishermen's canoes in Five Cowrie Creek, Lagos (Nigeria) ix Page 114 122 125 125 128 135 135 139 143 145 150 155 169 173 176 182 184 187 193 195 196 CHAPTER 1 INTRODUCTION The Study of Infectious Diseases by Medical Geographers A growing number of geographers are becoming interested in geographic relationships in the occurrence of infectious diseases. Study- hwy the geography of infectious diseases is but one facet of medical geography, whicri has been defined as "the application of geographic concepts and techriiques to health-related problems".1 This definition readily facil itates the incorporation of the wide spectrum of health-related problems being; studied by geographers, including psychosocial stress, location of healtri care facilities, and the impact of environmental pollution; and it broadenis the definition given by Jacques May in American Geography Inventxary and Prospect. May defined medical geography as "the study of the arwea distribution of disease and its relationship to the environment". 2 Interest in geographic aspects of infectious disease is not a new Phenonkenon. Consideration of environmental factors in health was a basic featurwa of Greek medicine. The period between 1835 and 1855 has been Cfl11eci the "Golden Age" of medical cartography by Gilbert.3 Cholera was one of’ the most intensively studied diseases during this era. John Snow's M 1 J-Pl. Hunter, "0n the merits of holism in understanding societal health "GEdS", The Centennial Review, vol. 17, No.1 (1973), p.4. ‘3- lflay, "Medical Geography", in P.E. James and C.F. Jones (eds.), Anmirican Geogrgphy: Inventory and Prospect, AAG, 1959, p.453. E 1' Gilbert, "Pioneer maps of health and disease in England", 9E5193~aphical Journal, vol. 124 (1958), pp. 172- 183. treatise on cholera in London in 1849 was a landmark study. By careful map- ping of cholera deaths in a part of central London, he was able to demons- trate a connection between the spread of cholera and the use of water from the pumps of a particular supply company.4 Previously, cholera had been considered to be an air-borne disease. Interest in medical geography declined during the second half of the nineteenth century with the shift of attention to the study of pathogenic bacteria and the search for antibiotics to combat them. Interest in medical geography has gradually revived since about 1950. J.M. May's World Atlas of Disease (1950), Ecology of Human Disease (1958), and series on the ecology of malnutrition, L.D. Stamp's Geography of Life and Death (1964), R.M. Prothero's Migrants and Malaria (1955) and the volume of selected papers edited by N.D. McGlashan, MGeography Technigues and Field Studies (1972) are significant landmarks in the recent study of infectious disease by geographers. There are four basic approaches used by medical geographers in the study of infectious disease.' These are the map approach, map correlation approach, quantitative approach and ecological approach. (a) Map approach: Spatial variations in the incidence of a disease are mapped. This is the approach used in most medical atlases. Maps SPOWI rig patterns of disease diffusion may also be included as part of the map a pproach. \— 4 )1. Snow, Snow on Cholera (reprint of 2 papers by J. Snow, 14.0.)- eW York: Commonwealth Fund, 1936. (b) Map correlation approach: The correlation between a disease and certain environmental conditions is demonstrated by showing the coincidence of their spatial distributions. This was the method used by Snow in his noted study of cholera.5 Allen-Price's map relating cancer deaths to water supply sources is another example of this approach. Computer mapping is a recent innovation in the analysis as well as the portrayal of disease distributions. (c) Quantitative approach: Geographers are making increasing use of analytical models and such quantitative techniques as correlation and regression, factor analysis and graph theory to analyse statistical data and thereby understand the distribution or dissemination of a disease. Two examples of the quantitative-model approach are Haggett's modelling of measles outbreaks in South Western England7 and Brownlea's development of a QEOQraphic model of infectious hepatitis.8 (d) ECological approach: Studies of the disease transmission cycle and the environmental conditions which facilitate the persistence of mfeclt'ion in an environment or its spread to a new environment may be called the ecological approach. The study by Hunter in Northern Ghana of \ 5 Show, op.cit. G~D. Allen-Price, "Uneven distribution of cancer in West Devon with FEFerence to the divers water supplies", Lancet, June 4, 1960, DD . 1 235-1238. 7 . P- Haggett, "Contagious processes in a planar graph", in N.D. McGlashan (ed.), Medical Geggraphy: Techniques and Field Studies, London: 8 Methuen, 1973, pp.307-324. A-A. Brownlea, "Modelling the geographic epidemiology of infectious hepatitis", in N.D. McGlashan (ed.), op.cit., pp.279-3oo. spatial and environmental relationships and implications of the river blindness transmission cycle demonstrates the potential of this approach.9 The four approaches listed are not mutually exclusive categories, but rather areas of emphasis. Distribution maps, consideration of the transmission cycle and some form of statistical analysis are basic to almost any study of infectious disease. The choice of the primary strategy for studying a problem generally is based on the researcher's training and philosophy of geography, on data availability and reliability, and on the scale of analysis. Disease Diffusion Either endemic or epidemic occurrence of a disease may be studied. Endemicity refers to the persistence of a disease in a given location. 'The transmission cycle is generally the key to this persistence. An epidemic may refer to either a seasonal or periodic flare-up of a disease in its endemic area, or its spread to a new area. Particularly with newly introduced epidemics, disease diffusion is an important concept. Disease diffusion refers to the process of spread of a disease through space and time. Considerable research interest has focused in recent years on Spatial diffusion. Many of the models and concepts used in studying the 9 J.M. Hunter, “River blindness in Nangodi, Northern Ghana: A hypothesis of cyclical advance and retreat", The Geographical Review, vol.56 (1966), pp.398-4l6. FIGURE 1 THREE TYPES OF SPATIAL DIFFUSION Contact diffusion is a wave-like outward movement to adjoining territory commencing at a central point. In hierarchical diffusion, the direction of diffusion is from the largest and most important cities in a central place urban hierarchy to progressively smaller and less important ones. Relocation diffusion involves a sudden transfer over hundreds or thousands of miles with no impact on the intervening territory. FIGURE I THREE TYPES OF SPATIAL DIFFUSION Contact Diffusion 0 fax 0 O O /. 1J0” o O a o o Hierarchical 5" 0 OR./ Diffusion 0 36 O . Y 0 O O O 0 Relocation Diffusion Spatial diffusion of innovations are also applicable in disease diffusion. Three basic types of diffusion, which are shown in Figure l, are expansion or contact diffusion, hierarchical diffusion and relocation diffusion.10 An infectious disease may Spread in any of these ways. Where transportation is evenly developed, a disease may spread outward like a wave from a central focus. This is a form of contact diffusion. If infection enters the largest city in a central place urban hierarchy, it is likely to diffuse in a stepwise manner down through the hierarchy. The long distance jump of an infection by means of a traveller on an airplane or other means of rapid transportation is essentially relocation diffusion. The amount and types of human interaction are crucial for both innovation and disease diffusion. Interaction is governed by such factors as distance, intervening opportunities, and the distribution of attractions and facilities in Space. The configuration of barriers and channels, both natural and man-made, is a very important determinant of diffusion patterns. The logistic curve of adoption roughly corresponds to the cumulative epidemic curve in a community. Mathematical epidemiologists have developed models to predict the spread of epidemics through Space and time. N.T.J. Bailey's Ihg_ Mathematical Theory_of Epidemics is one of the most important works in this field. A key element of the mathematical epidemiological models is the concept of a susceptible population, members of which are subject to 10 P. Gould, Spatial Diffusion, Washington: AAG. Commission on College Geography Resource Paper No.4, 1969, p.35. infection.) Immunity may be gained through exposure to infection and recovery, genetic immunity, or through vaccination or chemical prophylaxis. Unfortunately, the diversity of possible transmission media and the "iceberg effect of a preponderance of asymptomatic carriers in cholera would make it extremely difficult to use susceptible p0pulation-immune p0pulation models in the analysis of a cholera epidemic. Study,0bjectives This study focuses on the following objectives: 1. To determine and map spatial and temporal patterns of cholera diffusion in Africa since 1970. Emphasis will be placed on West Africa. Besides diffusion patterns at the national and international scales, more detailed local case studies from areas with differing environmental conditions will be examined. By examining cholera diffusion at different scales the effect of scale-sensitive factors in diffusion may be examined. For example, certain factors may operate only in diffusion within a small community, and the effects of others may be most clearly illustrated at the international scale of analysis. 2. To determine the key factors governing the patterns of diffusion. This requires a consideration of diffusion principles and their application under West African circumstances. A careful consideration is also needed of the cholera transmission cycle and its interactions with elements of the West African environment. 3. To determine ways in which this study of cholera in Africa can be shown to confirm, extend or refute existing spatial diffusion concepts and models. 4. To construct a systems model showing the key relationships involved in the survival and diffusion of cholera in Africa. 5. To determine areas where cholera has remained endemic following the initial epidemic invasion, and the characteristics of these endemic areas. 6. To use correlation analysis and analysis of variance to identify variables related to the observed diffusion pattern for selected countries. Data Sources and Data Reliability Cholera is one of four diseases which are governed by inter- national regulations requiring prompt reporting of confirmed and suspected cases to the World Health Organization. The number of cases and deaths for each reporting country is published each week in the Weekly Epidemiolo- gical Record. These reports are usually broken down on a district, county or provincial basis for each week. The Weekly Epidemiological Record, volumes 46 to 50 of 1970—74 comprised the main source of data. Published reports of specific cholera 12 13 outbreaks from newspapers, newsmagazines and medical journals formed 12 The following newspapers were employed: New Nigerian (Kaduna, Nigeria), The Daily_Times (Lagos, Nigeria), Gaskiya ta Fi Kwabo (Kaduna, Nigeria) and Fraternité Matin (Abidjan, Ivory Coast). 3 in particular, West Africa and Africa Digest. 10 a secondary source of data. The date of onset or of the first reported case from each political unit was recorded as well as the weekly reports of cases and deaths. The data, particularly those on the number of cases and deaths, are subject to various limitations. Four of these, which are discussed below, are lack of information on symptomless carriers, size of the reporting political units, failure to report the presence of cholera, and inaccurate information on the number of cases and deaths. (a) Lack of information on symptomless carriers: Cholera characteristically has up to 100 asymptomatic carriers for every patient with clinical symptoms.14 Unless there is very exten- sive bacteriological investigation of the general population, these asymptomatic carriers remain unrecognized. The carrier state may exist in a location and subsequently spread to other locations with no clinical case developing.15 There have been very few published studies focusing on the incidence of cholera carriers in Africa. (b) Size of the reporting units: The data published in the Weekly Epidemiological Record are typically at the district, county or provincial level. However, the 14 W.H. Mosley, "Epidemiology of Cholera”, in Principles and Practice of Cholera Control. Geneva: WHO 1970, p.26. 15 Joint ICMR-GWB-WHO Cholera Study Group, Calcutta, India. "Serological studies on cholera patients and their household contacts in Calcutta in 1968", Bull. WHO, vol.43 (1970c), pp.389-399, and 6.1. Forbes, J.D.F. Eockhart, and R.K. Bowman, "Cholera and nightsoil infection in Hong Kong, 1966", Bull. WHO, vol.36 (1967), pp.367-373. 11 level of reporting is inconsistent. Some Nigerian states, such as Lagos provide a divisional breakdown of cholera incidence, while others give only provincial or state totals. After the initial epidemic phase, Ivory Coast provided only a monthly national total. The large areal extent and p0pulation of some political units presents another problem. For example, Sokoto Province, Nigeria had a 1963 population of 3,193,01916 and an area of 36,477 square miles. When cholera is reported from "Sokoto Province", we have a very imprecise idea of its actual location and extent. Another problem arises from the wide range of size of popula- tions and areas among the political units of a study area. Other things being equal, a large division has a greater probability of being infected than a small one. Thus, the diffusion pattern mapped is largely a function of the areas and populations of the political units in the study area. (c) Failure to report the presence of cholera: The lists of cholera infected territories were frequently incomplete. In some cases, cholera was denied or not reported for such reasons as prestige and minimizing economic and social disruption. Guinea refused to admit the presence of cholera, and left the World Health 17 Organization in protest after WHO unilaterally announced the outbreak. Imported cases of cholera originating in Tunisia were reported from 16 Nigeria Year Book 1971. Lagos: Times Press, 1971, p.27. ‘7 west Africa, Sept.5, 1970, p.1039. 12 18 although Tunisia was Great Britain, Germany and Italy late in 1973, reporting no cholera at the time, probably to protect its tourist industry. Especially in rural areas lacking medical facilities, individual cases or even fairly sizeable outbreaks doubtlessly remained undiagnosed and unreported. (d) Inaccurate information on the number of cases and deaths: The actual number of cases and deaths far exceeded the reported number in several countries. Félix has estimated that the actual inci- dence of cholera in Nigeria, Mali and Dahomey was five to ten times the 19 This sometimes resulted from deliberate under- reported incidence. reporting to minimize adverse publicity. Several countries, including Liberia, Ivory Coast, Mauritania and certain Nigerian states furnished data to World Health Organization only sporadically, with the result that the apparent incidence of cholera is much smaller than the actual number of cases. Many thousands of cases occurred which were not reported to health officials. This was especially true in rural areas lacking medical facilities. In a study of attitudes toward Cholera in Trengganu, Malaysia, Chen found that many cholera cases and deaths were not reported ‘8 Week1y_Epidemiological Record, vol.48 (1973), p.358 and pp.363-364. 19 H. Félix, "Le développement de 1'epidemie de choléra en Afrique de l'Ouest”, Bull. Soc. Pathol. Exotique, vol.64, No.4 (1971C), pp.565 and 572. 13 because of skepticism about Western treatment methods and a fear that the authorities would interfere with traditional burial ceremonies and the functioning of markets.20 Similar fears of the consequences of reporting cases may well have existed in Africa. For example, the first chcilera cases in Togo were among Ghanaian fishermen. They returned at nigiht to Ghana with cholera patients and victims' bodies to avoid being caught by Togolese authorities.2] Another source of error in the Weekly Epidemiological Record 1nor~bidity and mortality data is the lack of a standardized procedure on 11ac2teriological confirmation of cases. In some territories, for example, Lagos State, Nigeria, all suspected cases were bacteriologically checked. Many territories (Driljy 26% of the suspected cases were cholera-positive. Analysis and comparison of reported only the number of suspected cases. rnc>r~t>idity data submitted with such inconsistent criteria would be difficult earici the results very tenuous. The major limitations of the data for the study have been The limitations apply in particular to considered in this section. Where cholera was diagnosed and reported morbidity and mortality data. \ I , F)..C. Chen, "Socio-cultural aspects of a cholera epidemic in Trengganu, 2(3 Eilaysia", Trop. Geog. Med., vol.23 (1971), p.298. ‘3 .J. D'Almeida et al., "L'épidemie de cholera au Togo", Medecine d'Afrique 231 ()ire, vol.20, Nos.8-9 (1973), p.639. 2322 , I361da, B.A.A., "First sixteen weeks of cholera in Lagos State", éngSoc. Hlth. Nigeria, vol.6,No.3 (1971B), p.133. 14 the deviation of the reported date of onset from the actual date is thought to be relatively small. There are certain countries and regions where relatively good morbidity and mortality are available, especially where detailed local epidemiological studies have been made and published. However, the nature of data from the Weekly Epidemiological Record does restrict the possibilities for statistical analysis. Data limitations should also be kept in mind when evaluating conclusions reached in the study. Interpretation of Diffusion Arrows Several of the maps in the thesis are designed to Show the major cholera diffusion routes. In the majority of cases, there are no let)T ished descriptions of particular outbreaks to provide conclusive evidence of the origin of infection. Where such evidence was unavailable, a SSLthjective selection was made of the probable origin of infection. By considering the distribution and seriousness of outbreaks in the three 0'" “Four weeks preceding and the amount of interaction between these Previously-infected places and the newly-infected location, a "most Pr‘Obable" origin was selected. Estimating the amount of interaction 1nV01 ves consideration of several factors, including total population, urban population, transport routes, diffusion barriers, urban function, the f1 ow of trade goods, seasonal p0pulation movements, ethnic distribution, i . . . . . . nteY‘national frontiers, and intervening opportun1t1es. 15 It is thought that the maps convey a relatively accurate picture of the actual diffusion system. Published reports describing the sequence of infection in particular areas seem to confirm that actual diffusion patterns bear a close resemblance to the expected pattern based on inter- actiorL However, it is important that individual arrows should be inter- pretmed probabilistically rather than deterministically. 16 CHAPTER 2 CHOLERA Causative Agent Cholera is caused by vibrio cholerae, curved rod-like bacteria of the genus vibrio.1 Cholera vibrios possess a complex chemical structure. 0n the basis of their antigenic properties, they may be divided into three serotypes. These are called Inaba, Ogawa and Hikojima.2 Evidence of spontaneous sero—conversion exists. For example, during the West African cholera Ogawa outbreak, vibrios of the Inaba serotype were 3 The East isolated in Southern Caneroun, Chad, Niger, and Liberia. African epidemic began as Inaba, but Ogawa vibrios were later found in Uganda and Western Kenya.4 The several types of non-agglutinable or NAG vibrios present special problems of classification.5 They possess most of the characteristics of agglutinable vibrios, but are usually non—pathogenic. NAG vibrios are conmonly found during the final stage of a cholera outbreak, probably develOping as a result of mutation. 1 R. Sakazaki, "Classification and characteristics of vibrios", in Principles and Practice of Cholera Control, Geneva: WHO, 1970, p.33. 2 R. Pollitzer, Cholera, Geneva: WHO, 1959, pp.232-234. 3 B. Cvjetanovic and D. Barua, "The seventh pandemic of cholera", Nature, vol.239 (Sept.15, 1972), p.138, and Felix, H., "Le cholera Africa1n”, Médecine Trgpicale, vol.31, No.6 (19710), p.627. 4 D. Barua, "The global epidemiology of cholera in recent years", Proc. Royal Soc. Med., vol.68, No.5 (May, 1972), p.425. 5 D. Felsenfeld, The Cholera Problem, St. Louis: W.H. Green, 1967, pp.29—30. 17 Cholera E1 Tor is a biotype of classical vibrio cholerae. It was first found in 1906 among pilgrims returning from the Mecca pilgrimmage at the El Tor health control station in Sinai.6 It was next found in Celebes (Indonesia). From 1937 to 1949 there were occasional minor out- 7 Isolated occurrences of breaks of cholera-like diarrhoea in Celebes. cholera El Tor were also reported from Djakarta, Singapore, and India.8 Prior to 1962, it was considered a separate disease, known as paracholera E1 Tor, because of its less severe clinical symptoms. It was reclassified by the World Health Organization in 1962 as vibrio cholerae biotype E1 Tor and hence became subject to international regulations.9 Concern about cholera E1 Tor had grown as a result of its spread in a pandemic form from Celebes commencing in 1961. The designation vibrio cholerae biotype El Tor was supported by Hugh as a result of his investigations which showed it to be identical to classical vibrio cholerae in thirty positive and twenty negative characteristics, and different in only minor ways. Differences in the epidemiology and clinical picture of classical and E1 Tor cholera will be discussed later in this chapter. 6 M. Khan, K.J. Bart and Z. Haq, "The changing pattern of cholera in East Pakistan: the appearance of Vibrio cholerae", J. Pakistan Med. Assoc., vol.20, No.2 (1970), p.43. 7 C.E. DeMoor, "Paracholera (El Tor)", Bull. wno, vol.2 (1948), pp.5-l7. 8 O. Felsenfeld, "Some observations on the cholera E1 Tor epidemic in 1961-62", Bull. WHO, vol.28 (1963), p.289. 9 D. Barua, and B. Cvjetanovic, "Cholera during the period 1961-70", in Principles and Practice of Cholera Control, Geneva: WHO, 1970, p.16. 10 R. Hugh, in Proceedings of the Cholera Research Symposium, PHS publication 1328, 1965. 18 Clinical Symptoms Cholera infection usually results from the ingestion of an infected medium such as food or water. The illness may take various forms along a continuum from carriers displaying no clinical symptoms, to mildly symptomatic cases, to very severe, fatal illness. Various factors govern the severity of illness. These include the level of immunity of the victim, his general health and nutritional status, the quantity of vibrios ingested, the virulence of the vibrio strain, the length of the incubation period, and the speed of diagnosis and treatment. The incuba- tion period is generally one to five days, but is occasionally longer.n Cholera patients may display a variety of clinical symptoms. Typically, there is a sudden onset of effortless and profuse diarrhoea with a "rice water stools" appearance. Vomiting occurs among about 80% of the cases.12 As the diarrhoea and vomiting continue, signs of dehydra- tion and disturbed electrolyte ~balance become apparent. These symptoms include clammy skin, sunken eyes and cheeks, a drop in body temperature and blood pressure, rapid respiration, cramps, and marked saline depletion.13 The progression of symptoms usually takes about five to twelve hours. If these symptoms are untreated, mortality may be 60% or higher. However, relatively Simple treatment involving rehydration with saline solution 1] Pollitzer, op. cit., pp.684—686. 12 Felsenfeld, op.cit., 1967, p.58. 13 Felsenfeld, op.cit., p.58. .i- u« ‘r- .~ . .6.‘ . . ”J“. '1’. 1 n ‘Ph l9 and the use of antibiotics to control vibrio multiplication may reduce the mortality to almost nil.14 Cholera does not assume the severe clinical picture outlined above in the majority of those infected. Most are carriers who have been infected by cholera vibrios and who may in turn reinfect the environ- ment but who have no clinical symptoms such as vomiting or diarrhoea. Carriers may be classified as incubatory carriers, convalescent carriers, and symptomless carriers.15 Incubatory carriers are persons who become infected and later develop clinical symtoms. Convalescent carriers are persons excreting vibrios during the period of their recovery from clinical cholera. Symptomless carriers generally excrete vibrios, but do not develop any clinical symptoms. The carrier state normally lasts from one to five days. However, the carrier state with E1 Tor infections lasts, on the average, 16 Long term, three times as long as that of classical vibrio cholera. intermittent carriers have been found. The most publicized long term carrier is “Cholera Dolores" from the Philippines, who has continued to 14 W.H. Mosley, K.J. Bart, and A. Sommer, ”An epidemiological assessment of cholera control programs in rural East Pakistan", Int. J. Epid., 1972, p.10. 15 R. Sinha et al., "Cholera carrier studies in Calcutta in 1966-67". Bull. WHO, vol.37, No.1 (1967), pp.90-91. 16 W.H. Mosley, K.J. Bart, and A. Sommer, 0p.cit., pp.9-lO. 20 excrete vibrios intermittently for over ten years.17 A comprehensive program of testing for cholera carriers and environmental contamination in Calcutta located several households in which the cholera infection seemed to be ingrained, repeatedly reappearing in a Carrier or the environ- ment despite the complete absence in the area of clinical cholera cases.18 The percentage of asymptomatic carriers is much higher with infections of vibrio cholerae biotype El Tor than with classical yjprjp_ cholerae. There are normally five to ten carriers for every clinical 19 cases of classical cholera, and 25 to 100 carriers for every E1 Tor case. This "iceberg effect" of numerous unrecognized infections is a major problem of cholera control, especially in outbreaks of cholera E1 Tor. Immunity Some degree of immunity may be obtained as a result of vaccination, recovery from an attack of cholera, and through the barrier effects of saliva, gastric acidity, and bacterial competition in the intestine.20 17 J.C. Azurin, et al., ”A long-term carrier of cholera - Cholera Dolores”, Bull. WHO, vol.37 (1967), pp.745-750. 18 R. Sinha et al., "Role of carriers in the epidemiology of cholera in Calcutta", Indian J. Med. Res., vol.56, No.7 (1968), pp.964-978. 19 W.H. Mosley, "Epidemiology of cholera", in Principles and Practice of Cholera Control, Geneva: WHO, 1970, p.26. 20 J. LeViguelloux, and J.C. Doury, "Epidémiologie du choléra moderne", Médecine Tropicale, vol.31, No.1 (1971),p.27. 21 Vibrio cholerae is acid—sensitive, and it would appear that stomach acids must be at least temporarily neutralized before cholera can survive.21 Stomach acid is governed by physiology, diet, and intestinal parasites. ‘Persons with heavy worm loads and inadequate diets are more susceptible to cholera.22 Eating a high protein meal may also neutralize stomach acid.23 Cholera antibody levels rise after the ingestion of cholera vibrios. Even inapparent infections may cause a significant rise in antibody titer.24 The longer the period of excretion the higher the probability of a significant increase in antibodies. In cholera-endemic areas, there is a linear increase of vibriocidal antibody with age.25 Therefore, children are moSt frequently affected when cholera occurs in an endemic area, while in newly-infected areas, adults are commoner among the early victims.26 2IS. DeMaeyer-Cleempoel, “Quelques réflexious a propos du cholé?a...", Arch. Belg, Med. Soc., Vol.30, No.6 (1972), p.381. An experiment showed that after neutralization of stomach acid among volunteer subjects, ingestion of 106 cholera vibrios caused heavy diarrhoea, while among those with un-neutralized stomach acid a dose of 1011 vibrios caused only light diarrhoea. 22C.K. Wallace, et al., "The 1961 cholera epidemic in Manila, Republic of the Philippines", Bull. WHO, vol.30 (1964), p.808. 23C.C.J. Carpenter, "Pathogenesis and pathophysiology of cholera", in Principles and Practice of Cholera Control, Geneva: WHO, 1970. PP. 53-54. 24W.E. Woodward and W.H. Mosley, "The spectrum of cholera in rural Bangladesh. II Comparison of E1 Tor Ogawa and classical Inaba infection", Am. J. Epidem., vol.96 (1972), pp.345-349. . 25W.H. Mosley, A.S. Benenson, and R. Barui, "A serological survey for cholera antibodies in rural East Pakistan", Bull. WHO, vol.38 (1968), pp.327-334. None of the sampled Pakistani children under one year had cholera anti- bodies, while 90% of the adults over 30 did. A control experiment in a non-endemic area (Czechoslovakia) showed no change in antibodies with age. 26W.H. McCormack, et al., “Endemic cholera in rural East Pakistan", Am.J. Epidem., vol.89 (1969), p.403. 22 Vaccination for cholera is a controversial subject. Available vaccines give only about 50% protection for three to six months.27 While vaccination may reduce the number of severe cases, it has been claimed that there is no reduction in the number of vibrio excretors or the length of excretion.28 Cost-benefit analysis has been used to show that, in most outbreaks, the small protection given by vaccination cannot economically justify its use. It is cheaper instead to treat clinical cases . 29 Transmission Cycle Man is the only known natural host of vibrio cholerae. The infection is maintained as a result of the establishment of a man— environment or person to person transmission cycle. The vibrios, which multi- ply in the gut of a carrier or cholera patient are excreted. Where environ- mental sanitation is poor, the vibrio-carrying feces may directly contaminate the soil, water, or food. The transmission cycle is completed when a new subject ingests the contaminated medium and becomes a carrier. Environ- mental contamination may also be indirect, such as vibrio transmission by 27 A.M. McBean et al., "Comparison of intradermal and subcutaneous routes of cholera vaccine administration", Lancet, Mar.4, 1972, p.528, and, . L. La Peyssonnie, ”Chemioprophylaxie de l'infection cholerique: interet, espoirs, et limites", Médecine Tropicale, vol. 31 (19718), p.127. 8 La Peyssonnie, op.cit., p.128. 9 D. Barua, "Cholera vaccination as a tool for cholera control", Bull.Soc. Pathol. Exotique, vol.64 (1971), pp.652-659. 23 FIGURE 2 CHOLERA TRANSMISSION CYCLE Cholera may spread directly from the infected to the susceptible population (person to person spread) or indirectly as a result of environmental contamination and subsequent ingestion of the infected medium by members of the susceptible population. 24 FIGURE 2 CHOLE RA TRANSMISSION CYCLE uoumndod Convalescent I \ Carrier Q Case Threshold 0 - of Infection 0 $3 Incubatory 7.: / 8503 Pl 11 JUL I a,qudeosng 1911103 Aqunou: uouoam ,0 181.110 waosamnuog (9 Susceptible Population i3 / 89 0 69 Source: LeViguelloux 8- Doury. I971 25 'flies from contaminated nightsoil to food, or from contaminated nightsoil fertilizer to vegetables grown with it.30 Cholera may also be spread directly from person to person. This commonly occurs when a patient is being treated for cholera at home, or 31 during the ritual preparation for burial of the body of a cholera victim. Cholera vibrios may survive on the body or clothing of a cholera victim and be passed to the hands of anyone in close contact. These vibrios are then ingested as a result of hand-mouth contact or the contamination of food or water. Cholera patients excrete several liters each day including abant 106 to 108 vibrios per milliliter.32 Thus contacts of cholera victims rur1 a considerable risk of infection. The survival of cholera vibrios in the environment depends on the: favorability of the transporting medium. Three important factors in sur~vival are moisture content, temperature, and acidity. Cholera vibrios 33 carnnot survive in high temperatures, although they are resistant to cold. 0 . x). Cohen, et al., "Epidemiological aspects of cholera E1 Tor in a non- €n1demic_area”, Lancet, July 10, 1971, pp.86-89. The article attributes ‘ttue spread of cholera in Jerusalem in 1970 to the sale of nightsoil- 1rifected vegetables. 31 [-63 Viguelloux and Doury, op.cit., pp.24-25. :12 £13 [-53 Viguelloux and Doury, op.cit., p.19. Felsenfeld, op.cit., 1967, p.32. 26 lhey are also sensitive to acid. A slightly alkaline medium, such as brackistilagoon water, is ideal.34 El Tor vibrios usually remain viable Under ideal conditions, they may survive for two for two to three weeks. 35 Vibrios weeks or more in foods such as milk, seafood and cooked rice. usually die within hours on acidic fruit. Sunlight, drying, and ultra- violet light are all fatal to cholera vibrios. In many areas with a long history of cholera occurrence, the disease displays a distinct seasonal pattern. However, areas with apparently identical seasonal climatic regimes may display different cholera regines. For example, cholera in Dacca peaks late in the monsoon and (ihsappears during the hot, dry season. The pattern in Calcutta is the opuaosite.36 The two cities are located relatively close to each other in the Ganges delta . Vibrio cholerae biotype El Tor survives for a longer period in the: environment and under more adverse conditions than does classical vibrio ctualerae.37 As a result, El Tor outbreaks are harder to control, and endemicity is more common than with classical cholera. 34- . ti. Wollf, "A quantitative approach to the epidemiology and control of Cflnolera", Bull. Soc. Pathol. Exotiqge, vol.64, No.5 (1971), pp.583-584. £35 erlsenfeld, op.cit., Ch.2, 1967, p.33 and D. Barua, I'Survival of cholera V“ibrios in food, water and fomites", in Principles and Practice of (Iiiolera Control, Geneva: WHO, 1970, pp.29-31. 1365 P4c>sley, op.cit., 1970, p.24. :3)? . ’\.. Sommer and W.E. Woodward, "The influence of protected water SUpleeS (3'1 the spread of classical/Inaba and E1 Tor/Ogawa cholera in rural East Benga1.", Lancet, Nov.1l, 1972, pp.985-987. 27 Control Measures38 Cholera is a very adaptable disease which can survive in a variety of environments and be transmitted in a variety of ways. With cholera E1 Tor, the large incidence of symptomless carriers relative to symptomatic cases, and its ability to survive in adverse environments multiplies the problems of cholera control. The key to control is environmental sanitation. Where there is adequate sewage and garbage disposal, it is improbable that a cholera transmission cycle could become established. However, modern sewage facilities are lacking in many large cities in the Third World, much less in the rural areas. Measures such as improved garbage collection, fly control, and construction of simple pit latrines are effective control measures. Sources of drinking water, whether wells, ponds or a piped water system may be protected by chlorination. Restrictions of large gatherings for festivals or markets is frequently used as a control measure. A large gathering can easily promote an epidemic eXplosion by bringing carriers into contact with a large sus- ceptible population from different locations. One of the earliest control nmasures was establishment of a travel cordon around an infected town or along an international boundary. These cordons have usually failed to stop travel and hence the dissemination of cholera.39 \ Extensive discussions of control measures are found in Pollitzer, 0p.cit., pp.893-987 and Felsenfeld, op.cit., 1967, pp.108-152. 39 Pollitzer, 0p.cit., p.965. 28 Public education is a key element of a cholera control program. The public must be told about the symptoms of cholera and the importance of environmental sanitation in stopping its transmission. Mass vaccination of the susceptible population is a popular control measure, largely because of it is an easily understood and visible approach. However, it is costly, limited in its effectiveness, and may instill a false sense of security in the population. Constant surveillance and bacteriological examination of suspected cases and their contacts, water sources and nightsoil are also necessary. It is only through such surveillance that early recognition of an outbreak or the determination of the means of infection transmission is possible. 29 CHAPTER 3 HISTORY OF CHOLERA Occurrence of Cholera Before 1817 Controversy surrounds early reports of cholera-like illnesses made by Greek, Indian, and Chinese physicians. There is no consensus on whether the descriptions are of cholera or other diseases with similar symptoms. Extensive reports of a cholera-like disease are found in the writings of the Indian physician Shushruta in the fifth century A.D.1 The earliest definite evidence is from the journals of European travellers to India following Vasco da Gamas pioneering voyage in 1498. Gaspar Correa, in Lendas da India described an illness which are almost certainly cholera which had been observed in 1503 and 1543 in India.2 There are many more well-documented, independent references to cholera outbreaks in various parts of the Indian sub-continent from 1503 to 1817. Before the nineteenth century, cholera did not spread westward from India. However, whether or not it had moved eastward into China and the East Indies before 1817 is less certain.3 Nineteenth Century Cholera Pandemics The year 1817 marked a crucial turning point in the history of cholera. The years 1815 and 1817 were exceptionally rainy, causing \— L.M. Bhattacharji, et al., "Changing phases of cholera in India", Indian g;_Med. Res., vol.52 (1964), p.751. 2 R. Pollitzer, Cholera, Geneva: WHO, 1959, pp.12-l3. 3 Pollitzer, op. cit., pp.14-15. 30 floods and crop failures. The weather in 1816 was extremely hot and dry. During this period, cholera assumed such a violent form that it was thought to be a new disease.4 It also spread out of the Indian sub-continent into South-East and East Asia and the Middle East. This marked the beginning of the first cholera pandemic. During the nine- teenth century there were six pandemics in which cholera spread from country to country and continent to continent. In each case, the pandemics originated in the Indian sub-continent. After epidemic outbursts in the newly-infected countries, the infection tended to disappear for a few years until reintroduced. Table 1 shows the extent of cholera infection during each pandemic and the approximate dates of the pandemics.5 Transportation routes and commercial centers played key roles in the dissemination of cholera during each pandemic. Caravans and sail- ing ships were especially important. Cholera reached the Shores of the Mediterranean from China via the Central Asian caravan route during the second pandemic.6 Mecca was repeatedly infected by caravans arriving 7 from Persia. In East Africa, caravans were responsible for bringing cholera from the Red Sea coast inland and south to the Somali coast.8 4 Pollitzer, 0p. cit., pp.17-18. 5 Maps of the diffusion and extent of cholera during each pandemic are found in J. May, The Ecology of Human Disease, New York: MD publications, 1958, pp.39-43. A’relatively detailed account of the occurrence of cholera and routes of diffusion is contained in Pollitzer, op.cit.,pp.l7-48. 6 May, 0p.cit., pp.38-40. 7 J. Christie, Cholera Epidemics in East Africa, London, McMillan, 1876, p.79. 8 Christie, op.cit., p.103. 31 TABLE 1 Extent of the First Six Cholera Pandemics A if. 3 +3 ft! r6 0) g- ‘1 CU I'D «‘6 W C °r- LLI (D D. (0 I‘d (U U U U OJ ‘5 U1 0- O O U U 'U 0r- r- rr- c: C -r- < (D O L L -r- .,... -r- L L L I'U'l— U) r-- S- : 3 ‘- L L- O) Q) .,... 4-: <1: 4-» “o :5 Lu Lu '4— u— H— g E -o c m '0 1.14 < < <1: «a: < E , C O LLJ (U 'r- 3 LLJ Pandem-IC Date H U (I) Lu 2 (I) Z Z Z Lu 3 Z L) U) First 1817-23 x x x x x Second 1826-37 x x x x x x x x x x x x Third 1842-62 x x x x x x x x x x x x Fourth 1865-75 x x x x x x x x x x x x x Fifth 1881-96 x x x x x x x x x x x x Sixth 1899-1923 x x x x x x Note: an "x" denotes the occurrence of cholera during a pandemic in a particular geographical area. 32 Sailing routes also served as cholera diffusion routes. Among the more important of these were the routes from India to the East India, Arabia and East Africa, from Europe to the Americas, and between the numerous ports of the European perimeter from the Baltic to the Mediterranean and Black Seas. Typically, the infection would be introduced at a sea port and then would spread inland. With the coming of the Railway Age in Europe and North America after the mid-nineteenth century, the infection diffused along railway lines from ports to their hinterlands. The most notorious cholera diffusion center was Mecca. The annual ngj_attracted tens of thousands of pilgrims from all parts of the Muslin World. Because of the lack of understanding of the causes or control of cholera, it could easily be introduced by any group of pilgrims. Pilgrims from other countries would then contact the infection and disseminate it on their homeward journeys. Between 1830 and 1912 there were thirty-three major cholera epidemics originating in Mecca.9 ‘0 has studied the patterns of cholera diffusion in the Pyle United States in 1832, 1849 and 1866. In 1832, the United States was still a frontier country lacking a well-developed transportation network. Cholera spread from Quebec and New York along the East Coast and the major Eastern river systems. Distance from the points of introduction Z May, op.cit., p.52. 10 G.F. Pyle, "The diffusion of cholera in the United States in the Nine- teenth century", Geog. Analysis, vol.1, No.1 (1969), pp.59-75. 33 was strongly correlated with the date of onset. By 1866, the United States had become a highly urbanized nation with a dense railway network. Cholera quickly spread from New York and New Orleans along railways to the largest towns and then diffused to nearby smaller centers. City size, rather than distance had become the key variable. The 1849 epidemic showed elements of both the 1832 and 1866 patterns. Mortality was very heavy during all of the nineteenth century cholera pandemics. For example, in 1866 there were over 90,000 cholera deaths in Russia, 115,000 in Austria-Hungary, 130,000 in Italy, 30,000 in Belgium and 20,000 in the Netherlands.11 During the nineteenth century two discoveries were made which had great significance for the possibilities of cholera control. The 'first was John Snow's discovery in 1849 that cholera was Spread by water and that epidemic control could be achieved by st0pping the distribution of 12 infected water. Robert Koch made the second important discovery in 1883, namely that cholera is caused by bacteria.13 These two findings formed the basis for subsequent cholera control strategies. Cholera in Africa during the Nineteenth Century Egpt Africa The cholera epidemics of East Africa prior to 1870 have been described in considerable detail in an 1876 book by Christie.14 Cholera ]] Pollitzer,op.cit., p.33. J. Snow, Snow on Cholera, New York: Commonwealth Fund, 1936. May, op.cit., p.35. Ctnjstie, op.cit. 34 was first noted in East Africa in 1821 when a slave ship arrived from 15 Zanzibar via Muscat and Oman with cholera on board. The second cholera invasion was in 1836-37 following the Mecca pilgrimmage. Dhows engaged in the coastal trade carried the infection from Arabia along the East 16 It was African coast to the Somali Coast, Zanzibar and Mocambique. also carried along caravan routes into East and Central Africa. During the Third Pandemic (1852-68) the West African coast was infected several times. Abyssinia was reached in 1853, 1855 and 1858. Speke referred to the frequent occurrence of cholera in Central Africa 17 Cholera arrived in Zanzibar from 18 (Uganda) in his journal of 1858-59. Oman on the ship Maryland on November 15, 1865. Very heavy mortality was reported in Zanzibar and the adjacent mainland around Kilwa. The diary of the explorer, Burton, included a very vivid picture of the impact of the epidemic. It also illustrates the prevailing lack of understanding of cholera epidemiology, particularly the need for environ- mental sanitation, ensured massive mortality. "The smooth, oily water was dotted with fragments and remnants of humanity; black and brown when freshly thrown in; patched, mottled and party coloured when in a state of half-pickle; and ghastly white, like scalded pig when the i mentum nigrum had become thoroughly macerated.“19 ‘Z 15 Christie, op.cit., pp.97-98. 6 Christie, op.cit., pp.102—103. 17 Christie, op.cit., p.107. 8 Christie, op.cit., p.111. Christie, 0p.cit., p.114. 35 Christie's most detailed account is of the 1864-71 East African epidemic. The pattern of diffusion is shown in Figure 3. Cholera was introduced to Berbera in November, 1864 by a ship which had probably come from Bombay. It then moved along caravan routes to the coast in the vicinity of the Jub River. In May, 1865 Jidda and Mecca were infected by a ship coming from Singapore via India. The most severe epidemic in the history of Mecca ensued.20 From Abyssinia in 1865, it gradually Spread south, reaching the Masai country in 1869. The Masai spread it to the south during raids on neighboring peoples. A party of traders carried the infection from Laikepya to the coast near Zanzibar in 1869. Of the 150 21 men in the party, only seven survived the trip. Cholera's arrival in the headwaters of the Congo beyond Lake Tanganyika was recorded by Living- stone.22 Livingstone apparently contacted cholera infection himself, but recovered. Traders carried the disease south to Lake Nyassa and on to the Mogambique coast in 1870. It spread from Mocambique north and south along the coast and to the Comoro Islands and Madagascar. In 1869, Zanzibar had been infected and sustained heavy mortality. In a single year, cholera claimed 70,000 victims in Zanzibar.23 Ships took water on board in Zanzibar from infected ponds and wells near 24 the harbour before setting sail. As a result, cholera was reintroduced to the East African coast from Cabo Delgado to the Horn in 1870. 20 21 22 23 24 Christie, op.cit., p.150. Christie, 0p.cit., p.228. Christie, 0p.cit., p.245. Pollitzer, op.cit., p.340. Christie, op.cit., p.420. 36 FIGURE 3 CHOLERA IN EASTERN AFRICA 1864-1871 This is a slightly simplified version of a map appear- ing in an 1876 book by James Christie on cholera empidemics in East Africa to 1870. It shows the southward movement of cholera from Berbera and Northern Abyssinia along caravan routes, eventually reaching Zanzibar, the headwaters of the Congo and Mocambique. Trading ships also carried cholera along the East African coast. The numbers on the map represent the year of onset in a particular area. 37 6190i?! Jiddo Mecca ‘5 (rm76,65) Suokuo N ARABIA Khartoum Socotra Is. - AIYSSINIA Ierbero 'k) SOMALI Nile loikepya |.. Victoria IN EASTERN AFRICA Innloarre 1864 - 1871 l. tangcnyilla ' 69 Gazembel omusrou IOUIE _='— MILES Zombili l - ' M0‘3mbique M T . K anarorwe Sour" — Christie. 1.76- 38 Several of Christie's conclusions are noteworthy. He emphasized the importance of highways of human travel for the propaga- tion of the disease. He noted that cholera dissemination peaked during 25 the dry season, the rainy season being a "time of isolation". He also stressed the importance of fixed local foci of dissemination, particularly where the cholefia generative agent was discharged into more or less stag- nant water. He concluded that because cholera is essentially a disease of filth; "a community can secure immunity by sanitary measures alone."26 Between 1879 and 1900, only the Red Sea and Somali coasts reported cholera in EaSt Africa. North Africa North Africa was subjected to repeated introductions of cholera during the Second through Fifth Pandemics. There were two main . . 27 sources of infection. Cholera was frequently introduced to the Alexandria-Cairo area by returning pilgrims from Mecca. It typically Spread UP the Nile into Upper Egypt and sometimes the Sudan. Tripolitania (Libya), Tunisia, Algeria, and Morocco were also infected via Egypt by epidemics originating in Mecca. The second source of infection was Mediterranean trading ships calling at North African ports. The 1867 epidemic in Tunisia was blamed on Tunisian smugglers returning from Sicily.28 25 26 27 28 Christie, op.cit., p.502. Christie, op.cit., p.505. Pollitzer, op.cit., p.28 and p.32. Pollitzer, 0p.cit., p.34. 39 Heavy mortality was frequently reported, especially from populous areas in Upper Egypt, Algeria and Morocco. For example, 80,000 died in Algeria in 1867 in an epidemic of French origin, and over 58,500 cholera victims were recorded in 1883 in Egypt.29 West Africa Cholera was introduced to Podor along the Senegal River by 30 caravans from Morocco in 1868. It then Spread to St. Louis, Bathurst (Gambia) and Bissau (Portuguese Guinea) in 1869. A reported 1,700 of the 5,000 residents of Bathurst died of cho1era.3' 32 It again arrived in the Senegal Valley by caravans in 1893-94. Worldwide Incidence of Cholera from 1923-1960 Between 1923 and 1960 cholera ceased to exhibit any pandemic tendencies. During this period there were no cholera deaths in EurOpe.33 In South-East and Eastern Asia the number of cases declined, and after 1945 only sporadic, minor outbreaks were reported. There was a steady decline in the number of deaths due to cholera in the Indian sub- continent. Between 1930 and 1949 there were 3,580,000 cholera deaths in 29 Pollitzer, op.cit., p.38. 30 ' Pollitzer, op.cit., p.34. 3] R. Schram, A_History of the Nigerian Health Services, Ibadan, 1971, p.102. 32 D.J.M. Mackenzie, "Cholera, its nature, management, and prevention", S. Afr. Med. J., vol.45 (1971), p.2. 33 Pollitzer, 0p.cit., p.59. ha ' L n f3 (LL) (" r) 40 India and Pakistan, compared to 8,050,000 deaths from 1900 to 1919.34 The geographical extent of cholera in the Indian sub-cantinent also became more confined. Cholera was restricted to a few endemic foci from which it periodically spread to neighboring regions. The major endemic areas were concentrated around the lower Ganges, Bramaputra and Hooghly 35 Rivers. The periodic epidemic outbursts from the endemic areas have been found to be highly correlated with the staging of large fairs and festivals}6 The most important cholera epidemic outside the Indian Sub- continent between 1923 and 1960 was in Egypt between September 18 and December 5, 1947. There were 32,978 recorded cases and 20,474 deaths.37 The infection is thought to have been introduced in conjunction with the movement of military personnel from India to Egypt.38 The first reported cases were at the village of E1 Korein which was thehome of many workers from a nearby British military base. A major date fair, attracting thousands of traders from all parts of Egypt coincided with the cholera outbreak. Either through the contamination of the dates or the infection 34 Calculated from annual mortality data in Pollitzer, op.cit., p.58. 35 S. Swaroop, "Endemicity of cholera in India", Indian J. Med. Res., vol.39, No.2 (1951A), pp.141-184. 36 S. Swaroop, and M.V. Raman, "Endemicity of cholera in relation to fairs and festivals in India", Indian J. Med. Res., v01. 39 (1951), pp.41-49, and A.C. Banjera, "Note on cholera in the United Provinces (Uttar Pradesh)", Indian J. Med. Res., vol.39, No.1 (1951), p.31. 37 Pollitzer, op.cit., p.63. 38 S. Abdou, "The cholera epidemic in Egypt: Mode of spread", Lancet, vol.2 (Nov.8, 1947), pp.696-698. 41 of some of the traders themselves, cholera was disseminated in a matter of days throughout the Nile Delta. Smaller outbreaks were reported along the Nile as far south as Qena.39 Because of its mode of introduction and transmission, its rapid dissemination and quick disappearance, and the very large number of cases and deaths the 1947 Egyptian epidemic remains somewhat enigmatic. Cholera El Tor and the Seventh Pandemic The unexpected development after 1960 ofia new cholera pandemic, involving a local and apparently mild biotype of vibrio cholerae from the Celebes forced a sudden reassessment of the consensus that cholera was a geographically-limited and declining disease. Cholera El Tor began to spread from its home territory in Southern Celebes following movements by troops and Celebes' Chinese population and after boats from the area had participated in a regatta 40 in Sarawak. The progression of the Seventh, or E1 Tor pandemic is shown in Figure 4.4] 39 "Trend of cholera in Egypt", NEE; V01-22 (1947), PP-383'385- 40 o. Barua, and B. Cvjetanovic, "Cholera during the period 1961-70", in Principles and Practice of Cholera Control. Geneva: WHO, 1970, p.5. 41 Several accounts of the dissemination of cholera during the seventh pandemic exist, including Barua and Cvjetanovic, op.cit., pp.15-21, J. Gallut, "Actualité du cholera. Evolution des problemes epidemiolo- giques et bactériologiques", Bull. Inst. Pasteur, vol.66, No.2 (1968), p.219-248, and S. Mukerjee, "Recent incidence of cholera outside India", Indian J. Med. Res., vol.52 (1964), pp.771-776. 42 FIGURE 4 TREND OF THE GLOBAL SPREAD 0F CHOLERA 1960-1974 Classical vibrio cholerae was superseded after 1960 by cholera El Tor. Originating in the Celebes, cholera E1 Tor had by 1974 advanced westward through Southern Asia and the Middle East into Africa and limited parts of Eur0pe. The map shows the migration of the "cholera front" from 1961 to June, 1974. 43 in... 5.... use 0.82. .8329...er 583 323m 38 9.0.2.0 Vs - 30— $3040 ¢ 0453.0. 93.2.0 5.39.0 ootooé .8933: co» m s 5.63 .8... _m E0.9.0 .3 deco ooBom .333? ’---‘ ml... umDO: 8.28. H 3E . 50.320 .0 302 0.6025 . “.0 “.0 02%: \el\¢ a‘ \II - .5 sJ \\a. ifs \ o. * \ A ‘ o s 1‘ _ . so . Q .1: s I. e l.‘ o I — . I o l I a . I. \l’nl‘HHIIIIo D \ % Italy .. P. is. . ‘ . cl fill l“--.\| «PO. «NJ-I 44 During 196l cholera reached several other islands of the Indonesian archipelago, Kwangtung Province in China, Hong Kong, Macao, and the Philippines. Taiwan and West Irian were infected in 1962. In l963, Malaya, Cambodia, Vietnam, Thailand, Burma, South Korea and Japan experienced cholera outbreaks. The most serious epidemics were in Indonesia, the Philippines and South Vietnam. Late in l963 cholera El Tor first appeared in the Indian sub- 42 By l966. cholera El Tor had completely elimin- 43 continent at Chittagong. ated classical vibrio cholerae in India, although classical cholera remained dominant in East Pakistan (Bangladesh). Since l968 there has been a resurgence of classical cholera in India and an increase in the 44 The two strains now appear frequency of El Tor isolation in Bangladesh. to be coexisting in both countries. Cholera El Tor moved into West Pakistan in l964. Afghanistan, Iran and limited areas in USSR, near the Iranian frontier were infected in l965, followed by Iraq in l966. The western cholera front remained relatively stable until 1970 when it advanced into all countries in the Middle East, as well as the Soviet Black Sea coast, eastern Czechoslovakia, North AfriCa, East Africa, and West Africa. The sudden appearance and rapid diffusion of cholera in West Africa was probably the most note- worthy of these develOpments. 42 S. Mukerjee et al., "A new trend in cholera epidemiology", Brit. Med. J., vol.2 (l965), p.837. 43 S. Mukerjee, "A decade's tracking of cholera with bacteriOphage", Indian J. Med. Res., vol.55, No.4 (l967), pp.3lO-3ll. 44 K.J. Bart, Z. Huq, et al., "Sero-epidemiologic studies during a simultaneous epidemic of infection with El Tor Ogawa and classical Inaba vibrio cholerae", J. Infect. Dis., vol.l2l (l970), pp.5l7-SZ4. 45 During l97l, cholera affected all countries in West Africa from Chad to Senegal, with the exception of Guinea-Bissau and Guinea (Conakry). In East Africa, cholera occurred in Ethiopia, Somalia, Kenya, Uganda, and the French Territory of Afars and Issas. In North Africa, Morocco and Algeria reported cholera for the first time in 197l. A new focus developed in Angola in l97l. Mocambique, Malawi, Rhodesia and Tanzania have reported cholera for the first time in 1973 and T974. There are indications that cholera El Tor may be becoming endemic in certain areas of Africa, in particular in West Africa. Health officials have been somewhat disturbed by the appearance of cholera in Europe since 1970. Because of the higher standards of living and environmental sanitation, and because of fifty years of cholera-free 45 However, status, Eur0pe had been thought to be immune from infection. EurOpean USSR and Eastern Czechoslovakia had cholera outbreaks in l970, Spain and Portugal in l97l, and Italy in l973. Portugal is reported to have had 827 cases of cholera between April and July, l974.46 In addition isolated imported cases have been reported from such countries as France, Germany, Britain and Canada. Although the epidemics have been confined to the poorest areas of Southern Europe, their size and the mere fact that cholera has been able to establish itself in EurOpe has been cause for concern. 45 "Tip of the iceberg shows", Nature, vol.227 (l970), p.766. 46 Radio station CKOX, Woodstock, Ontario. Noon news report, August 3, l974. 46 CHAPTER 4 CHOLERA EPIDEMIC ZONES IN AFRICA Following the initial establishment of cholera in Africa in l970, the disease has spread to thirty-two countries. These countries form five discrete clusters or infection foci, located in West Africa, North Africa, East Africa, Angola and South-Eastern Africa (Figure 5). The major axes of diffusion in each of these areas are described in this chapter. West Africa On August l8, 1970, Guinea summoned World Health Organization officials because of the outbreak of a serious epidemic suspected to be cholera.] Guinean students returning from USSR are suspected to have brought it from the Black Sea coast.2 Coastal Sierra Leone and Liberia first reported cholera in the period of September l9-25 and October 2-8, 1970, respectively.3 The infection was carried along the coast to the vicinity of Abidjan by 1 L. La Peyssonnie, "Le cholera, An II", Médecine Tropicale, vol.3l, No.6 (197lD), p.615. 2 L. La Peyssonnie, "Acquisitions récentes en matiére d'épidémiologie et de prophylaxe du cholera en Afrique", Bull. Soc. Pathol. Exotique, vol.64, No.4 (1971A), p.644. 3 All dates of onset and numbers of cases and deaths have been obtained from the Weekly Epidemiological Record, vols.45-49 (l970-l974) unless footnotes indicate another source. 47 FIGURE 5 CHOLERA-AFFECTED AREAS IN AFRICA l970-l974 Between l970 and 1974 cholera has established five discrete fronts in its invasion of Africa. These are in North Africa, West Africa, East Africa, Angola, and South-Eastern Africa. Long distance air travel played an important role in the initial establishment of three of the fronts. Despite diffusion over long distances in each of the five cholera zones, and despite the fact that thirty-two countries have had cholera outbreaks, roughly 80% of the continent has yet to experience any cholera cases. 48 CHOL E RA - AFFECTED Portugal ‘\ C halora - free area S Unofficial report I A AAA+I l I o :00 IOOO 1500 KM. L _A_ l A l J o no 1000 FIGURE 5 AREAS IN AFRICA 1970-74 l ram U.S.S.R. Middle East .4 Arabian Paninsula .5 ‘ .I _ . -’ '3 . .1 . ' I . ' ;’ Ethiopia ‘-',i- I ~ 5 l ‘l , ‘F’ ‘ N..." l > I Liv-lh.’ l. \1 / l x “h. LJ ‘ ‘\ A; r 1 ) "s l-{‘\ €I ‘M:J~’\§_J «N 1 ' ’ _\\‘ ’ Mo ambiqua ‘ . Pl I \’\‘&~(’ I ’f' l l J’ I {I /”'J #4 'k") f'": x,’ 49 FIGURE 6 DIFFUSION 0F CHOLERA IN WEST AFRICA After the introduction of cholera at Conakry in mid- August, l970, it spread along the coast as far as Southern Cameroun by February, l97l. There were three other types of diffusion. One was the riverine epidemic centered on Mopti (Mali) and extending from central Niger in the East to the Atlantic coast of Senegal and Mauritania in the West. Nigeria and South-Central Niger experienced urban-based hierarchical diffusion. Radial contact diffusion occurred in the vicinity of Lake Chad. Guinea-Bissau was the only country to completely escape infection. Substantial areas in the Middle Belt have remained cholera-free. 50 0.0.2.0 .0 8.50.2.0: 3.0.5.3 02 a 603320 .0280 .0501! 8.02:0 30282.... 035..-: . :03:th OECOZK ._ £2.02? 000 00.00 .0200 0... .0... 023:5 .2008 I 000.0 00.. n 0.0.2.0 00:032.... 0.0.2.0 .0 4003 ® 0050. 00.0035 0 I NU ® Gite VI fi/r m-J / .. a .2... \\ . \‘It‘. i .0. an.“ .30.; mu...! 001? 00_ 0 <>m_u_ — — «<43 .20. 000 00. n g .69; 0:02.00 DJ 65030.2 0. . Os... i deIOPQO Z(_o_83h / / 000...; .0 03.26.30... .50....0 .0 07.001 000.0...0aki00000m ._ 5000004 000 00.00 .0200 2.. \\ /A 30.2.: . .0“. l / J r {I 1 w. .300 / 3 Q0208 / 9 .3.‘°2— m_:=n—I \ \ \ ®\ 0:00 . .00.: 0505 8 M, 00.0.0 9 ‘ 00.20.0000 8 as IN SOUTH - EASTERN AFRICA N f . \ // / / - / N‘khoto any __‘___,__,’\ // T.” I \JA'N” / I f I \7 LMaIauI I ff f/ ZAMBIA \\ lie/NIOIO KO'C ', CABO DELGADO I NIASSA ,‘ ’ \ lTSKSaIIrnc:\ X A Lilongwe \ ....” /” \ (I \ fL’P/flf‘fl // / \ x”, J / . 1”” ’- MANGOSHE? // (" TETE 9i MOCAMBIQUE @Ktl’s‘upe / \ I , \ Zambezi R. m \4/ ~\ \ Mocamblque n-Ilu_—\\ ,§__,""’ If ‘\‘ 1\ \\\)~ ('33 Blan rz/ Mlange ; ILE ‘\‘\ \\ 09'1"“ / ‘\f\’/"\IA\ I.“ 69‘} \ U70 I Moruba ”.1 \ \J‘ «‘67 Neanoe \\ lg: ’ ’ I ‘% (Q) \'\ . Mtoko / (ITO (I 3 . Morrumbala ~ I ,1; \ \ \‘ “ \/ ‘)Vila Fontee I02“ I53 MOPEIA | c 0 \ /~ {‘1le \' .Mopeie“ \‘ Ouellmane ' - ‘>~ ,' ‘~”‘ ‘; R‘ ‘ 2‘ I ’ ‘-CHIMOIo I" \‘@ ”0mm" x. r .55 ' ’ \xMARROMEU a...“ °°°°Z @ /" I34 . ZIMBABWE\ “nun" - \j‘ié‘ . CHERINGOMA I ® Week of cholera Introduction. I Me\lsetter M F?" V'°'°"‘ \\ . — Diffusion routes , I Beira Chlplnooyf K/ Cholera - free areas ( f3 .5." S No official notification of cholera MANI ' .' cmmm(3 CA 3' SOFAL: TETE Province . ' 02 : , .I Q .' I TETE Division / \. ... s37: - ~—/ I / \—/ Tete Town /. .. \‘ Gownrrrfl ‘ ' , .3 " P , ._ _ -I—t—l- Railway ~ "“'~/; ‘ -I ,’ —— Main road _\ - MocAMBIQUE \ “. " I \ " I i \ I \\ , lFor the actual dates ~ \‘ GAZA \‘ INHAMBANE I see Appendix I RWSA I \ { Aluminum 1 “‘5 9‘ \ I , l I -’ I. IMANNIc'A “ .... \ (r o ' “ I O / \v t g In. 2 2 o / Mbebenev I M Lorenco Marques LL. . l l I \L OR o '00 l ZOO SWAZI- OEN O IILES LAND :4" ....3, 69 railway to Cheringoma and Mutarara Districts by the first week of October. Mutarara is north of the Zambezi River adjacent to the Malawi border. From Mutarara, cholera moved north-west to Tete and south-east to Marromeu District, following the trend of the Zambezi River and the railway line occupying the Zambezi Valley. Another major outbreak began in Lorenco Marques and Manhica in Southern Mocambique in the week of January 6-l4, l974. The infection was probably introduced by air or sea from Beira, some 500 miles to the north. Zimbabwe's eastern and north-eastern borders are shared with Mocambique. Cholera has affected these border regions in Northern, Central, and Southern Zimbabwe. The first cases were reported from Mtoko in North-Eastern Zimbabwe. The fact that cholera was reported in both North and Central Zimbabwe before it had been reported from districts on the Mocambique side of the border suggests that the Mogambique epidemic may have been more extensive than what has been reported. The introduction of cholera into Zimbabwe may have been related to movements of Zimbabwe liberation troops. The first outbreak in Zimbabwe was in the area with the most intensive liberation activity. The Zimbabwe African National Union (ZANU) sent requests to numerous inter— national organizations for assistance in combating the epidemic.24 There is probably considerable movement across the Mocambique border into Tete Province where the Mocambique Liberation Front (FRELIMO) is very active. 24 J. Pearpoint, "Cholera crisis reviewed", The cuso Forum, vol.2, No.3 (April-May, l974), p.24. 70 Cholera had appeared near the Malawi border in the first week of October, l973. On October 24 it was reported from Nsanje and Chikwawa in Southern Malawi. These two locations reported a total of 332 cases in the first seven weeks. However, it was only on December 31, l973 that it began to Spread to other locations in Southern and Central 'Malawi. The key to this widespread diffusion was apparently the infection of Malawi's former capital city, Zomba, on December 3l, 1973. By January 24, it had spread to eight other locations in Southern and Central Malawi. The infection followed the shore of Lake Malawi to Nkhata Bay and Karonga in Northern Malawi. The initial cases were reported early in March, l974 in Northern Malawi. Cholera has also crossed the Tanzanian border north of Lake Malawi. Five cases and two deaths were reported on June 6, l974 from Kyella District in Mbeya Region. As of early July, l974, cholera was established in four countries in South-Eastern Africa. The infection front is also within twenty-five miles of Zambia, Swaziland and Republic of South Africa. South Africa is in particular danger of infection because of the constant movement of migrant workers from Malawi, Zimbabwe and Moqambique into South Africa. The outbreak at Lorenco Marques poses a major threat as a result of its location close to the South African border and its function as a very important part for South African trade. The seriousness of the South-Eastern African outbreak is shown by the relatively large number of cases and deaths, and the persistence 71 of the infection. A total of l,570 cases and 204 deaths (l3.0% mortality) were reported from Mocambique up to June l, l974. .Beira had 5l6 of these cases. Tete, Mutarara, Quelimane and Lourenco Marques each had over lOO cases. Malawi has reported 99l cases, but the somewhat irregular publishing of reports in the Weekly Epidemiological Record suggests that the outbreak may in fact be more extensive than officially reported. There were 535 cases and forty-two deaths (7.8% mortality) reported from Zimbabwe up to May 24, l974. Cholera was still being reported at the end of May, l974 from all three Malawian provinces, four of the seven infected provinces of Mocambique, and two of the three infected Zimbabwe districts. However, the weekly number of reported cases has been declining since late March, l974. 72 CHAPTER 5 INTER-REGIONAL DIFFUSION OF CHOLERA IN WEST AFRICA Origin of Infection The first indication that cholera had become established in West Africa came on August 18, 1970 when Guinea sent an appeal to the World Health Organization to investigate an outbreak of a disease feared to be cholera.1 It was confirmed that the disease was cholera, but Guinea refused to make any official announcement of it. When the outbreak was unilaterally announced by the World Health Organization on September 4, 1970, Guinea withdrew in protest from the organization.2 The origin of the infection has not been definitely established. There were epidemics of cholera E1 Tor serotype Ogawa in the Middle East and along the Black Sea Coast of the USSR in mid-1970. The most likely explanation is that Guinean students in the USSR, who had holidayed in the Crimea just prior to their return to Guinea introduced the disease.3 1 L. La Peyssonnie, "Le cholera, An II, Médecine Tropicale, vol.3l, No.6 (l97lD), p.6l5. 2 West Africa, September 5, 1970, p.lO39. 3 L. La Peyssonnie, "Acquisitions recentes en matiere d'epidémiologie et de prophylaxe du cholera en Afrique", Bull. Soc. Pathol. Exotique, vol.64, No.4 (1971A), p.645. An alternate explanation of'introdUctibn by pilgrims from Saudi Arabia is given by La Peyssonnie, op.cit., 1971A and in "Cholera and politics", West Africa, September 8, 1972, p.ll75. Introduction by pilgrims isgimprobable because August, 1970 was off-season for pilgrimmages. 73 From this initial focus in Conakry, cholera has spread through- out West Africa. Guinea-Bissau is the only country to escape infection. It was noted in the previous chapter that four distinct patterns of diffusion characterized the spread of cholera between West African divisions or provinces. These were coastal, riverine, urban hierarchical and radial contact diffusion. A more comprehensive account of these four diffusion types follows. In addition, the chapter includes information on the morbidity and mortality associated with outbreaks in different parts of West Africa. Coastal Diffusion The term coastal diffusion is used for the movement of cholera along the coast from Guinea to Eastern Cameroun (Figure 11). The limited penetrations inland from the coast in several countries have been included where the baSe of the infection remained in the rural coastal zone. This definition excludes the majority of the Nigerian cholera epidemic which was essentially based in major cities, particularly Lagos. 0f the several West African ethnic groups specializing in fishing, the Fanti are most prominent. This homeland is coastal Ghana, but they are highly mobile and have established themselves along the length of the West African coast. These Fanti fishermen were apparently respon- sible for carrying cholera from Conakry along the West African coast. Comencing in Conakry, the migrating fishermen moved along the coast. Their periodic stops along the coast allowed the cholera carriers 74 FIGURE ll COASTAL DIFFUSION OF CHOLERA IN WEST AFRICA Following the initial outbreak in Conakry in August, 1970, cholera was spread along the coast by fishermen. The infection reached Southern Cameroun in March, 1971. There were only limited penetrations inland, and these occurred in areas of relatively high population density such as South-Eastern Ivory Coast, Ghana and Togo. The three insets show in greater detail the patterns of diffusion in South-Eastern Ivory Coast, Togo and Dahomey and Cameroun. 75 08.2.0 ._ 59.234 03 c230=2oc .22.. catoauo..c_ 80.2.0 .0 .2050 02 a .0300 2: 8m .83 @ 83:5 8.255 Ihlii ... sumo. >U20152 25.5% 2053......20 «02:00 302320.: .euomlI :35 III. .535. ........ .288 I 8555 8.5.5 :2.an :32 fill}... all». 4.4.5400 76 FIGURE 12 FANTI FISHERMEN, CAPE COAST, GHANA The central portion of the Ghanaian coast is the Fanti homeland, but colonies of Fanti dot the West African coast from Senegal to Cameroun. The migration of Fanti fishermen expelled from Guinea along the West African coast in their canoes has been blamed for the dissemination of cholera in coastal West Africa. 77 in the group to introduce the infection to a new location. Nhere conditions permitted the survival of the vibrios and establishment of a local transmission cycle, a potential point of diffusion develOped. The impact of a particular point depended largely on its location and the amount of interaction with other places. Lagoon villages near large cities such as Abidjan, Monrovia and Accra have been particularly dangerous sites. The swampy environment and brackish lagoon water are a suitable environment for long-term vibrio survival frcm which the infection may be repeatedly introduced to nearby cities. The risk of long distance transmission is greater from the cities because of the concentration of transport routes, commerce and travellers. The role played by fresh and smoked fish in the transmission of cholera is unknown. Cholera vibrios survive readily on fish, and the large extent of the trade in fish suggests a possible role of cholera- infected fish in Spreading the disease. According to Dr. Joseph N. Togba, the migration responsible for the diffusion of cholera started as a result of their expulsion from Guinea.4 There were major expulsions of Ghanaians from Sierra 5 but the major African news magazines contained Leone in January, 1969, no record of expulsions from Sierra Leone or Guinea in mid-1970. bnpublicized expulsions may have occurred, perhaps because of the existence of cholera infection among them. 4 University of Liberia Medical School. Lecture t0 I.D.C. 390 class at Michigan State University, February 7, l973. 5 . Africa Diary, vol.9 (January 8-14, l969), p.426l. 78 Guinea, Sierra Leone and Liberia Very little is known about the extent of the Guinean epidemic. The World Health Organization estimated 2,000 cases and sixty deaths up to September 4, l970,6 which were apparently confined to Conakry and vicinity. Sierra Leone's first reported cholera outbreak was in the fishing village of Bailoh located some 100 km. north of Freetown during the week of September l9-25, 1970. There were 293 reported cases in 1970 and 211 in 1971, but none since December, 1971. The infection was more persistent in Northern and Southern Provinces and Western Area which together occupy the coast than in landlocked Eastern Province or in the city of Freetown. Cholera has been more serious and persistent in Liberia. Unfortunately, because their early reports to the World Health Organiza- tion were very sporadic, little is known of the diffusion or extent of cholera in Liberia. Monrovia reported the onset of cholera during the first week of October, l970. All counties, except Lofa and Sinoe have reported cholera outbreaks. Data supplied to the World Health Organization indicate a persistent infection averaging about 100 cases per month in Monrovia. There has been no indication of any cholera outside Montserrado County since March, 1972. 6 West Africa, September 12, 1970, p.107o. 79 Ivory Coast After the infection of Monrovia, cholera next appeared l,000 km. to the east in Bingerville, a village located near Abidjan. The index case had apparently contacted the infection from a fisherman at Jacqueville, 40 km. west of Abidjan.7 The apparent lack of infection of the area between Monrovia and Abidjan may be related to differences in coastal morphology, population density, or accessibility. Eastern Liberia and Western Ivory Coasts are not lagoon coasts, so they provide a less favorable environment for vibrio survival. It is also a relatively remote area with poor communications, low p0pulation density, and no major cities. Small, local outbreaks could easily have gone untreated and unreported. An explosive outbreak resulted from the spread of cholera from Bingerville during the week of October 23—29, 1970. Twenty-five deaths and 447 cases were reported in Abidjan in the first week alone. A detailed bacteriological survey of a lagoon village near Abidjan in October showed that 20% of the p0pulation were carriers and that there was massive vibrio contamination of the environment.8 The infection of several villages along Aby and Ebrié lagoons resulted in about l,500 7 H. Felix, "Le develOppement de l'épidémie de cholera en Afrique de l'Ouest", Bull. Soc. Pathol. Exotique, vol.64 (l97lC), p.562. 8 M. Duchassin, et al., "Le vibrion cholerique. Diagnostic bacteriolo- gique", Médecine d'Afrjgue Noire, vol.20, No.3 (l973), pp.165al74. 80 cases and l20 deaths between October and December, 1970.9 Small out- breaks continued in Abidjan and nearby villages during l97l and early l972. The last report was in October, l972. Minor outbreaks of cholera were reported from the inland towns of Bouaké'in October, l970 and Dimbroko, Tiassale'and Abengorou between January and March, l97l. These towns are all located on major tranSport arteries extending north from Abidjan. Abidjan also was the starting point of the most spectacular long distance transfer of cholera, namely the l,500 km. diffusion to Mopti (Mali). Ghana Ghana's first cholera victim was a Togolese citizen arriving 10 by air from Guinea on September 1, l970. The disease did not gain a local base until the beginning of November. The first reported occurrence was at Half-Assini on the Ivoirian border .11 Outbreaks occurred during November in other places along the coast, including the Winneba, Accra, Keta and Ada areas. By mid-January the Upper Region was the only un- infected region. The impact of cholera has been quite serious in Ghana. The reported cases and deaths totalled 2,886 and 73, respectively, in 1970, 12,623 and 609 in l97l and l,l87 and 57 in l972-1973. Well over 9 A.Bourgeade, J. Rive et al., "L'epidémiologie du cholera et ses problemes", Médecine d'Afrigue Noire, vol.20, No.3 (1973), p.l81. 10 J.0. Pobee et al., "Case report of cholera", Ghana Med. J., vol.9, (1970), pp.306-309. ‘1 West Africa, November 7, 1970, p.1326. 81 90% of these cases have been in the Southern regions which are at least partly coastal, namely Western, Central, Eastern Volta and Accra. It has primarily been a rural epidemic affecting mainly the fishing pe0ples. Larger cities with protected water supplies have been repeatedly infected 12 from rural areas but reported few cases. The transportation of bodies and funerals were implicated in most of the new, explosive outbreaks in the early stages of the Ghana epidemic.13 The Upper Region of Ghana has not reported a single cholera case, and the Northern Region reported only nineteen cases. Cholera was also absent from Eastern Guinea, Northern Ivory Coast, Southern Upper Volta and Northern Dahomey. The absence of cases is surprising because of the large amount of travel between these areas and the coast, especially the movement of migrant laborers. Various hypotheses may be advanced to account for the cholera-free status of Upper Ghana and similar areas. The infection may have been introduced by individual carriers but gone unreported because of a scarcity of medical facilities. The dispersed settlement pattern and upland location of most settlement may have provided an unfavorable environment for the establishment and dissemination of cholera. The low population density and small urbanization of the Middle Belt would hamper contact or hierarchical 12 D. Barua, "The global epidemiology of cholera in recent years", Proc. Royal Soc. Med., vol.65, No.5 (May, 1972), p.427. ‘3 "Cholera in 1971", Weekly Epidemiological Record, vol.47 (July 28, 1972), p.282. 82 diffusion. The considerable width of the Middle Belt may also have a barrier effect. The distance factor prompts many travellers from coastal Ghana to interrupt their journey to Northern Ghana in such intermediate cities as Kumasi and Tamale. As a result, most cholera carriers would become vibrio-free or develop clinical symptoms preventing travel before reaching their destination in Upper Ghana. Western and Central Regions are two of the few places in Africa which have continued to report cholera relatively frequently. All but five of the 653 Ghanaian cases in 1973 were from these two regions. With the exception of the Half-Assini area, lagoons are less extensive in Central and Western Regions than in the adjoining areas in Eastern Ivory Coast and Eastern Ghana where cholera does not seem to be endemic. It is perhaps significant that the homeland of the Fanti people, who played such an important role in the initial dissemination of cholera, is in Western and Central Regions. Endemicity in this area therefore may be more a function of cultural than physical environmental factors. Togo and Dahomey Cholera was brought into Togo by Ghanaian fishermen the week of 14 November 13-19, 1970. The epidemic peaked in January, 1971, when 158 cases were reported. Sporadic, minor infections continued during 1971 in the coastal departments. Cholera has affected the fringes of Lome} 15 but areas with piped water have been almost unaffected. There were 14 J.J. D'Alneida et al., "L'épidémie de cholera au Togo", Médecine d'Afrique Noire, vol.20, Nos.8/9 (1973), p.639. 15 D'Almeida, Op.cit., pp.640-64l. 83 relatively small outbreaks in such Central Togo departments as Palimefl Klouto, Akposso, Sokode'and Lama-Kara. Most of these infections originated in the neighboring Volta Region of Ghana.16 Cholera was not reported from June, 1972 until January, 1974, when it reappeared near Lome. The cholera epidemic in Dahomey was also based in coastal lagoons and the delta of the Ouéme'River. The first case was found in the fishing village of Agone’Kame'near the Togolese border on December 7,'1970.17 Cotonou was infected on December 12 and Porto Novo on December 24. The epidemic in Cotonou was double-peaked, the first occurring in January, 1971 and the second during the short rainy season in September and October, 1971. Cotonou had 198 cases and Porto Novo had seventy-one.18 The national total during the serious epidemic phase (December 7, 1970 to April 10, 1971) was 1,812 clinical cases and 261 deaths (14.5% mortality). The only significant inland penetration occurred in November, 1971-January, 1972 when Borgou Division reported a total of 296 cases. Cholera persisted in the coastal lagoons of Dahomey until April, 1972, 19 but the prediction by Aubry of continued endemicity and new epidemics has not materialized. ‘6 D'Almeida, op.cit., p.643. 17 Comité Central de Lutte contre le cholera, "L'Epidémie de cholera au Dahomey", Médecine Tropicale, vol.31, No.6 (1971), p.644. ‘8 P. Aubry et al., "Une experience du cholera Africain...", Bull. Soc. Pathol. Exotique, vol.65, No.3 (1972), p.353. ‘9 Aubry, op.cit., p.356. 84 Nigeria Nigeria was the next country to be infected along the coastal diffusion axis. Cholera first appeared in Ajegunle-Owode village at the mouth of the Ogun River just north of Lagos.20 Five members of one household were hospitalized on December 26 and December 29, 1970. The first case in Metropolitan Lagos also occurred on December 26. The city of Lagos became an endemic cholera focus from which other Nigerian cities were infected. This urban-based diffusion is described later in the chapter. Coastal, lagoon-based diffusion continued beyond Lagos. Western Ijaw Division of Mid-Western State reported the presence of cholera in the last week of January, 1971. Western Ijaw is located in the Western Niger Delta. Cholera was apparently introduced by Ijaw fishermen and local gin dealers travelling by boat from Lagos to the 21 popular markets at Bomadi and Ojobo. The disease spread throughout the rural deltaic areas of Southern Mid-Western State, as well as to the cities of Warri, Sapele and Benin. The urban outbreaks were relatively minor compared to those in rural areas. There were 758 hospitalized 22 cases and 39 deaths up to March 7, 1971. However, a survey of only six 20 B.A.A. Dada, "First sixteen weeks of cholera in Lagos State”, J. Soc. Hlth. Nigeria, vol.6, No.3 (1971), p.133. 2] V.G.B. Amu, "Experience with cholera epidemic in Mid-Western State of Nigeria?, J. Soc. Hlth. Nigeria, vol.6, No.3 (1971), p.107. 22 Amu, op.cit., p.109. 85 villages revealed 135 suspected cases and 75 deaths before the provision of medical aid. There were at least 200 more unreported deaths in February from other riverine villages. The remoteness of the affected areas and the necessity of using boats to transport victims caused delays in treatment and elevated death rates. Cholera also spread into Otikpupa Division in Western State and Brass Division in Rivers State in mid-February. Both areas adjoin Mid-Western State. The size of the Rivers State outbreak is uncertain because of the failure to report cases. With the infection of Port Harcourt, a base was obtained for the diffusion of cholera into East Central State. — South-Eastern State reported cholera for the first time in the week of February 28-March 6, 1971. However, the fact that Southern Cameroun had reported cholera almost four weeks earlier suggests the infection of South-Eastern State may have been earlier than reported. All ’divisions were infected except Ikom and Ogoja in the Northern part of the state. There were only six weekly reports between March and July. However, the average incidence of over 250 cases per week indicates a relatively serious epidemic. Cameroun Southern Cameroun marked the farthest extent of the coastal diffusion axis. The introduction was again attributed to the movement of 86 fishermen along the coast.23 The first case was in a fishing village near Douala. There were 333 cases and 55 deaths during the four week Douala epidemic. The deltaic zone in the Victoria-Douala area became a persistent focus of endemic cholera.24 Minor occurrences continued to be reported periodically up to August, 1973. Several other divisions of Southern Cameroun were infected for short periods of time in 1971 and 1972 from the Victoria-Douala. These included Méfou (Yaoundé), the Southern border division of Kribi which adjoins Equatorial Guinea (Rio Muni), Lom et Djerem on the Central African Republic border , and Bamoun in the Bamenda Highlands. That cholera did not advance into these adjoining countries or into Central Cameroun is probably a function of low p0pulation density, poor transportation and relatively long distances from the endemic focus around Douala and Victoria. Generalizations A number of generalizations follow concerning the spread of cholera along the West African coast. These generalizations have also been incorporated into a systems diagram and a schematic map diagram (Figure 13). 23 J. Dutertre et al., "Le cholera au Cameroun", Médecine Tropicale, vol.32, No.5 (1972), p.608. 24 Dutertre, Op.cit., pp.607—624 includes a detailed account of cholera in Victoria and Douala. 87 FIGURE 13 THE COASTAL DIFFUSION SYSTEM SCHEMATIC MAP OF COASTAL DIFFUSION The model shows the three phases of coastal diffusion. The first phase involves the spread of cholera along the coast to fishing villages in association with the movements of fishennen. Diffusion to surrounding villages from the points of introduction follows. Urban and hierarchical diffusion may follow the infection of major urban centers. 88 FIGURE 13 (A) THE COASTAL DIFFUSION SYSTEM movement of importance barrier effects fishermen _ of transport of time-distance * 1 I 1 corridors ow density areas j coast as a rural ‘sh long distance channel establishment grade and hierarchical of cholera ' diffusion distance' suburban coastal population density establishment‘ urban diffusion lagoon environment of cholera outbreak (B) SCHEMATIC MAP OF COASTAL DIFFUSION .r‘w_:qmwwqfigflhflffgd ° '3': .1“ ’V‘ ,‘ "..A. Lagoone 3A Cities, towns, villages Movement of fishermen along coast (primary phase) - Diffusion from fishing villages (secondary phase) ---- Long distance and hierarchical diffusion (tertiary phase) 89 l. The sequential west to east movement of the infection demons- trates that this was essentially contact diffusion with date of onset being primarily a function of distance from the point of origin. 2. In each of the infected coastal countries except Guinea, fishermen were responsible for the introduction of theinfection and much of the subsequent internal diffusion. The Fanti were primarily responsible but other fishing people such as the Ijaw, Ewe, Kru and Ebrie'were also involved. 3. The coast served as a diffusion channel for cholera. This may be attributed to the lack of east-west roads near much of the coast, and the use of canoes for both short and long distance travel. 4. The fishermen first introduced the disease in rural or sub- urban areas. This may be termed the primary phase of the coastal diffusion (Figure 13). The areas of introduction were physiographically distinctive, being either deltas or lagoon coasts. Cholera has also persisted longer in these deltaic and lagoon areas. 5. When villages near a major urban center were infected, the disease spread to the urban center (secondary diffusion phase) and subsequently diffused along tranSport axes as hierarchical or long distance diffusion (tertiary stage). 6. Distance from coastal endemic foci and low population density regions such as the Middle Belt and Southern Cameroun seem to have acted as a barrier to the disease. Despite considerable movement by migrants 90 and traders across the Middle Belt, it was breached in only two places, namely, Nigeria and Mali. Denser population and better transport reduce the barrier effect of the Middle Belt in Nigeria. As for the long distance transfer of cholera from Abidjan to Mopti, it is probably not unreasonable to consider it a long-shot possibility which happened to materialize. Riverine Diffusion Primary_(River Valley) Phase The spread of cholera to M0pti (Mali) from Abidjan, a distance of 1,500 km. has been noted in the previous section. Mopti is a very important center for the marketing of dried and smoked fish from the Niger River's Inland Delta. The index case was a trader who had arrived on November 5, 1970 to buy fish in the market.25 He developed cholera symptoms on November 6 and spent two days frequenting the public latrines of the market.26 An explosive outbreak occurred two weeks later at the time of the weekly market in M0pti. The large attendance at the market, much coming from a considerable distance explains the rapid diffusion of the infection from M0pti. The Mopti epidemic corresponded to the classical epidemic model of four stages, namely, introduction of the contamination, latent phase with multiplication of the bacteria, explosive epidemic and finally 27 dissemination outside with the departure of people. 25 Félix, op.cit., 1971c, p.563. 26 Félix, op.cit., 1971C, p.563. 27 Félix, op.cit., 1971C, pp.563-564. 91 The diffusion of cholera from Mopti initially followed the axis of the Niger River. The week after the Mopti outbreak, five divi- sions in the Inland Delta, plus the important river town of Segou, located upstream from Mopti, reported cholera. The other two divisions partially located in the Inland Delta (Djenné and Douzentza) were infected in the third week. Tombouctou, Gao and Koulikoro also reported cholera in the third week of the riverine epidemic. Koulikoro and Gao are respectively the western and eastern terminuses of the Niger River navigation system, and Tombouctou is one of the main intermediate ports. The river is the main transportation channel in Eastern and Central Mali, and for most of the riverine areas the only access route from about July to November. The division containing the Malian capital, Bamako, was infected in the fifth week. The next significant advance was on December 27, 1970 when the border of theNiger Republic was breached. The first outbreak in Niger was at Famalé’in Tillabery Division. It subsequently spread within four weeks to Tera, Niamey and Gaya Divisions along the Niger and to three peripheral divisions. The only other occurrence of cholera in the first ten weeks ofthe riverine epidemic were in Ouagouiya and Tougan in Northern Upper Volta. Figure 14 effectively shows the connection between location along the Niger River and the early stages of cholera in Mali, Niger and Upper Volta. 0f the twenty-four divisions infected during the first ten 92 FIGURE 14 FIRST TEN WEEKS OF THE RIVERINE CHOLERA EPIDEMIC All but five of the first twenty-four divisions infected during the riverine cholera epidemic were located along the Niger River. This map emphasizes the high correlation between riverine environment and early cholera onset. 93 ON m. m. m FI 259% $5010 szmzm m5 9mm; 2m: 5m: mo S umawd 94 weeks, nineteen were located along a 2,000 km. expanse of the Niger River between the Nigerian border in Gaya Division and the Guinean border in Bamako Division. Various factors account for the channelling effect of the River Valley.. Where navigable the rivers themselves provide a transporta- tion route. When roads are constructed, river valleys often offer the path of least resistance. In Mali, the p0pulation density is higher in the valley than in the neighboring upland areas. Also, river valleys provide a suitable hydrological environment for the survival of vibrig_ cholerae. It is these characteristics of river valleys which facilitate cholera dissemination. The introduction of cholera in the riverine divisions of Mali was typically followed by an epidemic lasting about three to seven weeks. It is very difficult to estimate the extent of these outbreaks. Because of the inaccessibility of many of the infected areas and the total lack of prior preparation by health authorities, mortality was very heavy and the majority of cases went unrecorded. For example, a total of 917 cases and 33 deaths were listed in the Weekly Epidemiological Record from Goundam and Dire DiviSions in 1970. However, Félix reports that there were over 5,000 cases and 2,000 deaths among sedentary farmers alone, plus heavy mortality which is impossible to estimate among the nomads and fishermen.28 28 H. Felix, "Epidemie de choléra en Afrique. Note d'information sur l'evolution entre Aofit et Decembre, 1970", La Presse Médicale, vol.79, No.11 (1971), pp.476. 95 Secondary (Valley Periphery) Phase The second phase of the riverine epidemic in Mali comprises the period between the tenth week of the Mopti epidemic (week 25 out of Conakry) and the forty-seventh week. Cholera remained endemic in the Inland Delta, particularly around Mopti. There were periodic new outbreaks in the important river towns of Segou and Koulikoro. It was from this focus of infection that cholera gradually penetrated into the valley of the Bani River and the upland periphery of the Niger Valley via Mopti, Ségou and Koulikoro. There were four principal thrusts into the peripheral divisions. One was south from Mopti into the vicinity of the Bandiagara Escarpment and the northern border divisions of Upper Volta. A second axis of diffusion was the movement from Segou into three divisions in the Valley of the Bani River, a major tributary of the Niger. The three divisions were San, Koutiala and Dioila. There was also a major northward thrust from Koulikoro into Banamba and Nara in arid North-western Mali. Nara had a reported 1,129 cases and 332 deaths in thirteen weeks. Cholera reached South-Eastern Mauritania from Nara. Amourj, Néma and Timbédra were the infected Mauritanian divisions. The fourth diffusion axis, also originating in Koulikoro, extended along the Dakar-Mali railway into the headwaters of the Senegal River. Bafoulabe’and Kayes Divisions were infected in Mali. The infection subsequently progressed down the valley of the Senegal River, reaching the coast at the beginning of September, 1971. 96 FIGURE 15 CHOLERA DIFFUSION IN THE MIDDLE NIGER WATERSHED Cholera was introduced from Abidjan into the Niger Valley at Mopti in mid-November, 1970. During the first phase of the epidemic, cholera rapidly spread down the valley of the Niger River to Nigeria and upstream as far as Bamako. The secondary phase involved a more gradual penetration of the upland areas flanking the Niger River, the Bani River Valley and the valley of the Upper Senegal. Cholera has continued to Spread in Upper Volta in connection with the West African drought. This has been called the tertiary diffusion phase. 97 N x6593 oom . 030 33a 3:06.20 . ._ lacunae. com 393 8238-20323 32:07:... 393 3.93:3 3:03 Coucooomlll 393 23:; 392. CoEcml moSo. 573:5 o.o_czo 5:03.92: 9205 .o xoo>>® Ezoxqo \\ l.\ 3200 of .5“. e / \ll-.. c9220 «mam.» 6:80 _oco:ozono_2 .263 283 82m Concaon :EuESLEfluquduu..- /II.\ co cozouEBc .2050 02 m 393 396 3:. 229.0 000% \Is édzhxwo It II E 1. 3.2.65 anon” J Hmdoo >mO>_ A J \r . oo~ numfiImmllb \ .l/ \_ . . x J I) \ fi.\. (\IJ dmOuZan Me KL — C. \. Du] “no \ 8.937\ O Douala \ ’ \ (UZBO iv; %\ .Jnnxld 1.... .. x _ n '2 25232 SE waywnfl w... 8 \im ~no m30 o:2mo_n_w 8. 00h «ha. .n mmoxgmwm n 05. .N. KUOIuONO hm mmaoi uhdhm mos; ¢Ou u>¢30 Elmo-nu (CNJOIU Amy 151 major peak in the North-Western Periphery was on January 20. It was also preceded by two smaller seeding period peaks. Mainland Lagos had two early peaks on January 16 and February 27. They occurred at the same time as peaks in the North-Western Periphery and slightly after peaks on Lagos Island. The first peak in the Eastern Periphery, which is adjacent to only Lagos Island, followed the first major Lagos Island peak. Thus, during January, the earliest peaking of the epidemic was in Lagos Island, followed by peaks in other sections of the Lagos Metropolitan Area. These later peaks are thought to be the result of the transfer of infection from an earlier-infected source, namely, Lagos Island. At the end of January, there was very low incidence of cholera in all parts of Lagos. However, there was an outbreak at this time just north of the city in the vicinity of the initial cholera cases in Lagos State. The infection may have been introduced to Lagos for a second time, as this second outbreak on the outskirts of the Metropolitan Area was followed by a new rise in cholera incidence within Metropolitan Lagos. During this Second epidemic, the sequence of peaking was reversed. It was the North-Western Periphery and Mainland Lagos which peaked first (February 1) followed by Lagos Island on February 5 and February 11. Minor peaks in the Eastern Periphery followed on February 7 and February 11. 152 The next major peak was in Mainland Lagos on February 17, followed by simultaneous peaks on February 20 in Lagos Island and the North-Western Periphery. February 15-17 therefore appears to be the beginning of a third phase in which Mainland Lagos had become the center of diffusion from which cholera spread to Lagos Island and the North-Western Periphery. Mainland Lagos had four major peaks between February 16 and March 7. Small peaks in Lagos Island and the North- Western Periphery on March 5 followed an epidemic peak in the Lagos Mainland. After about a week of low activity there was a major peak in the North-Western Periphery on March 18. Data inavailability prevented consideration of its impact on adjoining regions. The sequence of peaks suggests that first Lagos Island. followed by the North-Western Periphery, and then Mainland Lagos acted as source areas for new epidemics in the other regions. It is hypothesized that vibrio contamination was relatively heavy throughout Metropolitan Lagos, and that the introduction of new infection from an adjoining region would be comparable to adding water to a full bucket, i.e. it would trigger an epidemic. The multiple peaks are therefore thought to represent essentially a chain reaction of repeated inter-regional diffusion within the city. It is perhaps significant that Mainland Lagos which is located centrally between Lagos Island and the North-Western Periphery had eight relatively clear peaks between January 6 and March 6, compared to only six in Lagos Island and seven in the North-Western Periphery. 153 The epidemic curves seemed to indicate the existence of patterns of spread between regions of the city. A single region was selected and the early cases of cholera were mapped to find if similar patterns of diffusion could be found within a city area. Lagos Island was selected because of its physical separation, except for bridges, from the other regions. The early occurrence of cholera on Lagos Island, between December 29, 1970 and February 14, 1971, is shown in Figure 28. The sequence of 1] The cases has been divided into cohorts of approximately ten each. homes of the first ten cohorts have been mapped. Cholera was initially concentrated in the center of Lagos (period one, Figure 28A). The first case in Lagos Island was within a block of Tinubu Square which is the focal point of the city. During the ensuing second, third and fourth periods cholera moved outward in all directions toward the edge of the Island. After this explosion from the center of the island outward the epidemic wave seemed to implode in periods five and six. The new cases were found ever closer to the center of the Island again. The pattern is less obvious during periods seven through ten. In periods seven and nine, the infection seemed to be relatively scattered toward the edge of the Island, while there appeared to be greater clustering in the center of the Island during periods eight and ten. The areas which had the heaviest incidence of cholera are in the center and north- western part of Lagos Island. These are high density residential areas. 1 . . . 1 Because all cases reported 1n one day were cons1dered as a s1ngle unit, some groups contain slightly more or less than ten cases. 154 FIGURE 28 EXPLOSIVE AND IMPLOSIVE CHOLERA DIFFUSION DURING THE FIRST SEVEN WEEKS 0F CHOLERA IN LAGOS ISLAND The locations of ten successive cohorts of ten cases each have been mapped. Beginning in the center of Lagos Island, cholera had spread outward to the edge of the island by periods three and four. The epidemic wave then reversed itself and imploded toward the center of the island (period six). The graph shows the dispersion and mean distance of cases from Tinubu Square which is the focal point of the Lagos Central Business District. The graph clearly shows the explosive and implosive diffusion of cholera in periods one to six and the less spectacular fluctuations characterizing periods seven to ten. 155 533:. A FIGURE 28 .50.. DEC. 29.870-JAN. 22.1971 A , EXPLOSIVE AND IMPLOSIVE A o L La can 0 m' ° CHOLERA DIFFUSION C] e 4 D . .0 D. . DURING THE FIRST SEVEN WEEKS 0 we.” a o ° ° OF CHOLERA ON LAGOS ISLAND ® \ D o o \‘ DUB \ A \\D #7 Lagos Harbour ‘J v ® TInubu Square \ Central Business DIstrICt Cl 0 (band an uaboounje, 1968. p.281) \"\ o COHORTS 0F :10 CASES Ikoyi D Period l-‘Dec.29- Jam? Note= Meet locatione are approximate. Locations are A pui“ 2. Jana-1| baeed on Road Mape at 'Greater Logo—3‘ , l973 0 Period 3-J0h12‘18 Twelve caeee were not located. 0 Period Oi Jan. 19-22 Victoria leland . B D - JAN- 23‘ FEB- 5-19'" DISTANCE or CASES FROM CENTER . ° 0F LAeos ISLAND (TINUBU SQUARE) 0.: / o D A A I {— Mean distance from Tinubu Square A I,” U A N ' D A A0 ° Kb D a I" C V r I or . 2 0 0 e \‘\ I6» COMORTS 0F :10 CASES U Piriod 5: Jan.23-25 [4» e A Period 6: Jan. 26-Feb.2 . 0 Period T= Feb. 3-5 /—N |.__A_A—_——~—A——’ 0 MILES 0.5 l.2’ . _- ..- ._. . . . 0 FEB. 6-14. I97| m ° . . 3 10* ' . a L up . l. A a: 0.6 \ / l O AD \, I \/A \ O D 0 e \\\,_ U o 08 06' O ‘K D o 0 A A ® 3 A o . \ A A . \f7 04» . . 7 t : : ' ° \\ O O O . . . . \ 02 o . COHORTS or 210 CASES / . . 0 Period e: Feb.6-8 ° 0 a nice 9: Foes-Io o ‘ g , . . ._.. ,_,_L__.__ 9 ..0 : . - 3 4 5 6 7 8 I 0 Periodic Feb ll 14 I 2 PERIOD NUMBER (SEE MAPS) 156 The southern and south-eastern portions which have the fewest cases comprise the Central Business District and governmental areas. The dispersion of cases from the focal point of Lagos Island, Tinubu Square, was measured for each cohort. Figure 280 shows the dispersion of cases and the mean distances of cases from Tinubu Square in each period. This graph confirms that the infection exploded away from the center of the island and after the fourth period began to implode again toward the center. A The explosion-implosion sequence described above is analogous to the behavior of a physical wave which is reflected by an impermeable barrier. The analogy is attractive because the water surrounding Lagos Island is a definite barrier to the Spread of the infection. However, the use of a physical wave analogy does not explain why an epidemic wave should appear to behave in such a way. It does not explain why the out- ward-spreading infection should not ”pile up" along the waterfront and cause a large epidemic there rather than receding from the edges of the Island toward the center. This distinctive explosive-implosive pattern of cholera dissemination on Lagos Island therefore remains unexplained. Neither the regional epidemic curves (Figure 27A) or maps provide evidence about the method of cholera dissemination. The long persistence of the epidemic suggests the existence of a well-established man-environment transmission cycle, as opposed to person to person Spread. The large water frontage and grossly inadequate sanitation would facilitate the establishment of such a cycle. The wide dispersion of cases suggests 157 that carriers were instrumental in disseminating the infection from endemic foci throughout the city. If carriers had not played an important role in diffusion, cases would have tended to be more highly concentrated around endemic foci. The numerous food sellers, whose products are prepared and sold in very unhygienic conditions may also have played an important role in the dissemination of cholera in Lagos. Age distribution of cholera victims The age distribution of cholera victims in Lagos State in the first sixteen weeks of the epidemic corresponded to the age patterns found elsewhere in Africa and to the pattern expected for a newly-infected area. The highest incidence was among young adults, with 47.8% of the first 922 cases being among persons aged twenty-one to forty. Normally, adult males have the highest attack rates due to their greater mobility and exposure to the risk of infection.]2 However, adult female cases out-numbered males in Lagos. This may be related to the female domina- tion of market trading among the Yoruba of Lagos. The incidence of cholera among children in newly-infected areas is typically low. In Lagos, the under ten years cohort had only 9.2% of the cases, although they probably account for one quarter to one-third of the total population. 12 W.H. MoSley, "Epidemiology of cholera”, in Principles and Practice of Cholera Control. Geneva, WHO, 1970, p.25. 158 TABLE 6 Age—sex distribution of the first 922 cholera cases in Lagos State Age cohort Male Female Total Percentage of cases 0-10 56 29 85 9.2 11-20 75 49 124 13.4 21-30 115 139 254 27.3 31-40 87 103 190 20.5 41-50 69 57 126 13.6 51-60 31 37 68 7.4 61-70 17 22 39 4.3 71 & over 6 12 18 2.1 Unknown 10 8 18 2.1 Total 466 456 922 99.9 Source: Dada, op.cit., 19718, p.138. Ibadan, Nigeria Ibadan experienced an explosive epidemic following the introduc- tion of cholera from Lagos on January 3, 1971. The epidemic peaked between its Sixth and ninth week. There were over 3,300 cases in this four week period.13 Treatment facilities were completely overwhelmed. 13 J.D. Adeniyi,"Cholera control: problems of beliefs and attitudes“, Int. J. Hlth. Education, vol.15, No.4 (1972), p.238. 159 St. Mary's Hospital, with a total of only eighty beds treated 4,303 cholera victims in eight weeks and admitted up to 256 new patients in a single day.14 The infection persisted at a lower level for over a year. Infected wells were implicated in the Ibadan Cholera epidemic.15 A study by Adeniyi16 showed that poor sanitation and the beliefs and attitudes of the population contributed to the massiveness of the epidemic. An estimated 10% of the population defecates in the streets, streams, and drains, while many pe0ple continue to rely on these streams for their water supply. Adeniyi's survey of 250 randomly selected people revealled widespread misinformation about the epidemic. For example, 70% thought that cholera was indigenous to Nigeria, confusing it with a disease known in Yoruba as origbameji (literally two calabashes, i.e. one for vomiting and one for diarrhoea). Because they had successfully treated origbamejj_at home, people were inclined to use the same tradi- tional treatment for cholera and go to the hospital only when it was too late. Over 18% of Adeniyi's sample believed that cholera was caused by angry Gods or the Yoruba masquerade glglg, Native medicine was preferred to Western medicine by 14.8%, and 15.2% said they did not believe in vaccination. Many of those preferring native medicine and rejecting vaccination were among the group who attributed cholera to angry Gods or Olulo. 14 E.A. Lewis et al., "Cholera in Ibadan", Amer. J. Trop. Med. Hygiene, vol.21 (1972), p.309. 15 U.L.C.M., op.cit., p.144. 15 Adeniyi, op.cit., pp.238-245. 160 A significant implication of the above findings is the diffi- culty of successfully implementing control measures when a segment of the population is unwilling to participate or cooperate because of their beliefs. The uncooperative group may easily become a persistent reservoir of infectivity. Zaria, Nigeria The distribution of cholera in the City of Zaria in 1971 has been studied and mapped by Schram.17 Most cases were in the Old City from the area surrounding the market. Three-quarters of the victims obtained their water supply wholly or partially from untreated wells or the river. The occupational division of a sample of 200 cholera patients included twenty-one Koranic pupils, twenty-one traders, fifteen tailors, nine foodsellers and butchers, three Koranic school teachers, one water seller and one beggar, but only three students and nineteen farmers and herdsmen.18 The concentration of market-associated occupations, as well as the spatial location of cases implicates the market as a source of cholera dissemination. The relatively large numbers of Koranic pupils and foodsellers infected suggests that food may have been an important vehicle of transmission. Koranic students in Northern Nigeria who sustain themselves by begging habitually congregate around the market to get left-over food from foodsellers. ‘7 R. Schram, "The 1971 cholera epidemic in Zaria, Nigeria", Savanna, vol.1,No.2 (1972), pp.213-222. 18 Schram, op.cit., p.221. 161 Goudoumefl Ivory Coast Goudoume is a lagoon village located a few miles east of Abidjan. It was the site of a bacteriological search for carriers in October, 1970. Of the 506 cultures obtained, 101 were positive, despite the fact that there were only four clinical cases during the entire outbreak.19 Virtually every resident of Goudoumé must have been a carrier at some point. Examination of wells and the lagoon edge in infected lagoon villages usually showed the presence of vibrio contamination.20 The Goudoumé example is probably typical of cholera outbreaks along the West African coast. This was an example of the classical cholera transmission model, with massive contamination of the environment, particularly water supplies. The epidemics in such places tended to be protratted but unspectacular. Among the factors which account for the high risk of establishing a carrier-environment transmission cycle are the location of villages along lagoons or delta channels, the predominance of fishing as an occupation, the use of untreated water from lagoons, rivers and shallow wells, and the suitability of the brackish water found along the coast for vibrio survival. 19 J. Le Viguelloux and J.C. Doury, ”Epidemiologie du cholera moderne”, Médecine Tropicale, vol.31, No.1 (1971), p.23. :20 A. Bourgeade, J. Rive et al., "L'epidemiologie du choléra et ses probiemes", Médecine d'Afrique Noire, vol.20, No.3 (1973), pp.177-187. 162 Lougg, Senegal The holding of gatherings such as markets, funerals, weddings and festivals has been noted in previous chapters as a key element in the inter-regional and long distance diffusion of cholera. Gatherings were equally important for local dissemination of the infection. For example, cholera was introduced to Thiaméne, near Louga, Senegal by 2] Cholera traders from the Senegal Valley coming to the periodic market. Spread radially from Thiaméne to seventeen villages up to fifteen kilometers from Thiaméne within five days. Most of the villages were infected by persons who had attended the market, although the actual means of infection, whether person to person or through contact with an infected medium, is unknown. In the Podor-Matam region of the Senegal Valley local diffusion followed a wedding.22 A second phase of the diffusion occurred as a result of the gathering of mourners for the funerals of first generation victims. Akodesséwa-Plage, Togo Following the introduction of cholera in late November, 1970 by Ghanaian fishermens to Akodessewa-Plage, a fishing village twelve 2] A. Sy et al., "L'epidémie de cholera dans le departement de Louga (region de Diourbel)", Bull. Soc. Med. Afrigue Noire Langue Frangaise, vol.17, No.4 (1972), p.663. A. Carvhallo et al., "L'épidéMie de cholera a Podor et Matam", Bull. Soc. 22 Med. Afrique Noire Langue Francaise, vol.17, No.4 (1972), pp.655-66Tl 163 kilometers from Lome, stringent measures were taken to control the epidemic.23 Quarantines, mass vaccination, establishment of treatment centers, prohibition of fishing and batheing in the lagoon or sea, and disinfection of vehicles used to transport cholera victims are some of the measures which were used in an attempt to stop the epidemic. The containment efforts failed. DeSpite the deployment of troops to enforce the quarantine, the nightly movement of fishermen into the restricted zones from infected areas in Ghana continued.24 They also failed to prevent the escape of villagers through the cordon. Fugatives from AkodesséWa-Plage carried cholera to Zolwa which served as a point of diffusion to several other villages. The intense social interaction of the pe0p1e and their failure to heed warnings against travel, gatherings and ritual treatment of bodies before burial were responsible for the survival and diffusion of cholera in coastal Togo.25 The establishment of cholera in Coastal Togo shows the diffi- culty of implementing control measures to contain thediffusion of the disease. The success of control measures may be affected by the attitudes of the people toward the disease and government-initiated controls. The controls are likely to fail if the understanding and cooperation of the p0pulation is obtained, and unless the controls are designed to only minimally disrupt normal activities. 23 The control measures are described by E. Amégée et al., "MéSures d'hygiéne du milieu au cours de l'epidemie de choléra au Togo", Medecine d'Afrique Noire, vol.20, Nos.8/9 (1973), pp.649-654. 24 J.J. D'Almeida et al., "L'epidemie de cholera au Togo", Médecine d'Afrique Noire, vol.20, Nos.8/9 (1973), p.642. 25 D'Almeida, op.cit., p.643. 164 Goulfey, Northern Cameroun The cholera epidemic at Goulfey is an example of apparent person to person spread. The outbreak began the day after a large circumcision ceremony attended by 15,000 to 20,000 pe0p1e from Cameroun, Chad and Nigeria.26 Between May 7 and May 28, 1971 there were 801 cases and 121 deaths among the 3,500 residents of Goulfey. All wells in the town were chlorinated and all latrines disinfected four or five times daily. Among the 700 boys who had been circumcized and whose customary post-circumcision isolation coincided with the epidemic, there was not a single case.27 The absence of infection among the isolated children and the treatment of water sources would seem to rule out spread by contaminated drinking water or food. Person to person spread seems to be indicated. It probably resulted from burials and caring for cholera patients at home. The attack rate in Goulfey was higher among women than among men.28 This is somewhat surprising because of the usual confinement of Muslin women and the more active role of men in the conducting of burial ceremonies. However, women might play a more active role than men in the care of the sick at home. This may account for the higher incidence among women. 26 B. Coulanges and P. Coulanges, "A propos de l'epidémie de cholera du Sultanat de Goulfey (Nord Cameroun: Mai-Juin, 1971). Considerations sur l'épidemiologie et la proplylaxe", Bull. Soc. Pathol. Exotique, vol.65, No.2 (1972), p.216. 27 Coulanges and Coulanges, op.cit., pp.222—223. 28 Coulanges and Coulanges, op.cit., p.221. 165 Another puzzling aspect of the Goulfey outbreak is the very small number of cases among residents of nearby villages and among guests at the circumcision ceremony. While the infection was imported at the time of the festival, it was only after the festival that widespread dissemination occurred. Generalizations The local dissemination of cholera has been shown to be very much a function of local conditions. The following generalizations summarize aspects of the local dissemination of cholera in the seven locations considered in the chapter. 1) Distinct patterns of cholera diffusion may be found in local areas, as well as in inter-regional diffusion. However, the complexity of urban environments and the considerable potential for Spread through person to person contact often complicates the analysis of these local diffusion patterns. 2) Markets, funerals and festivals were important factors in local, as well as inter-regional diffusion. 3) The means of dissemination varied from place to place. Infected water supplies were implicated in Goudoume and Ibadan. Person to person Spread apparently occurred in Goulfey. Contaminated food sold by market vendors is the suspected mode of transmission in Zaria. 4) Inadequate sanitation and absence of protectéd water supplies generally were characteristic of the infected areas. 166 5) Control measures must be understood and approved by the controlled population if they are to be successful. Control measures should disrupt normal activities as little as possible. Successful implementation of control programs is especially difficult where a part of the population is uncooperative because of their belief systems. 6) The highest incidence of cholera was among young adults. Children had a relatively low attack rate. 167 CHAPTER 7 SYSTEMS MODELS OF CHOLERA DIFFUSION The previous chapters dealing with the diffusion of cholera at different scales of analysis from international to intra-urban contain considerable evidence of the complexity of the cholera trans- mission cycle and diffusion process. Cholera vibrios are adaptable to considerably different environmental conditions and may be Spread in a variety of ways. Because of this chameleon-like adaptability, it is generally difficult to discover the basis of survival and transmission during a particular outbreak. A considerable number of detailed studies have been undertaken in recent years in Asian countries, especially India and Bangladesh, which have attempted to gain a better understanding of cholera epidemiology. The investigations seem to have achieved very few significant break—throughs. As even intensive local studies have generally failed to produce satisfactory explanation of cholera survival and diffusion processes in a particular epidemic or under particular environmental conditions, any attempted general model of the interrelationships of factors governing cholera survival and diffusion is bound to be in- adequate, especially when it is based only on secondary sources. Nevertheless, this chapter will focus on the synthesis of factors relevant to the survival and diffusion of cholera El Tor, with particular reference to the experience with cholera in West Africa. 168 FIGURE 29 LOCAL AND EXTERNAL CHOLERA DIFFUSION A model of cholera diffusion must include two essential components, namely, local and external diffusion. Figures 30 and 32 show the components of the local and external diffusion subsystems. 169 . 2053.935 34205 .20 -mw...z= Adzmmkxw ZO_mDmn=o ...z m22_<._.zoo _>m:m 1.400... 0.26 co_oa_Encot. /\ .024 u ..... .......... ............... . . . x. o _ ....................._..................................... .. . . o . . W. 3:56 L8: 388: $5.25 .. x. . cowpumwcw we mucmpmwo . AV . . . . xx mxmws cmzmwz \oewp mmcmpogu owcn_> . . . . . x . o . b . o if 2038 ........................ ..... ...... / II ..................... ....... Amcwcczpmc xx hmww :owuwcup< :,+a ........ H e / 1.1 .6 mctmpcmwowmflmfi A x/ :owpmwwwow ucfimmmmmicofiomwfi .N N 2395 ~98.an so cowuecpcmucou carom—ago; .0 xx mowpmwwnma Mmcmwuwouwwmw .m WW wIPm_uom..mmmm:wmt mNPm cowumpzaoa .m xx . mcoppupcumwc _m>mch .¢ mm .mumEo: .mucmcmwsv 3:333 .mmctcmv KllllIIIILILIIllllIIII N 223%...» m I . . .fififififi .. .. a move: “Loamcmch .m prmmmu cowum_:aoa.moq .N “W cowumuzvm .cupmm; .pcmE:cw>ow .N Ame ”MMmemmcflmfipmxmv . mm pcmscwmpcou Amumxcmsv ounce ._ . . a _ wm \. coruuocuu< mumpa chpcmu IEOOZ zo_m3n_u=o A_o-mm._.z_v 1.4.meme Om mmDoE 174 The length of carrier status is short, generally one or two days. Thus, the attrition of the vibrio is a major limiting faCtor in the Spread of cholera, especially with long distance diffusion. Key elements affecting the attrition of vibrios in the course of a journey are time and distance. The longer the journey the greater is the probability of a carrier arri- ving vibrio-free at the destination. Vibrio characteristics, especially their rate of mortality in the carrier state also govern vibrio attrition. Barrier, channel and central place effects combine to focus the probability of disease introduction on certain locations, while reducing the probability of infection in others. Barriers are features which impede or prevent movement. Examples of barriers are physical barriers such as mountain ranges or gorges, immune populations (vaccination barrier), travel restrictions created by travel cordons or international boundaries, and scarcity of transportation routes or transportation vehicles. Distance and inter- vening opportunities may also act as barriers to the flow of traffic between two locations. The effect of impermeable barriers is to prevent movement and as a result prevent the diffusion of cholera beyond the barrier. Permeable barriers impede and reduce the volume of traffic. Diffusion through a permeable barrier is possible, but the probability is relatively low. Channels are features which provide paths of least resistance facilitating movement between points along and near the channel. Movement may be facilitated by a reduction in the time, cost or difficulty of the 175 journey. Channels may be physical features, such as valleys. Trans- portation routes are man—made channels of particular importance for diffusion. The Spatial distribution of the disease-susceptible popula- tion also affects the Spread of the disease. Places located along physical or man-made channels have a relatively high risk of infection because of the possibility of contamination by persons passing through toward another center further along the channel. Population nucleations have unequal attraction for travellers. The more important a place, the more traffic it is likely to attract. Among the functions which combine to constitute the central place attraction of cities for travellers are the volume of trade, markets, location of government, health and educational facilities, and traditional functions such as the staging of traditional festivals. Important transportation nodes, such as junction points, railway centers or airports attract transit travellers. Population size and employment opportunities for migrants are other aspects of central place attraction. Barrier, channel and central place factors are Closely inter- related. For example, the Sahara Desert and the Middle Belt in West Africa constituted barriers to cholera diffusion. However, the barrier effect is primarily a result of the lack 0f central places With substantial attraction in the Sahara Desert and Middle Belt, and the relative scarcity of transport channels. Three types of barriers, namely, physical barriers, vaccination barriers and travel restrictions affected the diffusion of cholera in 176 FIGURE 31 MARKET SCENE IN MOKHA (NIGERIA) The role of markets as foci of cholera diffusion in West Africa is well established. For example, markets were implicated in intra-urban diffusion in Zaria (Nigeria), local rural diffusion at Thioméne (Senegal) and both long distance and regional diffusion at Mapti (Mali). The meeting of cholera carriers and cholera susceptibles from a wide area in the crowded and often unhygienic conditions of the market facilities cholera dissemination. 177 West Africa. The Atlantic Ocean provided an impermeable physical barrier along the West African Coast, and the Sahara Desert a virtually imper- meable barrier to the North. Areas of low population density in the Middle Belt, Southern Cameroun and Central Chad also provided barriers to the infection. Vaccination barriers were used in the vicinity of Bamako2 and Fort Lamy3 to protect the urban areas from cholera. Although both cities experienced outbreaks, it appears that the vaccination program was instrumental in reducing the size of the outbreaks and preventing these cities from becoming major diffusion foci. Travel cordons were 4 5 and Northern Cameroun6 in trying to contain the used in Togo, Mali infection. These cordons formed permeable barriers which reduced travel but did not prevent further dissemination of the disease. International boundaries were ineffective as cholera barriers in West Africa. Even if legal crossings are controlled, the uncontrollable movements of smugglers and migrating herdsmen and fishermen continue. Smugglers have been implicated in the introduction of cholera in Makary (Northern 2 J. Voelckel and G. Causse, "Apercus prophylactiques", Médecine Tropicale, vol.31, No.6 (1971), p.713. 3 O. Bono et al., "Installation du cholera aux alentours du Lac Tchad", Bull. Soc. Pathol. Exotique, vol.64, No.4 (1971A), p.389. 4 J.J. D'Almeida et al., "L'épidénie de choléra au Togo", Medecine d'Afrique Noire, vol.20, Nos.8-9 (1973), p.640. 5 G. Causse, J. Le Viguelloux and J. Voelckel, “Remarques sur 1'orgini- zation practique de la lutte contre le choléra en zone rurale dépouvre de médecin", Bull. Soc. Pathol. Exotique, vol.64, No.5 (1971), p.662. 6 Causse, Le Viguelloux and Voelckel, Op.cit., p.662. 178 Cameroun)7 and Tunisia.8 Nomads were responsible for Spreading cholera in East Africa9 and Mauritania.10 The role of migratory fishermen in the coastal diffusion phase has been well documented. Physical and man-made channels were influential in the diffu- sion of cholera in West Africa. The valleys of the middle Niger and Senegal Rivers were the most prominent physical channels. The coastal fringe, which includes the lagoon networks and inshore areas, also served as a channel for the movement of fishermen in canoes. In addition to providing physical channels facilitating movement, the river valleys and coastal areas tend to have relatively high population densities. This linear concentration of susceptibles also promotes a linear pattern of diffusion. Man-made transportation channels directed the patterns of diffusion in Angola and Mcmambique, where diffusion axes tended to follow railway lines, and in Nigeria, where the amount of transportation develop- ment was found to be Significantly correlated with the spread of cholera. The regularity with which large cities were infected with cholera in various parts of Africa attests to the importance of urban centrality. The impact of central place attraction was most prominent in Nigeria, where a composite index of urban importance was the most 7 B. Coulanges and P. Coulanges, ”Lépidémie de cholera du sultanat de Goulfey (Nord Cameroun: Mai-Juin, l97l)”, Medecine Trgpicale, vol.31, No.6 (1971), p.637. 8 H. Félix, "Le choléra Africain”, Medecine Tropicale, vol.3l, No.6 (19710), p.620. 9 B. Carteron and J.C. Artus, "Epidémiologie du cholera au Territoire Francais des Afars et des Issas", Medecine Tropicale, vol.33, No.3 (1973), pp.235-248. IO "Cholera and politics", West Africa, No.2881 (Sept.8, l972), p.1175. 179 important of six variables correlated with the time of cholera onset. Urban importance was not uniformly important in controlling diffusion patterns. The correlation between urban importance and week of onset in Mali was not statistically significant. In Kenya, cholera penetrated relatively close to such cities as Nairobi and Kisumu, but remained confined to rural areas. Particular central place functions were instrumental in the diffusion of cholera in various areas. Markets were diffusion foci in several areas, including Mopti, MaliH and Thioméne, Senegal.12 The attraction of thousands of people to a traditional ceremony in the emirate capital of Goulfey13 brought about the initial cholera outbreak in the vicinity of Lake Chad. The movement of seasonal labor migrants may also be considered an example of central place attraction because migrants' places of origin and destination are Spatially concentrated. Seasonal labor movements 14 have been linked to cholera diffusion in Bangladesh. While there is no conclusive evidence that labor migrants spread cholera in West Africa, 1] H. Félix, "Le développement de l'épidemie de choléra en Afrique de l'Ouest", Bull. Soc. Pathol. Exotiqge, vol.64, No.4(l97lC), p.566. 12 A. Sy et al., ”L'epidemie de cholera dans le département de Louga (region de Diourbel)”, Bull. Soc. Méd. Afrigue Noire Langue Frangaise, vol.17, No.4 (1972), p.663. ' 13 Coulanges and Coulanges, op.cit., p.636. 14 W.M. McCormack, W.H. Mosley et al., "Endemic cholera in rural East Pakistan", Amer. J. Epidemiology, vol.89, No.4 (1969), p.402. 180 their role must be suspected because of their large numbers and generally poor living conditions. It is suspected that the increased volume of migration associated with the West African drought is partially responsi- ble for the continuing diffusion of cholera in the West African Sahel during 1973 and 1974. Local Diffusigp_ The diffusion of vibrio cholerae in a community after its initial introduction may involve the complex interaction of a number of variables. Among the more important of these factors are the character- istics of the vibrio, the suitability of the environment for vibrio survival, the size and location of the susceptible population and use of control strategies. Under ideal conditions, a transmission cycle will be established and the disease will continue to spread until the equilibrium is disturbed through a change in the predisposing environ- mental conditions or a decline of the susceptible population. The infection will be contained if prevailing conditions are not conducive to its Spread. Figure 32 shows a model of the local transmission of cholera. The various subsections of the model will be described below, together with a number of African examples. Initial survival The survival or containment of cholera vibrios after their introduction to a community depends on vibrio characteristics, the form 181 FIGURE 32 LOCAL SURVIVAL AND DIFFUSION OF CHOLERA This model shows interrelationships between factors governing the local survival and diffusion of cholera from its initial introduction to its elimination. The diamond-shaped boxes Show major factors governing cholera survival and diffusion. These may be favorable or unfavorable for the survival of the vibrio. For example, a large susceptible population and ideal environmental conditions would be favorable, while a very small susceptible p0pulation and a suitable environment would be unfavorable. Unfavorable conditions, as a rule, lead to the contain- ment of the infection. AS the epidemic proceeds, the susceptible population decreases, eventually ending the epidemic phase. 182 LOCAL SURVIVAL, FIGURE AND 32 DIFFUSION OF CHOLERA water supply 'pollution of water supply unprotected water supply water pH, temperature sanitatiog: poor sewage. garbage disposal fly problem defecation habits susceptibles uncontrolled defecation Fl Favorable far cholera vlbrlae barriers U' Unfavorable for cholera vlbrlae vaccination — control program introduction treatment of virbio chlorination Chance of vibrio-environment\\F environmen fcontacft US for virbio 0"“ ° ival introduction urv f hange in h f d it virbio c an9€ 0 ens y rharacteristics environmental suscegziglf U condition popu U F en em City low profile, protracted u new introduction man-environment conta nment, of infection transmission elimination survival _ f - 1. .3: ecal contamination unerals. care of of environment preparation sick (home 57' of corpse vs. hospital) water. sewage system ga breakdowns (markets. eat no festivals) with hands ‘ person transmission man- to person cycle 1 environment It spread N I— water supply spread control programs ablutions ipcgia;:ng I piped treated - ti water Simply [fill V2031": 2" contaminationl * 0 con ac s - rivate well control of personal : cgnlnunity well. gatherings hygiene , change J. stream. pond of vibrio characteristic small remaining susceptible population increase of control measures F inapparent infections carrier/case ratio 1 decline in susceptible population large remaining 1 poor sanitation sewage disposal food consumption of uncooked foods vegetables grown with sewage foodsellers susceptible populatiggfggi 1' 183 of introduction, and environmental suitability. The hardiness of the strain of the introduced vibrios influences the duration of their survival and their ability to survive in less than ideal circumstances. An index cholera carrier who develops clinical symptoms is likely to seek medical aid and be confined to a hospital. In contrast, asymptomatic or mildly symptomatic carriers may continue to circulate freely and seed the environment with cholera vibrios. The suitability of an environment for cholera vibrio survival involves the interrelationship of a large number of factors. Cholera is a disease of poor sanitation. Overcrowded areas lacking adequate sewage and garbage disposal and with unprotected, polluted water supplies experience the highest attack rates. Such conditions are commonly found in the recently constructed, unplanned, low class suburbs which surround many large African cities. Rural communities generally have smaller populations and lower population density than the poor urban suburbs, but are also susceptible to infection because of the lack of sewage disposal and protected water systems. The practice of'defecating in gutters, along stream beds and "in the bush" is widespread in rural and low-class urban areas in Africa. Such defecation habits maximize the possibility of the pollution of food or drinking water with vibrios. The use of flush toilets or pit latrines greatly reduces the risk of virbio survival. In upper class, low density neighborhoods with protected sewage and water systems, cholera is not a serious threat. 184 FIGURE 33 URBAN SLUM HOUSING IN CAPE COAST (GHANA) This photograph illustrates some typical conditions in the West African urban slum areas which were frequently heavily infected with cholera. These conditions include poor quality housing, crowding, and a probable lack of piped water and sewage systems. The open gutter between the street and houses may provide an ideal surVival environment for cholera vibrios if it is used for defecation or the disposal of household refuse. 185 Various control measures may be undertaken to make the environ- ment less suitable for cholera survival. Construction of pit latrines, chlorination of water supplies, mass vaccination programs, environmental clean-up campaigns and ongoing surveillance to detect and isolate an introduction of cholera vibrios at an early stage are examples of control programs which increase the likelihood of vibrio containment. Where cholera has been established for a long time, such as India or Bangladesh, it often occurs seasonally. Cholera introduced during the off-season is likely to encounter unfavorable environmental conditions and be contained. Conditions during the cholera season wand tend to favor the survival of the infection. It was widely believed at one time that seasonal cholera epidemics were directly related to temperature or humidity.15 Cholera seasonality is probably related to climate indirectly as a result of such factors as an increase in travel in the dry season or a rise of water levels in ponds, wells, and rivers during the rainy season . Susceptible population The potential development of an outbreak of cholera is limited by the size of the cholera-susceptible population. Figure 34 Shows that the Size of the susceptible population in a community is influenced by a variety of factors. Susceptibility to cholera is relative, rather than 15 A.M. Kamal, "Endemicity and epidemicity of cholera", Bull. WHO, vol.28 (1963), p.284. 186 FIGURE 34 SIZE OF CHOLERA-SUSCEPTIBLE POPULATION The size of the susceptible population is important because it essentially defines the potential size of an epidemic. Such factors as population size, number who are initially cholera resistant, control measures, age structure and the socio-economic structure of the community affect the size of the susceptible population. However, susceptibility cannot be determined in isola- tion from the extent of vibrio pollution and virulence of the vibrio strains. Where there is exceptionally high vibrio pollution or a particular virulent strain, a considerably larger number will be susceptible than in an "average“ epidemic. 187 FIGURE 34 SIZE OF THE CHOL ERA - SUSCEPTI BLE SIZE TOTAL POPULATION Small POPULATION SUSCE PTIBILIT Y LEVEL OF Large Many HIGH ANTIBODY LEVELS EFORE ONSET ,Few STANDARD OF LIVING SOCIO-ECONOMIC STRUCTURE, High Low High Low CONTROL PROGRAMS MASSIVE VIBRIO CON TAMNATION A .e" ....... ........... IN AN 'hvERAOE" EPlDEMlC SUSCEPTISILITY Major AoE STRUCTURE V IBRD CONTAMINATIW ‘. 1.3:...- '.e‘ a", . '1' ' - ... ' Vi ' ' '. - '1 IIII Endemlc Area 188 absolute. Given a sufficiently large ingestion of vibrios, no person is immune. In a normal p0pulation, the level Of susceptibility Of the members varies from very low to very high. This susceptibility is a function Of such factors as vaccination, recovery from a cholera infection, sociO-economic status and age. Total population size is a basic variable in establishing the limits Of the susceptible population. In a village Of 1,000 persons, the limit of the local susceptible population is 1,000. The course Of an epidemic is likely to be much shorter than in a millionaire city such as Lagos or Ibadan. Within the total community, part Of the population may have a degree of immunity as a result of prior exposure to the disease or recent vaccination. The prOportion of initial susceptibles in the total population will be less in endemic than in newly-infected areas. Because cholera is associated with poor living conditions, knowledge of the sociO-economic structure is necessary for estimating the p0pulation's susceptibility. Control measures, such as mass vaccination and clean-up campaigns reduce the Size Of the highly susceptible p0pulation. However, vaccination provides only partial protection for less than six months, so even the vaccination Of the entire population would not eliminate the threat of cholera. The impact of control measures can be maximized by concentrating on low sociO-economic groups who run the greatest risk Of infection because Of their poor living conditions. Susceptibility is also a function of age. In endemic areas such as the Indian subcontinent, cholera mainly affects 189 children16. Adults, who generally have acquired high antibody levels as a result Of prior exposure to the disease, are less affected. In newly-infected areas, young to middle-aged adult males are infected first and most seriously. Adult males tend to be the most mobile section of the population and therefore have the greatest risk Of eXposure to infection. There is no change in antibody levels with age in previously uninfected countries.]7 The various factors diSplayed in Figure 34 govern the size Of the initial highly susceptible population in a community. With the progression of the epidemic the susceptible population declines in size relative to the cholera-resistant p0pulation. This results from the increase in cholera antibodies among individuals who have recovered from clinical or sub-clinical cholera infections. Consideration of the Size Of the susceptible p0pulation is incomplete if the extent Of vibrio contamination during an epidemic is not considered. Where there is only light contamination Of the environment, only the highly susceptible segment of the p0pulation is in dangercfl being infected. Where there is massive contamination Of the environment, virtually the entire population may be considered highly susceptible. The massiveness Of vibrio contamination accounts for the 16 McCormack, Mosley et al., Op.cit., p.403. 17 W.H. Mosley, A.S. Benenson and R. Barui, "A serological survey for cholera antibodies in rural East Pakistan", Bull. WHO, vol.38 (1968), pp.327-334. 190 heavy morbidity and mortality in the West African Sahel cholera outbreaks compared to other areas with similar initial susceptibilities. Transmission cycle If cholera is introduced to a new community containing a cholera-susceptible p0pulation and with conditions favoring its survival in the environment, an ongoing transmission cycle may be established. The conventional transmission cycle involves fecal contamination Of the environment and subsequent ingestion Of some contaminated element from the environment by a cholera-susceptible person. The cycle is completed by the multiplication Of the vibrios in his gut and the return Of the vibrios to the external environment in feces. In addition to dissemina- tion through this type Of man-environment transmission cycle, cholera may be transferred directly from a carrier's body to the body Of a cholera-susceptible person with elements of the environment such as water, fOOd or flies playing no intermediate role. This may be called person to person Spread. Many Observers, particularly those from Francophone Africa have emphasized the dominance of person to person spread in the West African cholera epidemic. For example, Felix has claimed that diffusion in Savanna West Africa was exclusively person to person spread.18 Although person to person spread may have been predominant in certain areas, other examples may be cited from many parts Of West Africa which I8 Félix, Op.cit., 1971c, pp.573-575. 191 implicate infected water supplies in the dissemination Of cholera. Therefore, one type Of transmission should not be over-emphasized at the expense Of the other. It is probable that man-environment and person to person Spread frequently occurred concurrently in a particular outbreak. Person tO person dissemination of cholera was frequently found to be associated with funerals and the care Of cholera patients in West Africa. The ritual preparation Of a corpse for burial and the collection and disposal Of the copious amounts of watery excrement passed by a cholera victim are particularly dangerous activities during a cholera epidemic. Contacts who are unaware Of the way in which cholera Spreads are unlikely to take the necessary precautions to avoid ingesting vibrios when in close contact with a cholera victim and the victim's environmental surroundings. Person to person spread may also occur in crowds of people such as those found at a market or festival. Personal hygiene and cultural factors influence the probability of ingesting vibrios. Persons washing infrequently and those who eat with their hands would have a relatively high risk of ingesting vibrios adhering to the hands or another part Of the body. It is also possible that the Mushim ablutions performed before all prayers, which involve the washing Of various parts Of the body such as hands, feet, anus, face and the inside Of the mouth could facilitate the transfer Of vibrios from the outer surface Of the body to the inside Of the mouth. 192 Control programs may be instrumental in reducing the risk Of person to person spread. The restriction Of gatherings and provi- sion Of adequate treatment facilities so patients need not be treated by members Of their family at home are examples Of control programs which may significantly impair person to person spread. Man-environment transmission occurs as a consequence Of fecal contamination Of the environment. This may result from the lack Of sewage collection and disposal systems, direct environmental pollution by uncontrolled defecation, floods, or temporary breakdowns Of the sewage system. One or'more of food, water, soil, or cholera-infected goods may form the environmental link in the transmission cycle. UnproteCted domestic water sources, such as streams, lagoons, or community wells are particularly susceptible to contamination. This contamination may occur as a result Of direct defecation into the water source disposal Of sewage and garbage, and use Of an infected container to draw water. In West Africa, areas along lagoons and river flood plains have tended to have persistent, water-borne epidemics. Food may be contaminated by flies which had come into contact with a cholera—infected medium. The transmission cycle in the Israeli epidemic Of 1970 involved consumption Of vegetables grown with untreated sewage containing cholera vibrios.19 Sewage is Similarly used for 19 J. Cohen et al., "Epidemiological aspects of Cholera El Tor in a non—endemic area", Lancet, July 10, 1971, pp.86-89. 193 FIGURE 35 A VILLAGE WELL, DOGON DOUTCHE (NIGER) One or a small number of community wells typically form the water supply for villages in the dry savanna. Because Of the crowds Of people who gather to draw water and herds Of animals brought to the wells the area around wells is frequently unsanitary. Leather devices used to draw water could support the survival Of cholera vibrios and be responsible for the transmission Of infection from one well to another. Therefore, it is possible for water to play a significant role in cholera diffusion, even in the dry environment Of the Sahel Savanna. 194 growing vegetables in West Africa in the vicinity of Kano, Zaria and other large cities. The numerous foodsellers found in Africa in markets and along roadsides are an important diffusion hazard. There is little or no supervision Of the preparation Of this food. Much of it is prepared unhygienically and sold in market places where the risk Of having the fOOd contaminated by flies or vibrio-infected people is at a maximum. In Schram's study Of cholera in Zaria, the infection was found to be clustered around the market.20 Relatively large numbers Of foodsellers other types Of merchants and koranic students who are largely dependent on food vendors' handouts for their subsistence developed cholera in Zaria. This evidence tends to implicate foodsellers as a focus Of the dissemination. Decline of an Epidemic Cholera epidemics in a community contain the seeds Of their own destruction. This results from the growth in the size of the cholera-resistant population and coinciding reduction Of the susceptible population. Decreased susceptibility results from the development Of vibriocidal antibodies as a consequence Of recovery from vibrio infection or from vaccination. Often, the end of an epidemics occurs after only a small proportion Of the population has developed clinical symptoms 20 R. Schram, "The 1971 cholera epidemic in Zaria, Nigeria", Savanna, vol.l, No.2 (1972), p.216. 195 FIGURE 36 ROADSIDE FOODSELLERS, PO (UPPER VOLTA) Markets, transport stops and even residential areas in African towns are frequented by large numbers of mobile or semi-mobile foodsellers. The preparation and sale of this food is generally unregulated. Foodsellers seem to have played an important role in the dissemination Of cholera in Zaria (Nigeria). It is probable that they were responsible for spreading cholera in other parts of West Africa as well. 196 _ .-.. _ LL_.-.-.-- _ A _ __ ---—___ ___,— FIGURE 37 FISHERMEN'S CANOES IN FIVE CONRIE CREEK. LAGOS (NIGERIA) This photograph illustrates conditions which favored the establishment and long-term survival of cholera in several parts of the West African Coast. The Creek, which is heavily polluted with sewage and refuse, contains brackish water suitable for vibrio survival. Fishermen are highly susceptible to Cholera infection because Of their close contact with water which may be highly contaminated with cholera vibrios. 197 Of the disease. This apparently premature disappearance may be explained by the typical high ratio of symptomless to symptomatic cases. As an illustration, if a town Of 1,000 peOple had only seven clinical cases, up to 70% of the population could have been infected if a carrier to case ratio of 100:1 is assumed. In addition to a decrease in the susceptible population, an increase in control measures, a mutational shift Of the vibrio or a deterioration Of the conditions facilitating surviVal Of vibrios in the environment could lead to a containment Of the epidemic before the critical point in the decline Of the susceptible population is reached. AS the susceptible p0pulation declines beyond a critical limit, or as other factors unfavorable to the continuation Of the epidemic become dominant, the epidemic peak passes and the disease again becomes inapparent. The infection may entirely die out, but if environmental conditions are conducive for its survival, a low profile, protracted man-environment transmission cycle may continue. Studies conducted in cholera-endemic areas in the Indian Subcontinent during an inter- epidemic period revealled that there was continuing vibrio survival and 2‘ This circulation, despite the total absence of clinical cases. protracted, invisible survival may be interrupted by a new introduction of cholera from an outside source, a change in the environnental conditions, or a change in the vibrio characteristics, possibly as a 2] R. Sinha et al., "Role Of carriers in the epidemiology Of cholera in Calcutta", Indian J. Med. Res., vol.56, No.7 (1968), pp.964-978. 198 result of vibrio mutation. The change in environmental or vibrio characteristics may be either unfavorable for vibrio survival, resulting in its total elimination, or favorable, thus promoting an increase in vibrio contamination and a new epidemic. 199 CHAPTER 8 SUMMARY AND CONCLUSIONS The organization of the concluding chapter will be based on the series Of study Objectives outlined in the first chapter. Spatial and Temporal Patterns Of Cholera Diffusion After six major cholera pandemics in the nineteenth century, cholera had virtually become confined tO the Indian subcontinent after 1920. In 1961, a previously minor cholera biotype known as cholera E1 Tor Spread from its endemic focus in Southern Celebes. This marked the beginning of the seventh cholera pandemic which has affected countries in South and South-East Asia, the Middle East, Southern Europe and Africa. There have been five cholera epidemic zones in Africa located in North Africa, West Africa, East Africa, Angola and South-Eastern Africa. The diffusion of cholera has followed different patterns in each Of these areas. In East Africa, it was mainly confined to nomadic peoples in Southern and Eastern Ethiopia and adjoining countries. The sudden appearance and disappearance Of Cholera in widely separated parts of North Africa complicates any determination of the means Of transmission. The Angolan epidemic has centered on Luanda. There have been four phases namely diffusion from Luanda.to other coastal cities, disappearance Of cholera for several months, reappearance in an endemic form in Luanda, and renewed diffusion from Luanda to coastal and railway towns. In 200 Mocambique and Malawi, cholera diffusion axes followed railway lines. The transfer of the disease across the Zimbabwe border is suspected to be a result of liberation trOOp movements. There were four types Of diffusion in West Africa, namely coastal, riverine, urban hierarchical and radial contact diffusion. From its point Of introduction in Guinea during August, 1970, the infection moved along the coast Of West Africa, reaching Southern Cameroun in February, 1971. The Spread of cholera was in conjunction with the migrations Of fishermen. The coastal epidemic was based in rural fishing villages along lagoons, which provided ideal vibrio survival conditions. Coastal cities were frequently infected from nearby fishing villages, but the duration Of infection was quite Short in these cities. The coastal epidemic was principally water-borne and charac- terized by relatively low morbidity and mortality. Riverine diffusion refers tO the cholera diffusion axis centered in the Middle Niger Valley. The infection, whiCh was introduced to Mopti by a traveller from Abidjan, quickly spread along the entire length Of the Malian and Nigerien portions of the Niger River. A secondary phase consisted of the more gradual penetration Of the upland valley periphery and the Senegal Valley. The infection proceeded down the Senegal Valley to the Atlantic Coast. The third phase is the continuing occurrence and diffusion to new areas related to the Sahelian drought. Distance from the point of origin in MOpti and riverine environment are the two variables found to significantly correlated with the occurrence and time Of onset Of cholera in Mali. 201 Urban hierarchical diffusion occurred in Nigeria and South- Central Niger. Cholera Spread from Lagos to the regional centers of Ibadan, Kano and Port Harcourt, and from these cities to lower ordered cities within their spheres Of influence. Diffusion followed the major road and railway lines. Distance from Lagos and urban importance were the two variables most significantly related to cholera occurrence and time Of onset. Radial contact diffusion was Observed in the vicinity Of Lake Chad in mid-1971. Because Of extremely high morbidity and mortality rates, survivors fled from infected vaillages and carried the disease to other villages. The epidemic front moved outward from Lake Chad like a wave. The containment Of the epidemic at the beginning Of the rainy season is perhaps a result of people remaining at home to farm instead Of fleeing. A study of local cholera transmission in Lagos revealed a complex pattern of epidemic centers shifting between the four major geographical regions Of Lagos Metropolitan Area. A peak in the epidemic curve of one Of the four regions was generally followed after a few days by peaks in the adjoining regions. These increases in morbidity apparently originated at différent times between December, 1970 and March, 1971 in Lagos Island, the North-Western Peripheral Suburbs and in Mainland Lagos. Within Lagos Island, the cholera epidemic began with an explosive outward movement Of cases from the center of the island followed by an implosive return to the island's center. This sequence 202 resembled the pattern Of a physical wave reflecting Off a barrier. However, the explosive-implosive sequence has not been satisfactorily explained. Spatial Diffusion Principles The three major types of spatial diffusion identified by Gould,1 namely, contact diffusion, hierarchical diffusion, and relocation (long distance) diffusion were Observed in connection with the diffusion Of cholera in Africa. For example, urban hierarchical diffusion occurred in Nigeria and possibly Angola. The initial importation of cholera by air travellers into Guinea, Angola and Mocambique, as well as the trans- fer Of infection from Abidjan to Mopti are examples Of long distance or relocation diffusion. Linear contact diffusion within a channel was Observed in the valleys Of the Niger and Senegal River and along the West African coast. Unchannelled radial contact diffusion occurred in the Lake Chad vicinity and in Northern Kenya. The patterns Of inter-divisional cholera diffusion Observed in West Africa may be compared to those Of cholera epidemics in the 2 United States in the nineteenth century which have been described by Pyle. The linear pattern Of diffusion along waterways with onset a function 1 Gould, Spetial Diffusion, Washington: AAG Commission on College Geography Resource Paper No.4, 1969, p.35. 2 G.F. Pyle, "The diffusion of cholera in the United States in the Nineteenth Century", Geog, Analysis, vol.l, No.1 (1969), pp.59-75. 203 Of distance from the source of infection, which was found in Mali, is very similar to the 1832 American pattern. The urban hierarchical diffusion system Of Nigeria corresponds to Pyle's finding of urban hierarchical diffusion in the United States in 1866. Pyle attributed the change in diffusion pattern in the United States to deve10pment Of an integrated transportation system and a significant increase in urbanization between 1832 and 1866.3 Parallel differences exist in Nigeria and Mali in the 1970's. Nigeria had twenty-five cities in l963 exceeding 100,000 population, while the second largest city in Mali has only 32,000 peOple. Central and Eastern Mali's main transportation artery is the seasonally-navigable Niger River. Nigeria has a rather well developed road and railway network. The relatively simple spatial concepts Of permeable and impermeable barriers, channels and central place attraction were found to be useful in explaining diffusion patterns occurring in various areas. The most prominent type of barriers were physical barriers. The Atlantic Ocean, and Sahara Desert were barriers defining the Northern Southern and Western boundaries Of the cholera-susceptible areas. Relatively unpOpulated areas with poor transportation networks were barriers to diffusion in the Middle Belt, Central Chad and Southern Cameroun. Vaccination barriers and travel cordons were barriers of local significance. 3 Pyle, Op.cit., p.71. 204 The channels most significant for diffusion were physical and transportation channels. River valley and coastal physical channels were most prominent in areas with relatively poor transportation and . where population density was greatest along the corridor. Roads and railways frequently were important diffusion axes. Central place attraction and the resulting large volume of travellers assured the infection of most important urban centers within the cholera-infected areas. Specific urban functions such as markets, transport functions and staging of traditional festivals were frequently found to be responsible for the introduction and dissemination of cholera into otherwise insignificant villages such as Goulfey (Cameroun) and Thiomene (Senegal). .Barrier, channel and central place effects are useful in explaining the patterns of inter-divisional diffusion of cholera, but are of little value in analyzing local diffusion, especially in large and environmentally complex urban areas. It is possible that a "probability of contact" model, perhaps using Markov Chain Analysis or the susceptible p0pulation 4 immune population models developed by mathematical epidemiologists could be successfully adapted for the analysis of cholera diffusion in local areas. Model of Cholera Survival and Diffusion The model of cholera survival and diffusion is composed of two parts, namely, local survival and diffusion and external or inter- 205 divisional diffusion. The effects of distance, barriers, channels and central place attraction, which are the main elements of external diffusion have been described in the previous sections. The cholera transmission cycle and its relationship to parti- cular environmental conditions in a community must be considered if local survival and diffusion processes are to be understood. General systems models are particularly useful for this purpose because they focus attention on the selection of relevant factors and a consideration of their interrelationships. The main elements in the local survival and diffusion system are the survival ability of the vibrio strain, fonn of introduction, environment for vibrio survival, size of the susceptible p0pulation and predisposing factors for the development of man-environment or person to person transmission cycles. The decline of an epidemic may result from changes in vibrio characteristics, the successful implementation of control measures or a substantial decline in the cholera-susceptible population. {Various facets of the standard of living, in particular environmental sanitation, pollution and protection of water sources, and population density are the major determinants of the suitability of an environment for cholera survival. Cholera primarily affects poor peOple living in crowded, unsanitary conditions. Programs such as cleanup campaigns, chlorination and vaocination may be used to reduce the favorability for survival. 206 The absolute and relative size of the cholera-susceptible p0pulation is a function of total p0pulation size, the number with initially high cholera antibody levels, the socio-economic structure of the community, the extent of control programs and age structure. However, the number of susceptibles in a population cannot be determined without knowledge of the extent of virbio contamination in a particular epidemic. The risk of person to person spread is greatest where there is close contact between a vibrio carrier and susceptibles. Funerals, home care of cholera patients, and attendance at markets have promoted person to person dissemination in Africa. Personal hygiene and such practices as eating with hands or performing ablutions are suspected to have been involved in person to person cholera transmission. The develOpment of a man-environment transmission cycle stems from unsatisfactory disposal of sewage or uncontrolled defecation, leading to the contamination of the environment with vibrio-bearing feces. Establishment of a transmission cycle through pollution of water sources is the most comnon fonn, but soil, food or other goods may also become contaminated. Although the dominance of person to person Spread has been emphasized by several observers of cholera in West Africa, both types of transmission were found. Simultaneous man-environment and person to person spread probably occurred in many communities. A variety of control programs were used to combat cholera in Africa. Mass vaccination programs were most prominent. It has been 207 claimed that vaccination cannot be justified when its cost is compared to the cost of treating cholera cases.4 However, the scarcity of treatment facilities in rural Africa, the excessive mortality experienced in areas such as rural Mali and Chad where treatment was unavailable, compared to the small outbreaks in vaccination protected cities show that vaccination is a valuable and perhaps essential cholera control method in Africa. Other controls such as the prohibition of travel and market attendance were only partially successful. Belief systems and attitudes which were incompatable with the functioning of organized control programs created a problem in various areas, including Ibadan and Coastal Togo. The ultimate control program involves the elimination of environments suitable for vibrio survival through improved sanitation and housing and provision of protected water supplies. However, this would require huge sums of money and take decades to achieve. Endemicity Cholera appears to have disappeared from East Africa, where there have been no reported outbreaks since l972. Reports of cases have continued from the other four African epidemic zones during l973 and l974. Twelve West African countries reported cholera in l973 and the first half of l974. These reports have been relatively infrequent 4 D. Barua, "Cholera vaccination as a tool for cholera control", Bull. Soc. Pathol. Exotique, vol.64 (l97l), pp.652-659. 208 and scattered, suggesting a significant under-reporting of cases or that there are many small, scattered pockets of endemic cholera in West Africa. The areas which have reported cases most frequently include Coastal Ghana, the Douala-Victoria area in Cameroun, Montserrado County (Monrovia), Liberia and the Lagos-Ibadan region in Nigeria. These locations all include areas of lagoon coast. Rainy season cholera epidemics, followed by the apparent disappearance of the infection have occurred in the Senegal Valley in l97l, 1972 and l973. The continuing presence of cholera in the Sahel is related to drought conditions. Cholera has persisted for four years, so permanent endemicity in Africa must be considered to be a definite possibility. If cholera continues to survive in Africa, new epidemic outbursts comparable to those of l970-l97l are probable. With time, antibody levels in the population decline and number of children with no antibodies due to their never having been exposed to cholera increases. As the level of susceptibility of the population rises, the risk of a new epidemic grows. However, the apparent elimination of cholera from East Africa and such West African countries as Sierra Leone, Ivory Coast and Togo, which would seem to be favorable locations for cholera survival shows that permanent endemicity is not inevitable. * 209 APPENDIX 1 MAP CODE FOR WEEK OF CHOLERA ONSET* Week Qatg§_ l Aug.lS—21, ljflgl 2 Aug.22-28 3 Aug.29—Sept.4 4 Sept.5-11 5 Sept.12—18 6 Sept.l9-25 7 Sept.26-Oct.1 8 Oct.2-8 9 Oct.9—15 10 Oct.l6-22 11 Oct.23-29 12 Oct.30-Nov.7 13 Nov.8-l4 14 Nov.15-21 15 Nov.22-28 l6 Nov.29-Dec.5 l7 Dec.6-12 18 Dec.13-l9 l9 Dec.20—26 20 Dec.27—Jan.2, ngl_ 21 Jan.3-9 22 Jao.lO-16 23 Jan.17-23 24 Jan.24-3O 25 Jan.31-Feb.6 26 Feb.7-13 Week 27 28 29 3O 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 Dates Feb.l4-20, 1271. Feb.21-27 Feb.28-Mar.6 Mar.7-l3 Mar.l4-20 Mar.21-27 Mar.28-Apr.3 Apr.4-1O Apr.ll-l7 Apr.18-24 Apr.25-May 1 May 2-8 May'9-15 May 16-22 May 23-29 May 30—June 5 June 6-12 June 13-19 June 20-26 June 27—July 3 July 4-10 July 11-17 July 18—24 July 25-31 Aug.l-7 Aug.8-l4 Week 1 coincides with the first confirmed outbreaks in Africa (Guinea and Libya). Meek 53 54 55 56 57 58 59 60 61 62 63 64 65 66 67 68 69 70 71 72 73 74 75 76 77 78 79 80 Aug Aug. Aug. 210 APPENDIX 1 (cont'd) Dates .15-21, 1971 22—28 29-Sept.4 Sept.5-ll Sept.12-18 Sept.l9-25 Sept.26-Oct.2 Oct. Oct. Oct. Oct. Oct. .7-13 Nov Nov. Nov. .28-Dec.4 Nov Dec. Dec. .19-25 Dec Dec. Jan. .9-15 Jan Jan. .23—29 .30-Feb.6 .7-13 Jan Jan Feb Feb. Feb. 3—9 10-16 17-23 24-30 3l-Nov.6 14-20 21-27 5-11 12-18 26-Jan.l, 1972 2-8 16-22 14-20 21-27 Week 81 82 83 84 85 86 87 88 89 9O 91 92 93 94 95 96 97 98 99 100 101 102 103 104 105 106 107 108 Feb. Mar. Mar. Mar. Mar. Apr. Apr. Apr. Apr. May May May May May June June June June July July July July July Aug. Aug. Aug. Aug. Sept Dates 28-Mar.5, 1272 6-12 13—19 20-26 27-Apr.2 3-9 10-16 l7-23 24-30 1-7 8-l4 15-21 22-28 29-June 4 5-11 12-18 19-25 26-July 2 3-9 10-16 l7—23 24-30 3l-Aug.6 7-13 14-20 21-27 28—Sept.3 .4-10 211 APPENDIX 1 (cont'd) Week Date§_ Week Dat§§_ 109 Sept.11-l7, 1972 137 Mar.26-Apr.1, 1213 110 Sept.18-24 138 Apr.2-8 111 Sept.25-Oct.l 139 Apr.9-15 112 Oct.2-8 140 Apr.16-22 113 0ct.9-15 141 Apr.23-29 114 Oct.l6-22 142 Apr.30-May 6 115 Oct.23-29 143 May 7-13 116 Oct.30-Nov.5 144 May 14-20 117 Nov.6-12 145 May 21-27 118 Nov.13-19 146 May 28-June 3 119 Nov.20-26 147 June 4-10 120 Nov.27-Dec.3 148 June 11-17 121 Dec.4-lO 149 June 18-24 122 Dec.1l-l7 150 June 25-July 1 123 Dec.18-24 151 July 2-8 124 Dec.25—31 152 July 9- 15 125 Jan.1-7, 1973_ 153 July 16-22 126 Jan.8-14 154 July 23-29 127 Jan.15-21 155 July 30-Aug.5 128 Jan.22-28 156 Aug.6-12 129 Jan.29-Feb.4 157 Aug.13-19 130 Feb.5-1l 158 Aug.20-26 131 Feb.12-18 159 Aug.29-Sept.2 132 Feb.19-25 160 Sept.3-9 133 Feb.26-Mar.4 161 Sept.10-16 134 Mar.5-ll 162 Sept.l7-23 135 Mar.12-18 163 Sept.24—3O 136 Mar.19—25 164 Oct.l-7 212 APPENDIX 1 (cont'd) West we; Week Bares. 165 Oct.8-l4, 1212 184 Feb.18-24 166 Oct.15-21 185 Feb.25—Mar.3 l67 Oct.22-28 186 Mar.4-lO 168 Oct.29—Nov.4 187 Mar.11-l7 169 Nov.5-11 188 Mar.18-24 170 Nov.12-18 189 Mar.25-3l 171 Nov.19-25 190 Apr.1-7 172 Nov.26-Dec.2 191 Apr.8-14 173 Dec.3-9 192 Apr.15-21 174 Dec.10-16 193 Apr.22-28 175 Dec.17-23 194 Apr.29-May 5 176 Dec.24-30 195 May 6-12 177 Dec.3l-Jan.6, 1212_ 196 .May 13-19 178 Jan.7-13 197 May 20-26 179 Jan.14-20 198 May 27-June 2 180 Jan.21-27 199 June 3-9 181 Jan.28-Feb.3 200 June 10-16 182 Feb.4-1O 183 Feb.11-17 213 APPENDIX 2 MAP PLACE NAME CODE Code Division 9612131221 K5 Kribi L5 Lomet Djerem M5 Mefou Dahomey A1 Atlantique M1 Mono 01 Ouémé' Zl Zou Ghana A2 Ashanti 82 Brong-Ahafo C1 Central E1 Eastern V1 Volta Wl Western Ivory Coast A6 Aboisso A4 Adiaké A8 Alépé 83 Bingerville Dl Dabou 02 Grand Bassam Liberia B4 Bong C2 Cape Mount G3 Grand Bassa G4 Grand Gedeh M2 Maryland M3 Montserrado N1 Nimba Mali B6 Banamba BlO Bankass 05 Djenne K7 Kadiolo K8 Kangaba M6 Macina T3 Touminian Y3 Yorosso Code Division Nigeria All Aba A12 Abakaliki A14 Aboh A16 Auchi 818 Badagry 825 Brass 010 Degema E4 Eket E5 Egbado I7 Ikot Ekpene 18 Ikom 113 Ilesha K15 Kabba N7 Ngwa (Umushia) 05 Owerri O6 Okene 010 Opobo 012 Oshun 015 Okitipupa U3 Ughelli W3 Western Ijaw W4 Warri Sierra Leone E4 Eastern N4 Northern S4 Southern Togo A5 Akposso A7 Atakpame A10 Anecho L6 Lama-Kara M9 Maritime N6 Nuatja P6 Palime’ $9 Sokodé T6 Tabligbo T7 TséVie V7 Vogan Code Division Upper Volta 831 832 833 834 D30 D32 D33 034 K30 K31 K32 K33 Boulsa Barsalago Bogandé Boromo Djibo Dédougou Diébougou Diapaga Kaya Kombissiri Kongoussi Koudougou APPENDIX 2 (cont'd) Code K34 M30 033 P30 P31 530 S31 T30 T31 T34 Y30 Division Koupéla Manga Orodara Pissila Po Saponé Seguénéga Titao Tiébéle Tenkodogo Yako 215 F\P\oP \\00m mm Emcmmmumoz .\F\_ we Meagan: F\O\m \\mmp mm Emcm< _m \\om~ mm mewpcmumcoo \\Nom em mcm_mp< ”\O\oF P\oxp \\~N wm mm23< P\o\op \\N mo mnmcc< _* _o cowuumecw mo mxmmz\m;pmmu\mmmmu mo xmm: .mpnmp mg» mo mama ummp on» mom .muocuoou ~* upon mpmpaeoucH + mcucoz z .mrnmpwm>m «you 0: “xmmgnuao umpcoamm m .vgoomm PmuwmopowEmqum xpxmmz on» cw vmagoamg ago: “so .mommu umuumamam m .xppasoca mmmp “Loam; mmvgpczoo mEom mm .mmpgm> covgma mew“ csmfi mzh .mpmxomgn cw Lupumr map .umpmwp mew mmgzm_w :uon .mucmnwucw agmpocu we meEE:m ngoomm PMUvmopovsmuwmm.x~xmmz Pmsccm mcp cw vmpgoamg Longs: ms» Eogm mgmmewv mummu we pupa» mg» can: .vmumwp :mmn m>mg mmmgu .mwugzom gmguo Eogw mrnmpww>m mew mumu mumpanu mgoe mews: "mmogzom Agmucouom .mgmpogu mo mpgoamg xpxmmz .¢Nmpuonm~ .ucoumm pmuwmoPoPEmvwam apxmmz ”mugzom xcmewgm .¢nmp .>4:w u ommp .Hm:un< “onHommzH no onHh~4HHonmoz .hmmzo mwo covuuwmcw ..vo mxmm3\mcummv\mmmmu ....o xwmz Aum::_pcogv m xHozwaad 217 Nomv o\m O\a wN\mmm omm\~mp.~ beech op\m\-~ me _ecmo axe: ~\o\w oe wszewo P\O\m op\m5_\ewop oe _Lmzu-mcomoo P\o\P mm apoecez-w=mzmz ~\N\o mm uppmx pm mco»z _\_\N we segues pm 504 F\O\_ F\P\F on caOEem P\0\F om inex _\0\P ~\o\F Pm sm¥.z use: m\M\N “\m\ue mm «sweezmz-mmecem m\P\m ~\N\Nm mm Ee¥.z P\O\_ ~\o\m mm zoewz N\o\m a\m\mm _N\e\NmF mm e_eoeu_> F\o\m m\e\mm e\ o\NN mm cease mp\m\Fm e\~N\Nm mm omen: F\,_\mu KN Am_m=oov m\0\v ¢¢\~P\Nmm mp\mm\mmm mm peso: 230mmz02 eump mump mum. .~m_ o~¢_ ummco co_mw>wo cozummcw ....o mxmm3\mguwmu\mmmwu ..vo xmmz Aem==_ocoov m xHszaaq 218 o~\pm. ~e~\mofim cm\mem 4uzozea ~P\em P\m ammm\mmmm Aeeoe m\F\NN Ne .m_wheeme .\M\e we _eemao mm\mm\upm ee wnnmx-osmz _\op\mm amp Am_eahe:_¥ «_Hv _\P\e m\_wN\Nmm me one ~\o_\me he Emcee N\m\NN mm x5e; “cod M\P\m mp\mmm_\momo Ne wagvzmmm-wzezu o_o .530me $0 mxmmz\m;pmmu\mmmmu mo xmmz Aemzewpcouv m xHozmaa< 219 Aoeev ~\m mm\omm m\o 4um «mop mump Nump pump cmm— pmmco cowm_>wo :oPpumwcw mo mxmmz\m;ummu\mmmmo mo xmmz Aemacwucouv m xHozmaa< 220 a F\_ o_ m__w>cmmewm a N% F_ m___>onm< ze\+0\wpm mm _meee ZM\+M\+MON mm\amm __ cmfiewn< z_\o\e Bea __ czomcmne quou >mo>~ oo\ooo~ F sowcwuv> a acxmcou ¢\M\mp Fm newcozoz m\o\e P\o\e M\o\~ me\mM\mFoP m\e_\©om m_ Amcuu_o :owuumwcw mo mxmm3\m:pmmw\mmmmu ....o xwwz Aemacwpcoov m xHozmaa< apomv 221 O\oe m~\~m 4zu¥ Amoev Amway 4m=cumo 2F\O\m mm co:MJ ncmgo zN\+o\om Pm oxogaewo ze\+O\¢o om zones «mop mump Nump meF o~m~ ummco co_mw>_o :owuumma mo mxwm3\m£pmmv\mmmmu mo xmmz Aumzcwucouv m xHozmaa< 222 «\0\_. om_ mgmomcmz ¢\+o\we P\O\m me_ masmmx m\+o\Po .\o\m QN_ =F=~ecmzwzu e\+o\mm o_\ep\~mp NoP azazxrzu e\o\mm P\O\m cup atsucmFm onwmm zmmxpzom _3<4mH4 Auemv Amoov A_~_v ~\~PP mp\mmoF m\ee_ ee\emF o\om 4occozv IM\~\N_~ 20\wp\~mop z_\m\mo z~\o\~m N oeazzamocoz 2N\m\mm Pm ucm_»cmz zr\o_\oe om :mumo eceeu ZM\m\oF om emmam tango zN\e\o 2m\cm\mo om peso: mama ZM\o\o om .ou meom «Hmmeo eemp mem_ Num_ Pea. o~m_ “mace co_mw>_o :owuuwmcw mo mxmm3\m£umwt\mmmmu ....o xmmz hemgcwpcouv m xHozwaaq 223 .\o\m om omc0mc< ¢\m\~ e\+em\opm o_ ego owo cowuommcw ....o mxw03\mzpmmU\mmmwu mo xmwz Aemacwpcoov m xHozmaaq 224 A~m~.ev P~\mm~ +mm\+m- o\~ “me\oomm mo~\mmmm 4<¥ m\m\mw mp =ooo:anop e\mp\me mm execmz m\o_\+mm_ m\mp\+mmp m\o~\mm m*e\+m~\o~o mp Emeczou P\_\o F\F\m mp\mM\NFF wem\+op\mm~ m_ .mcwo _\F\F mm smzzom cum. mnm_ NNmP Fem, o~m_ ommco :o_mw>_a covuummcw mo mxmwz\m;umwv\mmmmu $0 gum: Aemzcvgcoov m xHozuaaq 225 o\n o\om Jehoe to: m2 8.25 F\o\m _\o\mm we Loemz .\o\F eop acovcmx ~83 2: mm“. me ouoomoz Ame_v Amm_.0 mp\mm_ N.\¢N +No\+mmm_ 4 P\o\n P\O\~ _~_ meow _\M\m ~\O\m .5, :aeocemz m\FF\me ~\o\m eo_ maven: m\e\mm o\_\m_ mo_ owoewcu _\o\. M\F\m mo_ esom=_em;u m_\em\omm mp\ep\om~ mop newmm F\o\mm o~_ macaw <4ou wo cowpum$cw mo mxmmz\m;pmmn\mmmmu $0 xmmz Aemchpcouv m xHazmaaq 227 w\~me\m~o Rm “scene ~\+o_\+~op ~\N~\ea Fe ewccwz F\~\mm oe «swampmz N\N_\.m mm ~_memz _\eo\om NM .02300 «mozHN ~\N\mp mm\omp\mmpp a _N szanmp_we m\m\mm _M\¢mm\-ep a _N acme a\em\mPF mm saw eF\NM\mmP Fm Em_-=o M\o\+e F\O\m mm\mm\omm _N swamwz P\O\_ _N wamcwpva >wz_o :owuum$:.~ $0 mxmmz\m;pwmfi\mmmmu $0 x003 Aum=:_0couv m xHozmaa< 228 m 00 annex ~P\-\mo~ om cwcopm m\mp\mep <¢<3¥ m om ccmzugoz +e\o~m\mm- mm ocmx m oz0LQ 30000—0 0\mp\+wm 0m wcmxzz x mm wugaxmz a 0m 0x000 M\mp\0m om .>0La mzcmm 3wo coruum$cw $0 mx003\m:ummu\mmmmu $0 x002 Aem=CPp=ouv m xHazuaa< 229 _N\+m\+oom mm m$LMN mp\+~\+mm mm mc$mum¥ mp\mm\+mpm om ocaumx a 0:0:30 0\O\PP m\o\m pxo\¢ ~N\~m\moo_ 4_o cowuum$cw $0 mxmmz\mgummc\m0mmu $0 x003 $025008 m x8503 230 N\m mm 00H ¢\m mm mcmaxu poxH 0\m mm umxm m\m mm guanou N\m mm an< m\¢\mmp m\mop\m~op $mam a $gzm> m\~\+m mm ouoxom m 00 _nnmx m ammzo a 00 smgzmg< 0\m mm .>oca ouoxom m 00 anew: m we mgommpcox x mm muwm m 00 on=n< M\m .>oza camwz m\m_m\+mmm $mmzz$moz m msmm m munmm N\O\ee mm Awgsmsewmzv F\N\Nm x mm scgom F\m\~P x mm Peusem m . mgm~< we mzmsmu< m\~$\opp _\N\+o . $m$o cowpum$c$ $0 mxmmz\m;ummu\mmmmu $0 x003 Aemscwucoov m xHozmaae NM\~¢¢ m\mop 4<$0$ 231 «\F\N Nap vNemzwnu m\~\~. ~w_ 2800mm_oz ¢F\w\mo_ F\O\_ e~_ unevawzu o\O\oe o“. 0000: o~\om\~m~ N\Q\Ne we, ewzzen «Hmmooxm o~\Nm. em\mmo.~ Aoe~.mv Axe~.-v e\mp 4<$oe x mm annex$po mp\a om oso x mm ozo m_\m mm cacao m,\m om oeco _N\m mm mgma_H x mm Aauo saoth a mu 350023 NN\¢ mm anaflH ¢F\m mu aCH m_\m~\oem mm\m_P\+Nmn NN ceemnH op\m NM 00000 x mm “oemnmmv a mN Aenmmv a NP\¢ mm augxomn< «\e\n mm\-\eeu m\N\om_ oN\Pw~\m¢op zmMHmmz “\m mm 0»: ~\m mm . ogoao ~\m we «tango eum~ mum, N~m_ Fam_ on_ “mace :owm_>0o 00$u00$c$ $0 mxwmz\mcp000\mwmmo $0 x003 Aem==_pcouv m xHozmaae 232 AFPNV Ammmv m0\$0~ m~\+mop 4<$0$ 0\N\m “\mp\mm m Aczoammzav m\m\om m\m\m op 2mmhmm3 P—\m\ov F\o\0 mp 2mmz$aom mm\mm\P$$ m\0\+Pm m zmmz$moz 0\0\m mp 2mm$m=mqm m\0\0m m\0p\Fw 00 00:04 ~m\mop\mmnp P\$\mop —\0\P 00 Jummaofio m\m\m~ ¢\~\$_ m\F\m mm u$mm> ¢<0 4$o 00$uum$c$ $0 mxmmz\m:¢mmn\mmmmu $0 3002 Aem=:_pcoov m xfiozwaae 233 N\0\m mm\0\$~ $\$\mm op mac; m\0\0$ MZOJ m ommoaxq $\o\$$ M\o\0m om _wEmaxmp< ~\o\$$ $N ws$$ma _\o\e Pm enumsz 2 m\m\mm NN\0\mo m\m\m$ m_ osuwc< o$ mzH$Hm$o co$uum$c$ $0 mxmmz\mcumwu\mmmou $0 x003 Aemzcwpeouv m xHozmaa< 234 Ammuv op\$$0 4<$O$ m0 umuaa< m0 000002 «0 0x0004 m\o$\$$0 Wm$ «map m$m$ N$m$ $$m$ ommp ummco co$m$>$o c0$uum$c$ $0 mxmmz\m:ummv\mmmmu $0 x002 Aumgcwucouv m xHozmaa< Apmmpv 235 mom\mem o\F on\mmmP Ammv m<$0$ P\o\m . mm? oxm» m~\oe\mm_ NN mmm30$ F\O\_ . em, wpmmmm$ m\mm\_em em mmm$$ m\o\P om_ mmwmmmmmm ~\o\m mm_ mmommm ~\o\m _m_ mmmmm_m m\m~\mm “\em\m_m oe emmmmo m\P\o_ e\e\m m_ mm$=omm=o ~N\m~\mm~ _\o\p oe mommommmmmo F\o\P m\_m\_mP mm mmmoz m\m\~m mm” m_mm=mx m\o\m mm? amazommox _\O\_ Nm_ Pmmmommox m\o\m mm, Pewmm$meox m\P\m $m_ mmmx mmm mstmmm.z mmmm m\mm_\mme om\m_~\fimm m_ $.om m\op\,m m_\Fm\mem m em mmmmm ~\~\~ mmp mmmmom P\o\P KN mmmm_=omm omom <$40> mung: ¢$m$ m$m$ ~$m$ Pump cmmp “mmco co$m$>_o co$uum$c$ $0 mxmmz\m;ummu\mmmmu $0 x003 Ammmmmmcmmv m xmmzmmmm 10. 11. 12. 13. 14. 236 FOOTNOTES 1971 Algerian data from Mered, 1971. Briefly noted by Felix, 1971C. No details given. The following data are reported by Comité’centrale de lutte contre 1e cholera (1971) for 1970 and up to February 13, 1971: Mono (1970): 1,018/95, (l97l);’1,222/203; Atlantique (1970); ggz21, (1971): 1,794/401; Oueme (1970): 30/5, (1971): 1,455/ 8. i No official report. According to the New York Herald Tribune, August 1, 1970 there were 1,500 victims in AlexandrTa and several in Cairo. Outbreak at Shep mentioned in The London Free Press, London, Ontario, September 24, 1974, p. 33. Reported in Félix, 19710. Reported in Bourgeade, Rive, et. al., 1973. Félix, 1971A reports the occurrence of at leaSt 5,000 cases and 2,000 deaths in Dire and Goundam in 1970 among sedentary farmers, plus an unknown number among fishermen and nomads. The Moroccan epidemic was almost certainly more serious (space, time, number of victims) than reported. Lagos State 1970-1972 data from B. Dada, 1972. Mid-Western State 1971 data from Amu, l97l. Reported by Félix, 19718 and 19710. According to Ben Rachid et. al., 1971, the 1970 Cap Bon outbreak included 22 patients at Kéliéba, 13 at Boukrim, 2 at La Goulette, and 1 at Kamara and Menzel Teminé. Imported cases in Europe in August - September, 1973 had come from Tunisia. 237 BIBLIOGRAPHY Books, Theses and Statistical Documents Africa Diary: Weekly Record of Events in Africa, vol. 9 (1969) — vol. 12 (1972). . Annuaire Statistique: 1971. Bamako: République du Mali - Direction Nationale dB plan at de la statistique, 1973. Annual Abstract of Statistics: Nigeria, l966. Lagos: Federal Office of Statistics, 1966. Bailey, N.T.J., The Mathematical Theory of Epidemics. New York: Chas. Griffin 8 co., T957. Blalock, H.M., Jr., Social Statistics (second edition). New York: McGraw-Hill, 1972. Christie, J., Cholera Epidemics in East Africa. London: McMillan, 1876. Felsenfeld, 0., The Cholera Problem. St. Louis: W.H. Green, 1967. Fox, J.P., Hall, C.E. and Elveback, L.R., Epidemiology: Man and Disease. London: MacMillan, 1970. Gould, P.R., Spatial Diffusion. Washington: Association of American GeographersCommission on College Geography, Resource Paper no. 4, 1969. Guid' Ouest-Africain 1971-72. Paris, 26th edition, 1971. Mabogunje, A.L., Urbanization in Nigeria. London: University of London Press, 1968. MacMahon, B. and Pugh, T.F., Epidemiology Principles and Methods. London: J. & A. Churchill, 1970. May, J.M., World Atlas of Diseases. New York: American Geographical Society,*1950. ., The ecology of human disease. New York: M.D. publications, 1958. McGlashan, N.D. (ed.), Medical Geography, Techniques and Field Studies. London: Methuen and co., l973. . 238 Morgan, W.B. and Pugh, J.C., West Africa. London: Methuen. 1969. Nigeria Yearbook 1971. Lagos: Daily Times Press, 1971. Pollitzer, R., Cholera. Geneva: World Health Organization, 1959. Prothero, R.M., Migrants and Malaria. London: Longmans, 1965. Road Maps of 'Greater Lagos'. Shomolu-Lagos: Cardinal Survey, 1973. Sada, P.O., Metropolitan Region of Lagos, Nigeria: A Study of the PoliticaT'Factor in'UFban Geography,TIndiana University, Department Ef'Geography, Ph D. dissertatifin, 1968. Schram, R., A History_of the Nigerian Health Services. Ibadan, 1971. Snow, J., Snow on Cholera (reprint of 2 papers by J. Snow, M.D.). New York: CommonweaTth Fund, 1936. Stamp, L.D., The Geography of Life and Death. Ithaca, New York: Cornell University Press:—1964. Weeklyngidemio1ggjcal Record (weekly notification of cholera cases). Geneva: maria HealthEOrganization, vols. 45-49, 1970-1974. Articles in jgurnals, books, and newspapers. Abdou, S., “The cholera epidemic in Egypt: mode of spread", Lancet, (November 8, 1947), pp. 696-698. Adeniyi, J.D., "Cholera control: problems of beliefs and attitudes", Int. J. Hlth. Education, vol. 15, no. 4 (1972), pp. 238-245. Allen-Price, E.D., "Uneven distribution of cancer in West Devon with reference to the divers water-supplies", Lancet, (June 4, 1960, pp. 1,235-l,238. Amedome, A., "Etude clinique et therapeutique du cholera au Centre Hospitalier Universitaire de Lome (A propos de 150 cas bacteriologique", Méflecine d'Afrique Noire, vol. 20, nos. 8/9 (1973), pp. 655-656. Amegee, E., Nenonene, J., Kloutse, 1., Nabede, A. and Amedome, A., "Mesures d'hygiene du milieu au cours de l'épidémie de cholera au Togo", Medecine d'Afrique Noire, vol. 20, nos. 8/9 (1973), pp. 64 9-654 0 239 Amu, V.B.G., "Experience with cholera epidemic in Mid-Western State of Nigeria", J. Soc. Hlth. Nigeria, vol. 6, no. 3 (1971), pp. 07- O9. Araoz, J. de and Subrahmanyam, D.V., "Environmental sanitation in cholera", in Principles and Practice of Cholera Control. Geneva: World Health Organization, 1970, pp. 95-110. Artus, J.C. and Carteron, 8., Données bacteriologiques sur la cholera "E1 Tor" (serotype Inaba) dans le Territoire Francais des Afars et des Issas“, Medecine Tropicale, vol. 33, no. 3 (1973), pp. 249-256. Aubry, P., Flavigny, C., Rouhier, D. and Giraud, J.,,"Une experience du cholera Africain. A propos de 395 cas observes dans deux centres de traitement du Sud-Dahomey“, Bull. Soc. Pathol. Exotique, vol. 65, no. 3 (1972), pp. 351-356. Aubry, P3, Giraud, J., Hazoume, F., Massacrier, A., and Pignol, F., "Experience du centre de traitement du cholera de Cotonou (Dahomey)", Medecine d'Afrigue Noire, vol. 20, nos. 8/9 (1973). pp. 659-661. Azurin, J.C. et. a1. (9 others), "A long-term carrier of cholera - Cholera Dolores", Bull. W.H.O. vol. 37 (1967), pp. 745-750. Banjera, A.C., "Note on cholera in the United Provinces (Uttar Pradesh)", Indian J. Med. Res., vol. 39, no. 1 (1951), pp. 17-40. Bart, K.J., Huq, 2., Khan, M., et. al., "Sero-epidemiologic studies during a simultaneous epidemic of infection with E1 Tor O awa and classical Inaba Vibrio cholerae", J. Infect. Dis., vol. 121 (1970), pp. 517- $24. Bart, K.J., Khan, M. and Mosley, W.H., "Isolation of vibrio cholerae from nightsoil during epidemics of classical and El Tor choTera in East Pakistan", Bull. W.H.0., vol. 43, no. 5 (1970), pp. 421-429. Barua, D., “Survival of cholera vibrios in food, water and fomites", in Principles and Practice of Cholera Control. Geneva: World Health Organization, 1970, pp. 29-31. ., "Cholera vaccination as a tool for cholera control", Bull. Soc. Pathol. Exotique, vol. 64 (1971). pp. 652-659. ., "The global epidemiology of cholera in recent years", Proc. Royal Soc. Med., vol. 65, no. 5 (May, 1972), pp. 423-428. Barua, D. and Cvjetanovic, B., "Cholera during the period 1961-1970", in Principles and Practice of Cholera Control. Geneva: World Health Organizationjfll970, pp. 15-22. 240 Baylet, R. and Diop, S., "Enseignement d' une "epidemie" de cholera au Senegal en zone Sahelo-Soudanienne“, Bull. Soc. Pathol. Exotique, vol. 66, no. 1 (1973), pp. 54-65. Benenson, A.S., Ahmed, 5.2. and Oseasohn, R.O., ”Person-to-person transmission of cholera", in Proceedings of the Cholera Research Sgggosium, P.H.S. Pub. 1328, U25. Govt. Printing OffiCe, Washtington, Ben Rachid, M. S., Stroobant, A. Ben Salem,M . ,Ben Jaafar, C., and Essafi, M., ”Etude epidemiologique du foyer de cholera de Keliéba (Tunisie)", Arch. Inst. Pasteur Tunis, vol. 48, no. 3 (1971), pp. 255-291. ‘T' Bhattacharji, L.M. and Bose, 8., “Field and laboratory studies on the transmission of the V. cholerae in the maintenance of cholera endemicity: A prelimifiary report", Indian J. Med. Res., vol. 52, no. 8 (1964), pp. 777-783. Bhattacharji, L.M., Choudry, K. and Bose, D.K., "Changing phases of cholera in India", Indian J. Med. Res., vol. 52 (1964), pp. 751-759. Bono, 0., Coulm, J., Coulanges, 8., Sirol,J ., Tachon, J. and Félix, H., “Installation du cholera aux alentours du Lac Tchad", Bull. Soc. Pathol. Exotigue, vol. 64, no. 4 (1971A), pp. 387- 398. Bono, 0., Coulm, J., Coulanges, B. ., Sirol, J. and Tachon, J., "Installation du cholera aux alentours du lac Tchad", La Presse Médicale, vol. 79, no. 54 (19718), pp. 2, 483- 2, 484. Bourgeade,A ., Dobeschu, A., Duchassin, M, Kadio, A. and Naim, C., "Etude clinique du chole’ra", Medecine d' Afrique Noire, vol. 20, no. 3 (1973), pp. 189-197. Bourgeade, A, Duchassin, M., Kadio, A. and Tiendebreogo, H., "Traitement antibiotique du cholera par la Sulfadoxine", Medecine d' Afrique Noire, vol. 19 (1972), pp. 93- 98. Bourgeade, A. Rive, J., Duchassin, M., Koffi, E. and Seka, A., "L' Epidemiologique du cholera et ses problemes", Medecine d' Afrique Noire, vol. 20, no. 3 (1973), pp. 177-187. Bouvry, M., Bourgeade, A, Potet, F., Groussard, O. and Loukiere, R., "L' estomac des choleriques: etude anatomique et functionnelle", Medecine d' Afrique Noire, vol. 20. no. 3 (1973), pp. 231-238. Brownlea, A.A., "Modelling the geographic epidemiology of infectious hepatitis", in N.D. McGlashan (ed.), Medical Geography, Techniques and Field Studies. London: Methuen and co.,11973, pp. 279-300. 241 Carpenter, C.C.J., “Pathogenesis and pathophysiology of cholera“. in Principles and Practice of Cholera Control. Geneva: World Health Organization, 1970, pp. 53-56. Carteron, 8.," Le cholera dans la Territoire Francais des Afars et des Issas: evolution generale et donne’es bio-geographiques", Me’decine Tropicale, vol. 33, no. 3 (1973), pp. 229- 234. Carteron, B. and Artus, J.C., "Epidemiologie du cholera au Territoire Francais des Afars et des Issas", Médecine Tropicale, vol. 33, no. 3 (1973), pp. 235-248. Carvalho, A., Baylet, R. ., Djop, 5., Kane, 1, Dia, M. C. and Diop, 0., ”L' epidemie de cholera a Podor et Matam" , Bull. Soc. Pathol. Exotique, vol. 17, no. 4 (1972), pp. 655- 661. Castets,M .. Sarrat. H. and Mora, M., "Donnees bacteriologiques sur les souches de Vibrion cholérique isoiees au Senegal", Bull. Soc. Med. Afrique Noire Langue Frangaise, vol. 17, no. 4(1972). pp. 631-635. Causse, 6., Le Viguelloux, J. and Welckel,J ., "Remarques sur 1' organization practique de la lutte contre le cholera en zone rurale dépouvre de medecin", Bull. Soc. Pathol. Exotique vol. 64, no. 5 (1971), pp. 659- 662. Chen, P.C., "Socio-cultural aspects of a cholera epidemic in Trengganu, Malaysia", Trop. Geog. Med., vol. 23 (1971), pp. 296-303. "Cholera and Politics", West Africa, no. 2,881, September 8, 1972, p. 1.175. "Cholera in 1970", Weekly Epidemiological Record, vol. 46 (August 6, 1971), pp 0 325-328 '0 "Cholera in 1971", Weekly Epidemiological Record, vol. 47 (1972), pp. 281- 283. "Cholera in 1972", Week1y_Epidemiological Record, vol. 48 (1973). pp. 297- 299. "Cholera in 1973", Weekly Epidemiological Record, vol. 48 (1973), pp. 281- 282. "Cholera - Sri Lanka“, Weekly Epidemiolog_cal Record, vol. 49 (1974), pp. 173-176. Chuttani, c. s., Murty, D. K., P01, 5. c.. Bhatia, G. 5.. Shrivastau, a. B. and Pandit, C. G., “Epidemiological and clinical characteristics of cholera in Delhi 1966", Indian J. Med. Res. ., vol. 55 (l967A), pp. 815- 824. ,"Epidemiological significance of the carrier state in E1 Tor infection", Indian J. Med. Res. ., vol. 55, no. 8 (19678), pp. 825- 832. 242 Cohen, J., Klasmer, R., Ghalayini. H., Schwartz, 1., Pridan, D. and Davies, A.M., "Epidemiological aspects of cholera in a non- endemic area“, Lancet, July 10, 1971, pp. 86-89. Comité’Central de Lutte contre le Choléra, "L'epidémie de cholera au Dahomey", Medecine Tropicale, vol. 31, no. 6 (1971). pp. 643-663. Coulanges, B. and Coulanges, P., "L'Epidémie de choléra du sultanat de Goulfey (Nord-Cameroun: Mai-Juin l97l)", Méflecine Tropicale, vol. 31, no. 6 (1971), pp. 635-642. .. "A propos de 1'épidémie de cholera du sultanat de Goulfey (Nord Cameroun, Mai 1971). Considerations sur 1'epidémiologie et la prophylaxe", Bull. Soc. Pathol. Exotique, vol. 65, no. 2 (1972), pp. 2 7-230. - _ Cox, P.S.V., Cholera in Northern Kenya. The use of prophylactic treatment in an isolated community“, E. Afr. Med. J., vol. 49, no. 6 (1972), pp. 440-447. Cvetanovic, 8., "Cholera as an international health problem", in Principles and Practice of Cholera Control. Geneva: World Health Organizatibn, 1970, pp. 9-14. . Cvjetanovic, B. and Barva, 0., “The seventh pandemic of cholera“, Nature, vol. 239 (September 15, 1972), pp. 137-138. Dada, B.A.A., "Prevention and control of cholera in Lagos State“, J. Soc. Hlth. Nigeria. vol. 6, no. 3 (1971A), pp. 130-133. ., "First sixteen weeks of cholera in Lagos State“, J. Soc. Hlth. ngeria. vol. 6, no. 3 (19713), pp. 133-138. ., “Cholera in Lagos State: 26/12/70 - 2/9/729. Lagos: Lagos State Government, Ministry of Health and Social Welfare, Medical Statistics Division, 1972, l p. (mimeo.). D'Almeida, J.J., Kekeh, J.D., Mikem, A., Nabede, A. and Amedome, A., "L'épidemie de choléra au Togo", Medecine d'Afrique Noire, vol. 20, nos 8/9 (1973), pp. 639-644. De Maeyer - Cleempoel, S., "Quelques réflexions a propos du cholera et de la vaccination anticholérique a l'issue du symposium organise par l'OMS a Bamako (Mali) en Mars, 1972", Arch. Belg. Med. Soc., vol. 30, no. 6, (1972), pp. 380-393. DeMoor, C.E., PParacholera (El Tor)", Bull. W.H.O., vol. 2 (1949), pp. 5-17. 243 Desmoulins, G., Charpin, M., Yekpe, M. and Piacentini, M., "Etude sur le portage asymptomatique dans deux villages du Sud Dahomey infectes par le cholera", Medecine Tropicale, vol. 31, no. 6-(1971). pp. 665- 672. T Diop, I., Sow, A., Cousin, B. and Agbestra, M., "Cholera et diarhees choleriformes", Bull. Soc. Med. Afrique Noire Langue Francaise, vol. 17, no. 4 (1972), pp. 636-651. . Diop, 5., and Baylet, R., "La survie du vibrion El Tor dans certaines eaux de boissons“, Bull. Soc. Med. Afrique Noire Eangue Francaise, vol. 17, no. 4 (1972), pp. 652-654. Diop, S., Gaye, P., Baylet, R., Sy, M. and Toure, M., "Considerations cliniques et epidemiologiques a propos de l'epidemie de cholera dans 1e departement de Bakel (Senegal)", Bull. Soc. Med. Afrique Noire Langue Francaise, vol. 17, no. 4 (1972), pp. 670-674. Dizon, J.J. et. al. (15 others), "Studies on cholera carriers", Bull. World Health Organization, vol. 37 (1967), pp. 737-744. Dodin, A. and Felix, H., "Du role de la sueur dans l'epidemiologie du cholera en pays sec", Bull. Acad. Natn. Med., vol. 156, nos. 26/27 Dorolle, P., "Surveillance epidemiologique du cholera a l'echelon inter- national", Medecine Tropicale, vol. 31, (1971), pp. 149-156. ., "International problems of communicable-disease control", Lancet, September 9, 1972, pp. 525-527. Duchassin, M., Dodin, A., Bourgeade, A. and Rive, J., "Le vibrion cholerique. Diagnostic bacteriologique. (Bilan de 22 mois d'activite du laboratoire du cholera)“, Medecine d'Afrique Noire, vol. 20, no. 3 (1973), pp. 165-174. ‘ ‘ Dutertre, J., Huet, M., Gateef, C. and Durand, 8., "Le cholera au Cameroun", Medecine Tropicale, vol. 32, no. 5 (1972), pp. 607-624. Favreau, P., "L'evolution du cholera type E1 Thor en Afrique Noire", Lg_ Press Medicale, vol. 79 (1971), p. 553. Felix, H., "Epidemie de cholera en Afrique. Note d'information sur l'evolution entre Aout et Decembre, 1970”, La Presse Medicale, vol. 79, no. 11 (1971), pp. 475-477. ., "Epidemie de cholera en Afrique (suite). Evolution de la situation pen ant 1e premier semestre, 1971", La Presse Medicale, vol. 79, no. 41 (19718), pp. 1,801-l,804. 244 Felix, M., "Le developpement de l'epidemie de cholera en Afrique de 1' Ouest", Bull. Soc. Pathol. Exotique, vol. 64, no. 4 (1971C), pp. 561-586. ‘ "Le cholera Africain", Medecine Tropicale, vol. 31, no. 6 (19 710), pp. 619- 628. Felsenfeld, 0., "Some observations on the cholera E1 Tor epidemic in 1961-62", Bull. W.H.O., vol. 28 (1963), pp. 289-296. , "Cholera research, 1966", Bulletin Endemic Disease, vol. 8, "“fiB§'i/4 (1966), pp. 1-28. "Fighting the cholera crisis", West Africa, January 9-15, 1971, p. 33. Forbes, G.I. ,Lockhart, J. D. F. and Bowman, R. K. , "Cholera and hi htsoil infection in Hong Kong, 1966“, Bull. W. H. O. ., vol. 36, no. 3 11967), pp 0 367-373 0 Forbes, G. I. ,Lockhart, J. D. F. Robertson, M. J. and Allan, W. G. L., “Cholera case investigation and the detection and treatment of cholera carriers in Hong Kong", Bull. W. H. O. , vol. 39, no. 3 (1968), pp. 381-388. . Francis, T.I., "Effect of chemotherapy on the duration of diarrhoea, and on vibrio excretion b cholera patients", J. Trop. Med. Hygiene, vol. 74 (August, 1971 , pp. 172-176. Gallut, J., "Actualite’du cholera. Evolution des problémes epidemiologiques et bacteriologiques“, Bull. Inst. Pasteur, vol. 66, no. 2 (1968), pp. 219-248. ., "La septiéme pandemie cholerique. 1961-1966, 1970...", Bull. Soc. Pathol. Exoti ue, vol. 64, no. 4 (1971), pp. 551-560. Gilbert, E. W., “Pioneer maps of health and disease in England", Geog. J. vol. 124 (1958), pp. 172-183. Great Britain Medical Research Council, “Cholera in Northern Nigeria“, Medical Research Council Annual Report, April, 1972 - March, 1973. Haggett, P., "Contagious processes in a planar graph", in N.D. McGlashan (ed.), Medical Geography,_Techniques and Field Studies. London: Methuen ana'co.,—1973, pp. 307-324. Hugh, R., “Comparison of classical and E1 Tor cholera", in Proceedings of the Cholera Research Symposium. Washington: U.S. Dept.‘HfEJWTT PGbTic HealthTService—Pfiblication no. 1,328, 1965. Humponu-Wasu, 0.0., “Epidemiological aspects of an E1 Tor cholera outbreak in Kaduna, Nigeria", Trop. Geog. Med. ., vol. 25 (1973), pp. 277-281. 245 Hunter, J.M., "River blindness in Nangodi, Northern Ghana" A hypothesis of cyclical advance and retreat”, The Geographical Review, vol. 56 (1966), pp. 398-416. ., "0n the Merits of holism in understanding societal health needs", The Centennial Review, vol. 17, no. 1 (1973), pp. 1-19. Joint ICMR - GWB - WHO Cholera Study Group, Calcutta, India, “Cholera carrier studies in Calcutta, 1968", Bull. W.H.O., vol. 43 (1970A), pp. 379-387. Study on Vibrio cholerae infection in a small community in Calc utta", Bull. W. H. 0., vol. 43 (19708), pp. 401-406. . "Serological studies on cholera patients and their household contacts in Calcutta in 1968", Bull. W.H.O., vol. 43 (l97OC), pp. 389-399. Joseph, P.R., Tamayo, J.F., Mosley, W.H., Alvero, M.G. Dizon. J.J. and Henderson, 0. A. "Studies of cholera E1 Tor in the Philippines: 2. A retrospective investigation of an explosive outbreak in Bacolod City and Talisay, November, 1961", Bull. W. H. 0., vol. 33 (1965), pp. 637-643. Kamal, A.M., "Endemicity and epidemicity of cholera", Bull. W.H.O., vol. 28 (1963), pp. 277-287. "Keeping cholera at bay: Army helps to clean Kaduna", New Nigeriap, Kaduna, no. 1,592 (March 5, 1971), p Khan, A.Q., "Role of carriers in the intrafamilial spread of cholera", Lancet, February 4, 1967, pp. 245-246. Khan, M., Bart, K.J. and Haq, Z., "The changing pattern of cholera in East Pakistan: the appearance of E1 Tor Vibrio cholerae", J. Pakistan Med. Assoc.,_vol. 20, no. 2 (1970), pp. 43-48. La Peyssonnie, L., ”Acquisitions recentes en matiere d' epidemiologie et de prophylaxie du cholera en Afrique“, Bull. Soc. Pathol. Exotigue, vol. 64, no. 4 (1971A), pp. 644-652. ., "Chimioprophylaxie de 1' infection cholerique: intérét, espoirs et limites", Medecine Tropicale, vol. 31 (19718), pp. 127-132. ., “Chimioprophylaxie de 1' infection chole’rique. II Modalites et c1rconstances d' emploi“, Medecine Tropicale, vol. 31. no. 6(19710), pp. 719- 722. 246 La Peyssonnie, L., "Le cholera, An 11", Medecine Tropicale, vol. 31, no. 6 (1971B), pp. 615-618. ., "Le cholera: pourquoi?", Medecine Tropicale, vol. 31, no. 1 (1971E), pp. 9-14. Le Viguelloux, J. and Causse, G., "Réflexions sur l'epidemiologie du cholera en Afrique Occidentale", Medecine Tropicale, vol. 31, no. 6 (1971), pp. 677-684. Le Viguelloux, J. and Doury, J.C., "Epidemiologie du cholera moderne", Medecine Tropicale, vol. 31, no. 1 (1971), pp. 17-29. Lewis, E.A. et a1, “Cholera in Ibadan", Amer. J. Trop. Med. Hygiene, vol. 21 (1972), pp. 307-314. Mackenzie, D.J.M., "Cholera - Its nature, management and prevention", S. Afr. Med. J., vol. 45 (1971), pp. 1-7. Martin, A.R., Mosley, W.H., Sau, 8.8., Ahmed, S. and Huq, 1., "Epidemiological analysis of endemic cholera in urban East Pakistan, 1964-1966", Amer.g1TEpidemio19gy, vol. 89, no. 5 (1969), pp. 572- 582. May, J.M., "Medical Geography", in P.E. James and C.F. Jones (eds.), American Geograppy_1nventory and Prospect. Association of American Geographers, 1954, pp. 453-368. McBean, A.M., Agle, A.N.. Compaore, P., Foster, S.0. and McCormack, W.M., "Comparison of intradermal and subcutaneous routes of cholera vaccine administration", Lancet, March 4, 1972, pp. 527-529. McCormack, W.M. Islam, M.S. Fahimuddin, M. et. al., "A community study of inapparent cholera infections”, Amer. J. Epidemiology, vol. 89 (1969), pp. 658-664., McCormack, W.M., Mosley, W.H., Fahmidden, M. and Benenson, A.S., "Endemic cholera in rural East Pakistan", Amer. J. Epidemiology, vol. 89, no. 4 (1969), pp. 393-404. Mered, B. et. al., "L'épidémie de choléra en Algerie en 1971", Arch. Inst. Pasteur Alger, vol. 49 (1971), pp. 85-97. Mngola, E.N., “Cholera epidemic in Kenya", E. Afr. Med. J., vol. 49, n0. 2 (1972), pp. 151-1580 Modal, A. and Sack, R.B., "The clinical picture of cholera", in Principles of Cholera Control. Geneva: World Health Organization, 1970, pp. 57- 60. 247 Mosley, W.H., Alvero, M.G., Joseph, P.R., Tamayo, J.F., Gomes, C., Montague, T., Dizon, J.J. and Henderson, D.A., "Studies of cholera E1 Tor in the Philippines: 4. Transmission of infection among neighbourhood and community contacts of cholera patients", Bull. W.H.O., v01. 33 (1965), pp. 651-660. Mosley, W.H., Aziz, K.M.A., Minzanur Rahman, A.S.M., Chodhury, A.K.M.A., Ahmed, A. and Fahimuddin, M., "Report of the 1966-67 cholera vaccine trial in East Pakistan", Bull. W.H.O., vol. 47 (1972), pp. 229-238. Mosley, W.H., Bart, K.J. and Sommer, A., "An epidemiological assessment of cholera control programs in rural East Pakistan“, Int. J. Epidemiology, 1972, pp. 5-11. Mosley, W.H., Benenson, A.S. and Barui, R., “A serological survey of cholera antibodies in rural East Pakistan", Bull. W.H.O., vol. 38 (1968), pp. 327-334. Mukerjee, 5., "Recent incidence of cholera outside India", Indian J. Med. 332., vol. 52 (1964), pp. 771-776. ., "A decade's tracking of cholera with bacteriophage", Indian J. Med. Res., vol. 55, no. 4 (1967), pp. 308-313. Mukerjee, S. and Basu, 5., "Cholera E1 Tor in India: Effect on epidemiology of classical cholera", Trop. Geog. Med., vol. 19, no. 2 (1967) pp. 138-143. Mukerjee, S., Basu, S. and Bhattachary, P., "A new trend in cholera epidemiology", Brit. Med. J., vol. 2 (1965), pp. 837-839. Neogy, K.N., Mukerjee, M.,. and Manji, P., “Interplay of Vibrio cholerae strains in Calcutta", Lancet, January 18,,1969, pp. 159-160. Ogunleye, Femi, "Filth" A health hazard in Kano", Daily Times, Lagos, no. 20, 381 (September 8, 1973), p. 7. Pal, S.C., Murty, D.K., Murti, G.V.S., Misra, 8.5., Chuttani, C.S., Pandit, C.G. and Shrivastav, J.D., "Bacteriological investigations of cholera epidemics in Gurgaon District and in Delhi during 1965-66", Indian J. Med. Res., vol. 55, no. 8 (1967), pp. 810-814. Pandit, C.G., Pal, S.C., Munti, G.V.S., Misra, 8.5., Murty, D.K. and Shrivastav, J.8., “Survival of vibrio cholerae biotype El Tor in well water", Bull. W.H.O., vol. 37, no. 4 (1967), pp. 681-685. Payet, M., "Epidémiologie du cholera. Son aspect actuel. Ses possibilités d'avenir", Bull. Soc. Pathol. Exotiqgo, vol. 64, no. 4 (1971), pp. 580- 582. 248 Pearpoint, J., "Cholera crisis reviewed", The CUSO Forum, vol. 2, no. 3 (April-May, 1974), p. 24. Philippines Cholera Committee, "Study on the transmission of El Tor cholera during an outbreak in Can-Itom community in the Philippines", Bull. W.H.O., vol. 43, no. 3 (1970), pp. 413-419. Pobee, J.O., Grant, F.C. and Salles, C.A., “Case report on cholera", Ghana Med. J., vol. 9, no. 9 (1970), pp. 306-309. "Positive cholera cases and deaths in Lagos State“. Lagos: Lagos State Government, Ministry of Health and Social Welfare, Medical Statistics Division, 1971. (mimeo.). Pyle, G.F., "The diffusion of cholera in the United States in the Nineteenth Century", Geographical Analysis, vol. 1, no.1 (1969), pp. 59-75. Raska, K., "Surveillance and control of cholera", in Principles and Practice of Cholera Control. Geneva: World Health Organization, 1970, pp. 111-122. Roberts, J.M.D., "Cholera detection and surveillance in Kenya", E. Afr. Med. 2,, vol. 49, no. 2 (1972), pp. 159-162. Saba, A.L., Chakrakorty, A.K. and Chowdhury, K., "Measurement of variations in epidemic prevalence of cholera in India", J. Indian Med. Assoc., vol. 54, no. 1 (1970), pp. 12-20. Sakazaki, R., "Classification and characteristics of vibrios", in Principles and Practice of Cholera Control. Geneva: World Health OrganizatTon, 1970, pp. 33-38. Salles, C.A., Trikedi, 8.8. and Markell, E.C., "Cholera agglutinins in Ghana", Bull. Soc. Pathol. Exotique, vol. 65, no. 3 (1972), pp. 356- 364. Schram, R., "The 1971 cholera epidemic in Zaria, Nigeria", Savanna, vol. 1, no. 2 (1972), pp. 213-222. Sehgal, P.N. and Pandit, C.G., "Behaviour of an epidemic of Vibrio cholerae E1 Tor in Gurgaon District, Haranya State", Indian J. Med. Res., vol. 56, no. 7 (1968), pp. 982-1,001. Sen, R., Sen, D.K., Chakrabarty, A.M. and Ghosh, A., "Cholera carriers in a Calcutta slum", Lancet, November 9, 1968, pp. 1,012-1,014. Shousha, A.T., "Cholera epidemic in Egypt 1947", Bull. W.H.O., vol. 1 (1948), pp. 353-381. 249 Singha, P. and Pandit, C.G., “A statistical approach to the evaluation of public health measures in the control of cholera, with special reference to the cholera epidemic in the Gurgaon District, Haranya State“, Indian J. Med. Res., vol. 56 (1968), pp. 1,002-1,018. Sinha, R., Deb, B.C., 0e, S.P., Abou-Gareeb, A.H. and Shrivastava, D.L.. “Cholera carrier studies in Calcutta in 1966-67", Bull. W.H.O., vol. 37, no. 1 (1967), pp. 89-100. Sinha, R., Deb, 8.0., De, S.P., Sirkar, B.K., Abou-Gareeb, A.H. and Shrivastava, D.L., "Role of carriers in the epidemiolog of cholera in Calcutta", Indian J. Med. Res., vol. 56, no. 7 (1968 . Pp. 964- 978. Sirol, J., Félix, H., Delpy, P. and Bono, 0., "A propos de l'epidemie choléra de Fort-Lamy (Tchad) en Mai-Juin, 1971", Medecine Tropicale, vol. 31, no. 6 (1971), pp. 629-634. Sommer, A. and Woodward, W.E., "The influence of protected water supplies on the Spread of classical Inaba and El Tor/Ogawa cholera in rural East Bengal", Lancet, November 11, 1972, pp. 985-987. Swaroop, S., "Endemicity of cholera in India", Indian J. Med. Res., vol. 39, no. 2 (1951), pp. 141-184. Swaroop, S. and Raman, M.V., "Endemicity of cholera in relation to fairs and festivals in India", Indian J. Med. Res., vol. 39, no. 1 (1951), pp. 41-490 Swartz, T.A., "Recent Israeli experience with cholera", Proc. Royal Soc. Med., vol. 65 (May, 1972), pp. 428-432. Sy, A., Gangue, Y, Gaye, P., Baylet, R. and/Diop, S., "L'epidémie de cholera dans 1e département de Louga (region de Diourbel)", Bull. Soc. Med. Afrique Noire Langue Francaise, vol. 17, no. 4 (1972), pp. 662- 669. Tamayo, J.F., Mosley, W.H., et. a1. "Studies of cholera El Tor in the Philippines. 3. Transmission of infection among household contacts of cholera patients", Bull. W.H.O., vol. 33 (1965), pp. 645-649. "The seventh cholera pandemic", W.H.O. Chronicle. V01. 25 (1971). PP- 155-160. "Three die of cholera attack in Lagos District: First victims in Nigeria", Daily Time§,_Lagos, no. 19,459 (December 30, 1970), p. 1. ”Tip of the iceberg shows", Nature, vol. 227 (1970), p. 766. 250 "Trend of cholera in Egypt", Weekly Epidemiological Record, vol. 22 (1947), pp. 383-385. - “Trips t? Mecca held up", Daily Times, Lagos, no. 14,472 (January 14. 1971 . p. 3. ' ULCM Cholera Research Team: Departments of Community Health and Micro- biology, College of Medecine, University of Lagos, “Prevalence and behaviour of vibrio cholerae in a newly infected country (Nigeria)", J. Nigerian Med. Assoc., vol. 1, no. 3 (1971), pp. 143-144. Voelckel, J., "Assainissement et lutte anticholérique“, Medecine Tropicale, vol. 31 (1971), pp. 133-144. Voelckel, J. and Causse, G., "Apercus prophylactiques", Médecine Tropicale, vol. 31, no. 6 (1971), pp. 711-716. Wallace, C.K., Carpenter, C.G.J., Mitra, P.P., et. al., "Classical and E1 Tor Cholera: a clinical comparison", Brit. Med. J., vol. 2 (1966), pp. 447-449. Wallace, C.K., Falsie, A.E., Mangubat, E.V., Jrinio, C. and Phillips, R., "The 1961 cholera epidemic in Manila, Republic of the Philippines", Bull. W.H.O., vol. 30 (1964), pp. 795-810. ' Watanabe, Y. and Verwey, W.F., "Immunity in cholera", in Principles and Practice of Cholera Control. Geneva: World Health Organization, 1970, pp. 77-86} "Water shortage worsens in Lagos", Dai1y Times, Lagos, no. 19,460 (December 31, 1970), p. 5. Wilson, A.M., "The spread of cholera to and within Nigeria 1970-71”, J. Clin. Pathol., vol. 24 (November, 1971), p. 768. Wollf, H.L., "A quantitative approach to epidemiology and control of cholera", Bull. Soc. Pathol. Exotiqge, vol. 64, no. 4 (1971), pp. 582-589. Woodward, W.E. and Mosley, W.H., “The spectrum of cholera in rural Bangladesh. II Comparison of E1 Tor Ogawa and classical Inaba infection", Amer. J. gpjdemiology, vol. 96, no. 5 (1972), pp. 342-351. "1111111111111111111111“