|
|
||||||||
| ABSTRACT |
|
|
|---|
| INTRODUCTION |
|
|
|---|
Despite the severity of the problem, gaps remain concerning urban malaria transmission in SSA. Several explanations exist to account for why urban areas seem to have lower malaria transmission rates compared with rural areas. One explanation is that pollution affects larval habitats, the life cycle of mosquitoes, and vectorial capacity. Other explanations include mosquito avoidance behavior by urban populations, such as screens, doors, insecticides, and bed nets. Higher human population densities may reduce biting rates, owing to the higher ratio of humans to mosquitoes. Few data are available to confirm these theories or to prioritize strategies for malaria control. Many questions remain for public health specialists hoping to regain control over malaria in urban environments.
The objectives of this article are to synthesize existing literature on malaria transmission in urban areas of SSA and to generate a conceptual framework. Following a summary of the main terms, we present a meta-analysis of entomologic inoculation rates (EIR) from urban, periurban, and rural studies. This analysis provides background to a review of the entomologic and social and behavioral literature related to urban malaria. The resulting conceptual framework illustrates factors that may explain the heterogeneity of malaria transmission within and among urban environments in SSA.
| BACKGROUND |
|
|
|---|
The postWorld War II era has been characterized by a rapid increase in the worlds population, especially in tropical areas. In 2000, the population in SSA was estimated to be 784 million; by 2025, this figure is expected to be >1.2 billion.3 Despite intriguing accounts of Africas ancient cities, some with 40,000 inhabitants,1618 the continents population remained predominantly rural until recently. In 1900, <10% of Africans lived in an urban area, and by 1950, only 15% lived in an urban area. Today, almost half of the population in SSA lives in urban or periurban areas (45% in 1999), and this proportion is expected to increase dramatically over the next 25 years. Although rapid urban growth is a relatively recent phenomenon in SSA, the cities of approximately 0.5 million population are growing faster than in any other area of the world.19 This is an urbanization rate of 2% for the past 30 years, a doubling of the urban population every 37 years. Pollution, poor housing, lack of sanitation, unprotected water reservoirs, weak services, low productivity, and widespread economic disparity have accompanied this urbanization.19
Similar to most urban settlements, African cities are complex, dynamic structures. Western definitions emphasize characteristics that differentiate urban from rural areas, including land use patterns, increased density of households, differences in housing material, economic differentiation in relation to type and abundance of work, access to public transport, access to utility services, and access to social services. In many instances, urban SSA does not fit this conventional definition. Rice fields exist in the heart of Bouake,20 and market-garden wells are common in Dakar.21 Livestock often are herded through central business districts and are a common sight in residential, market, and commercial areas. Urban areas can encompass suburban development, affluent neighborhoods, older settlements that have become established shantytowns, business districts, and periurban slum settlements on the outskirts of towns and cities. People who settle in densely populated, undeveloped periurban areas often are met with communities already trying to overcome widespread poverty, pollution, and environmental degradation. These communities often lack the programs, personnel, and resources to combat infectious disease, a situation exacerbated by poverty, low levels of education, deteriorating infrastructure, and continued rural subsistence activity.
It is thought that about 10 anopheline species are responsible for malaria transmission in SSA. By contrast, in the urban context, there are normally 3: Anopheles gambiae, Anopheles arabiensis (both belonging to the An. gambiae complex), and Anopheles funestus. The first 2 species are the most important. An. funestus has not been found in most cities but was found in a district of Franceville, Gabon, and in Maputo, Mozambique.22,23
The value of the EIR lies in the fact that it provides an estimate of the passage of malaria parasites from infective anopheline species to human populations. It is calculated as the product of the human-biting rate (an estimation of the density of mosquitoes per person) and sporozoite index (an estimation of the proportion of vectors with sporozoites in their salivary glands). The EIR is expressed as the number of infective bites per person per year. It ranges from <1 bite per year in low-transmission areas to 2,979 infected bites per person per year in 1 sentinel house.24 Although the EIR is a good indicator of transmission intensity in the context of high mosquito density, logistical problems exist when estimating EIR in areas of low mosquito density.25 The determination of anopheline density at low densities can be difficult, and malaria transmission does occur below the entomologic threshold for detection.
The EIR cannot be considered an exact measure of transmission because not all bites from infected anopheline species succeed in infecting humans. In experiments, Rickman et al,26 using persons without any antimalarial immunity and anopheline species with P. falciparum sporozoites in their salivary glands, showed that 1 or 2 infected bites per person infected only 5 of 10 human volunteers. This limiting factor, which is called the infectivity success rate (usually parameter "b" in literature), rarely is taken into account when calculating an EIR from the sporozoite index and human-biting rate. In endemic areas, the infectivity success rate ranges from 5% to 26%, with inverse variations with EIR resulting from higher protective immunity, more multiple infections, and longer episodes with parasitaemia.24,27
| MATERIALS AND METHODS |
|
|
|---|
The EIRs reviewed were assigned a category: central urban, periurban, and rural. Categorical assignment of the respective EIRs was based on the description of the study area in the articles. Assignment in the rural category never posed a problem, whereas differentiating between central and periurban categories posed a challenge. Maps and habitat descriptions were used to differentiate between central urban and periurban when ambiguity was evident. Areas adjoining marshlands, areas adjoining rice fields, market-garden areas, or areas close to a lake were considered periurban; some central quarters of Pikine,11,29 Bouake,20,30 and Ouagadougou31 were classified as periurban. Overall, 39 studies satisfied the inclusion criteria. Because several studies provided estimated EIRs from >1 site, the final data set included 90 values of EIRs distributed across 21 urban centers, 14 periurban sites, and 55 rural sites. In a separate analysis, we categorized study areas according to their ecologic niche with 33 values of EIRs in dry savannas plus Sahel versus 57 values in wet savannas plus forest zones.
| RESULTS |
|
|
|---|
|
|
Variability in EIRs among and within cities is evident (Table 1
). Studies in the urban centers of Cotonou, Kinshasa, and Edea report the highest EIRs.3537 The highest EIR reported for any urban area is 30/year; the lowest is 0. The univariate shape of the distribution in central urban areas is skewed toward the lower end, with more than two thirds of the studies reporting EIRs <4/year. The variation across the periurban areas is more diverse, ranging from 0.4/year in Dakar29 to 126/year in Bouake, Ivory Coast.30
Important interquarter or interdistrict variations can be observed. The studies conducted in 3 areas of central Bobo-Dioulasso estimated EIRs at 0.1, 0.5, and 2.38,39 In Edea, this figure was 4/year and 30/year across 2 distinct quarters.37 In 1 quarter in Pikine, large differences were detected for parasitologic incidence in children who lived close to marshland: about 1 infection per year for children residing at a distance of 10 to 160 m and 1 infection every 4 or 5 years for children residing at a distance of 785 to 910 m.4 In Pikine, during the dry season, 93% of the An. arabiensis mosquitoes were collected in dwellings situated <285 m from the marshland.29 In Ouagadougou, Burkina Faso, at a distance of 415 m from a lake, the density of An. gambiae s.l. per room decreased by 67% compared with the density observed closer to the lake.40 Comparable results were obtained in the suburb of Maputo, Mozambique.41
The heterogeneity of malaria transmission in urban areas results in large differences in malaria prevalence in humans living in different parts of the cities. In Brazzaville, the prevalence among schoolchildren varied from 3% in a central quarter to 81% in peripheral areas, and these differences were reflected in annual EIRs that varied from 1 to 100.13 The impact of urbanization is more marked in areas where the mean rainfall is low and seasonal (Table 2
). The mean EIRs are 0.96 in urban centers of cities located in dry savannas and Sahel and 12.62 in urban centers of towns located in wet savannas and forest zones (P < 0.0009 by Mann-Whitney U test). In periurban areas, these values are 14.67 and 77 (P < 0.0027). In rural areas, they are 94.03 and 197.98 and are not significantly different. When comparing the mean EIR of urban centers with the mean EIR of rural areas, a 98-fold decrease in EIR is seen in dry savannas and Sahel, and a 16-fold decreaseis seen in wet savannas and forest zones.
|
| DISCUSSION |
|
|
|---|
Figure 2
presents a summary of the relationships across the major components of urban malarial ecology. The outcome is illustrated on the right as increases or decreases in urban malaria transmission. The central boxes reflect the entomologic factors that influence the level, timing, and intensity of vector competence. The boxes to the left represent an array of socioeconomic human factors and factors related to the physical environment thought to influence mosquito habitats. Human and vector components are influenced by climate and topographic variations, as indicated along the top of the figure. This framework was designed to facilitate the understanding of relationships that influence malaria transmission in urban areas across SSA.
|
There is some evidence that anopheline species may be adapting to urban ecosystems. Chinery42,43 observed some adaptation of An. gambiae s.s. to urban aquatic habitats, such as water-filled domestic containers and polluted water habitats created as a result of urbanization in Accra, Ghana. In a recently urbanized area of Kenya, Khaemba et al44 concluded that An. gambiae showed a strong preference for man-made, temporary aquatic sites over permanent aquatic habitats in the rainy season, although dams and swamps remained the preferred sites during the dry season.44
Although some adaptation may be occurring, many observers assume that the pollution of aquatic sites most likely has a negative effect on anopheline species density and longevity.45 Robert et al21 noted that larval populations of An. arabiensis in market-garden wells in Dakar are regulated by interacting environmental variables, such as high nitrate concentrations, low pH, murky water, and other factors. Other studies indicated that detergents are important pollutants that may inhibit the development of anopheline larvae.13 Barbazan et al45 concluded that pollution was a major factor responsible for the scarcity of larval sites in Maroua, a large city in the savanna region of Cameroon.
Another explanation for the low EIRs relates to high human population density, which provides easier access to human blood meals for mosquitoes. This increase in potential hosts is thought to reduce the chance of any single host receiving an infective bite, lowering the overall human-biting rate. Sabatinelli et al40 found that An. gambiae tend to bite near their breeding places in densely settled environments, however, and that their dispersion is restricted. An alternative hypothesis for low EIR argues that rapid urbanization in periurban areas and increases in population density in and around anopheline aquatic habitats increase the risk of malaria transmission. In other words, a large number of people living in close proximity to mosquito habitat increase blood meals available, which increases the number of mosquitoes an area can support, resulting in a greater number of total infective bites.46
Vector biting habits in urban contexts also may help account for some of the measurement challenges. An. gambiae s.s. and An. arabiensis feed on humans inside and outside houses. Preferential biting times may vary across and within small areas. In areas where humans typically stay outside later in the evenings, the protective effect of using bed nets or protection inside their houses is nullified if they are exposed during evening hours outside the home. Given this fact, the low EIRs calculated for urban areas may reflect the fact that finding female mosquitoes that have fed on a human outside as opposed to finding females resting inside houses may reflect challenges to existing entomologic field methods.
Human migration patterns also may explain variation in malaria transmission patterns among and within urban locations.18,47 For many urban residents, travel to rural areas on a regular basis is predictable and regular. Depending on immunity, people can be categorized as either active transmitters or passive acquirers.48 Passive acquirers are exposed to the parasite while traveling to an endemic area, whereas active transmitters harbor the parasite and increase the transmission risk when they travel to an area with an efficient vector. We can predict that when the relatively nonimmune individuals from urban areas travel to rural endemic zones, they may acquire infection and become transmitters when back home. Highland cities may be at risk of becoming malarious through the action of the rural migrants bringing infection with them.49,50
Municipal initiatives. Heterogeneity in malaria transmission may be explained by minor variations in the quality or type of water and waste management. In many urban contexts, the central business district is the only area with working water and sewage systems. It also may be the most sparsely inhabited area, especially at night during peak biting hours. From these central districts, the urban development is often characterized by growth in approximate concentric circles, with the newest construction in more marginal areas. These marginal areas can include swamplands or steep denuded hillsides. Generally, malaria transmission seems to decline from these peripheral areas toward the center. There are, however, several scenarios that may explain unexpected variations in this general trend. In Niamey, Niger,51 and Karthoum, Sudan,52 the city centers are located along rivers, which provides generous larval habitats for anopheline species.
Individual and household factors. In general, poorer populations are at greater risk of vector contact and infection, owing to physical proximity to water sources and lowered capacity (lack of education and resources) to use health care services and preventive measures to protect against malaria.53 Human-vector contact is influenced by housing type (e.g., number of screens, doors), housing and roofing material, and house location (gradient, surrounding drainage, and cleanliness of immediate environment).53 The use of screens, insecticides, prophylaxis, and bed nets, which is a function of income and education, also can affect the life expectancy of the mosquito, although whether this operates to influence vectorial capacity is unclear. On the one hand, shorter survival times of mosquitoes (reflected by the proportion of mosquitoes that had already laid eggs) were found in cities such as Pikine,11,29 Kinshasa,36 Edea,37 and Bobo-Dioulasso,38 in which the parity rate was significantly lower than in the neighboring rural areas. On the other hand, in other cities, including Brazzaville,13 Franceville,22 and Yaounde,54,55 the parity rate was similar between cities and their neighboring rural areas.
Malaria surveys were conducted during the 1930s in the port cities of Mombassa, Dar es Salaam, and Tanga.79 The authors of these studies believed that the problem of malaria in urban areas could be solved. The fact that it has not been solved has become a matter of concern.2,5658 The lower level of transmission, shown through this meta-analysis in urban areas, has important parasitologic, immunologic, clinical, and control implications. With urbanization, cities become populated with individuals who reach adulthood without significant malaria immunity. An increasing proportion of adults have little immunity. Given the rapid urbanization in recent years, such a situation is new for large segments of the population. This transition from a low endemic situation to a potentially epidemic one may be of great public health significance. This situation of islands of low immunity in an ocean of high endemicity is alarming, especially given the fragile ecologic state of the periurban areas and the considerable and regular migration patterns between these islands and their hinterlands.
Our analysis shows that these populations with low immunity do not constitute a fragile situation per se and that most cities rarely experience severe, widespread malarial epidemics. This situation is due, for the most part, to the major impact that urbanization (e.g., pollution, human density, increased built environment, protective and treatment measures taken by urban inhabitants) has on the larval ecology of the anopheline mosquitoes. Sudden changes in weather patterns (e.g., El niño) or civil unrest could affect the validity of these respective conclusions.
We propose that the city may be one of the most favorable African environments in which to envisage efficient and efficacious antimalarial activities. Although studies in rural areas report marginally significantly higher EIRs compared with periurban studies, these latter areas are much more densely inhabited. It can be assumed that if vector control can be successful in periurban areas, a greater number of people would be affected. Vector control must take into account variation in transmission patterns at the scale of a district, subdistrict, or quarter. In many urban areas, larval sites are few. They also are easily located and accessible. This fact can constitute a keystone for effective control. The arsenal for the antivectorial includes antilarval insecticide treatments, larvivorous fishes, house spraying, impregnated mosquito nets, and curtains.59 In some favorable situations, the final goal can be to make cities malaria-free areas. The implementation of a malaria-free area, small in surface area but large with regard to population, would stimulate neighboring areas, local authorities, and eventually funding agencies for their extension.60
Many valuable future research areas emerge from this analysis. We need more information not only on the human behavioral determinants, but also on the entomologic characteristics of different districts within cities before we can advise health organizations confidently how best to proceed toward reducing the burden of malaria in urban settings. Vector control seems to be an efficient weapon to reduce transmission with direct consequences in lowering the incidence of malaria cases. This reduction of malaria transmission in the most favorable situations within urban SSA might be efficient enough to create malaria-free areas.
Received June 10, 2002. Accepted for publication October 10, 2002.
Financial support: This study was funded in part by Institut de Recherche pour le Développement and by NIH grants U19 AI45511, D43 TW01142, D43 TW00920, and NSF DEB-0083602.
Authors addresses: Vincent Robert, UR paludologie afro-tropicale de lInstitut de Recherche pour le Développement, and Institut Pasteur de Madagascar, B.P. 1274, Antananarivo 101, Madagascar, E-mail: robert{at}pasteur.mg. Kate Macintyre and Joseph Keating, Department of International Health and Development, Tulane University, 1440 Canal Street, Suite 2200, New Orleans, LA 70112. Jean-François Trape, UR paludologie afro-tropicale de lInstitut de Recherche pour le Développement, B.P. 1386, Dakar, Senegal. Jean-Bernard Duchemin, Institut Pasteur de Madagascar, B.P. 1274, Antananarivo 101, Madagascar. McWilson Warren, Retired Public Health Service Officer, Grafton, NH. John C. Beier, Department of Tropical Medicine, Tulane University, 1501 Canal Street, Room 505, New Orleans, LA 70112.
Reprint requests: Vincent Robert, Institut Pasteur de Madagascar, B.P. 1274 Antananarivo 101, Madagascar, E-mail: robert{at}pasteur.mg
| REFERENCES |
|
|
|---|
This article has been cited by other articles:
![]() |
J. Stoler, J. R. Weeks, A. Getis, and A. G. Hill Distance Threshold for the Effect of Urban Agriculture on Elevated Self-reported Malaria Prevalence in Accra, Ghana Am J Trop Med Hyg, April 1, 2009; 80(4): 547 - 554. [Abstract] [Full Text] [PDF] |
||||
![]() |
I. Peterson, L. N. Borrell, W. El-Sadr, and A. Teklehaimanot Individual and Household Level Factors Associated with Malaria Incidence in a Highland Region of Ethiopia: A Multilevel Analysis Am J Trop Med Hyg, January 1, 2009; 80(1): 103 - 111. [Abstract] [Full Text] [PDF] |
||||
![]() |
M. F. Somi, J. R. G. Butler, F. Vahid, J. Njau, S. P. Kachur, and S. Abdulla Is There Evidence for Dual Causation Between Malaria and Socioeconomic Status? Findings From Rural Tanzania Am J Trop Med Hyg, December 1, 2007; 77(6): 1020 - 1027. [Abstract] [Full Text] [PDF] |
||||
![]() |
L. E. G. Mboera, E. A. Makundi, and A. Y. Kitua Uncertainty in Malaria Control in Tanzania: Crossroads and Challenges for Future Interventions Am J Trop Med Hyg, December 1, 2007; 77(6_Suppl): 112 - 118. [Abstract] [Full Text] [PDF] |
||||
![]() |
M. L. H. MABASO, M. CRAIG, A. ROSS, and T. SMITH ENVIRONMENTAL PREDICTORS OF THE SEASONALITY OF MALARIA TRANSMISSION IN AFRICA: THE CHALLENGE Am J Trop Med Hyg, January 1, 2007; 76(1): 33 - 38. [Abstract] [Full Text] [PDF] |
||||
![]() |
G. F. KILLEEN, A. ROSS, and T. SMITH INFECTIOUSNESS OF MALARIA-ENDEMIC HUMAN POPULATIONS TO VECTORS. Am J Trop Med Hyg, August 1, 2006; 75(2_suppl): 38 - 45. [Abstract] [Full Text] [PDF] |
||||
![]() |
F. TEDIOSI, N. MAIRE, T. SMITH, G. HUTTON, J. UTZINGER, A. ROSS, and M. TANNER AN APPROACH TO MODEL THE COSTS AND EFFECTS OF CASE MANAGEMENT OF PLASMODIUM FALCIPARUM MALARIA IN SUB-SAHARAN AFRICA. Am J Trop Med Hyg, August 1, 2006; 75(2_suppl): 90 - 103. [Abstract] [Full Text] [PDF] |
||||
![]() |
H. BOGREAU, F. RENAUD, H. BOUCHIBA, P. DURAND, S.-B. ASSI, M.-C. HENRY, E. GARNOTEL, B. PRADINES, T. FUSAI, B. WADE, et al. GENETIC DIVERSITY AND STRUCTURE OF AFRICAN PLASMODIUM FALCIPARUM POPULATIONS IN URBAN AND RURAL AREAS Am J Trop Med Hyg, June 1, 2006; 74(6): 953 - 959. [Abstract] [Full Text] [PDF] |
||||
![]() |
A. K Rowe, S. Y Rowe, R. W Snow, E. L Korenromp, J. R. A. Schellenberg, C. Stein, B. L Nahlen, J. Bryce, R. E Black, and R. W Steketee The burden of malaria mortality among African children in the year 2000 Int. J. Epidemiol., June 1, 2006; 35(3): 691 - 704. [Abstract] [Full Text] [PDF] |
||||
![]() |
G. F. Killeen, M. Tanner, W. R. Mukabana, M. S. Kalongolela, K. Kannady, S. W. Lindsay, U. Fillinger, and M. C. de Castro HABITAT TARGETING FOR CONTROLLING AQUATIC STAGES OF MALARIA VECTORS IN AFRICA Am J Trop Med Hyg, April 1, 2006; 74(4): 517 - 518. [Full Text] [PDF] |
||||
![]() |
D. P. MATHANGA, C. H. CAMPBELL, T. E. TAYLOR, R. BARLOW, and M. L. WILSON REDUCTION OF CHILDHOOD MALARIA BY SOCIAL MARKETING OF INSECTICIDE-TREATED NETS: A CASE-CONTROL STUDY OF EFFECTIVENESS IN MALAWI Am J Trop Med Hyg, September 1, 2005; 73(3): 622 - 625. [Abstract] [Full Text] [PDF] |
||||
![]() |
L. OSORIO, J. TODD, and D. J. BRADLEY TRAVEL HISTORIES AS RISK FACTORS IN THE ANALYSIS OF URBAN MALARIA IN COLOMBIA Am J Trop Med Hyg, October 1, 2004; 71(4): 380 - 386. [Abstract] [Full Text] [PDF] |
||||
![]() |
G. F. KILLEEN, A. SEYOUM, and B. G. J. KNOLS RATIONALIZING HISTORICAL SUCCESSES OF MALARIA CONTROL IN AFRICA IN TERMS OF MOSQUITO RESOURCE AVAILABILTY MANAGEMENT Am J Trop Med Hyg, August 1, 2004; 71(2_suppl): 87 - 93. [Abstract] [Full Text] [PDF] |
||||
![]() |
M. C. DE CASTRO, Y. YAMAGATA, D. MTASIWA, M. TANNER, J. UTZINGER, J. KEISER, and B. H. SINGER INTEGRATED URBAN MALARIA CONTROL: A CASE STUDY IN DAR ES SALAAM, TANZANIA Am J Trop Med Hyg, August 1, 2004; 71(2_suppl): 103 - 117. [Abstract] [Full Text] [PDF] |
||||
![]() |
J. KEISER, J. UTZINGER, M. C. DE CASTRO, T. A. SMITH, M. TANNER, and B. H. SINGER URBANIZATION IN SUB-SAHARAN AFRICA AND IMPLICATION FOR MALARIA CONTROL Am J Trop Med Hyg, August 1, 2004; 71(2_suppl): 118 - 127. [Abstract] [Full Text] [PDF] |
||||
![]() |
J. SHILILU, T. GHEBREMESKEL, S. MENGISTU, H. FEKADU, M. ZEROM, C. MBOGO, J. GITHURE, R. NOVAK, E. BRANTLY, and J. C. BEIER HIGH SEASONAL VARIATION IN ENTOMOLOGIC INOCULATION RATES IN ERITREA, A SEMI-ARID REGION OF UNSTABLE MALARIA IN AFRICA Am J Trop Med Hyg, December 1, 2003; 69(6): 607 - 613. [Abstract] [Full Text] [PDF] |
||||
![]() |
S. G. STAEDKE, E. W. NOTTINGHAM, J. COX, M. R. KAMYA, P. J. ROSENTHAL, and G. DORSEY SHORT REPORT: PROXIMITY TO MOSQUITO BREEDING SITES AS A RISK FACTOR FOR CLINICAL MALARIA EPISODES IN AN URBAN COHORT OF UGANDAN CHILDREN Am J Trop Med Hyg, September 1, 2003; 69(3): 244 - 246. [Abstract] [Full Text] [PDF] |
||||
| |||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |