Conceived and designed the experiments: BAO IKM CM. Performed the experiments: BAO IKM AK SM NM JN CM. Analyzed the data: BAO IKM SM. Contributed reagents/materials/analysis tools: BAO TA CM. Wrote the paper: BAO.
The authors have declared that no competing interests exist.
During an entomological survey in preparation for malaria control interventions in Mwea division, the number of malaria cases at the Kimbimbi sub-district hospital was in a steady decline. The underlying factors for this reduction were unknown and needed to be identified before any malaria intervention tools were deployed in the area. We therefore set out to investigate the potential factors that could have contributed to the decline of malaria cases in the hospital by analyzing the malaria control knowledge, attitudes and practices (KAP) that the residents in Mwea applied in an integrated fashion, also known as integrated malaria management (IMM).
Integrated Malaria Management was assessed among community members of Mwea division, central Kenya using KAP survey. The KAP study evaluated community members' malaria disease management practices at the home and hospitals, personal protection measures used at the household level and malaria transmission prevention methods relating to vector control. Concurrently, we also passively examined the prevalence of malaria parasite infection via outpatient admission records at the major referral hospital in the area. In addition we studied the mosquito vector population dynamics, the malaria sporozoite infection status and entomological inoculation rates (EIR) over an 8 month period in 6 villages to determine the risk of malaria transmission in the entire division.
A total of 389 households in Mwea division were interviewed in the KAP study while 90 houses were surveyed in the entomological study. Ninety eight percent of the households knew about malaria disease while approximately 70% of households knew its symptoms and methods to manage it. Ninety seven percent of the interviewed households went to a health center for malaria diagnosis and treatment. Similarly a higher proportion (81%) used anti-malarial medicines bought from local pharmacies. Almost 90% of households reported owning and using an insecticide treated bed net and 81% reported buying the nets within the last 5 years. The community also used mosquito reduction measures including, in order of preference, environmental management (35%), mosquito repellent and smoke (31%) insecticide canister sprays (11%), and window and door screens (6%). These methods used by the community comprise an integrated malaria management (IMM) package. Over the last 4 years prior to this study, the malaria cases in the community hospital reduced from about 40% in 2000 to less than 10% by 2004 and by the year 2007 malaria cases decreased to zero. In addition, a one time cross-sectional malaria parasite survey detected no
The usage of a combination of malaria control tools in an integrated fashion by residents of Mwea division might have influenced the decreased malaria cases in the district hospital and in the school children. A vigorous campaign emphasizing IMM should be adopted and expanded in Mwea division and in other areas with different eco-epidemiological patterns of malaria transmission. With sustained implementation and support from community members integrated malaria management can reduce malaria significantly in affected communities in Africa.
The highest burden of global malaria rests upon the population of sub-Saharan Africa which accounts for 80% of the estimated 2 million deaths that occur around the world annually. The reasons for this include the presence of ideal climatic conditions for the breeding of the malaria vector mosquito, the shortage of committed resources and trained personnel, the lack of commitment from governments and communities ravaged by the disease and poor infrastructural capacities for malaria control. The world's most effective malaria vector mosquito species,
As advocated by the World Health Organization, the combined use of all the proven tactics and available malaria control tools is the most effective way to check the spread of malaria
We conducted an entomological study in Mwea division in central Kenya in preparation for malaria control activities. During this study, we noticed that malaria cases in the local referral hospital were declining and sought to find the underlying reasons. In this paper, we analyze the potential factors that may have contributed to the decrease in malaria cases in this community and report that the community has embraced integrated malaria management, which might have contributed to the reduced malaria cases in Mwea division, Kenya.
The study was conducted in Mwea Division of Kirinyaga district in Central Kenya (
The green marking show the canals running in the rice irrigation paddies. The red lines are the major roadways running through the area and the blue are the several rivers and their tributaries that originate from Mount Kenya and run through Mwea division.
The ecological and climatic conditions in Mwea are ideal for the breeding of malaria vectors. The mean annual rainfall in this area is in the range of 1200–1600 mm per year and varies by the time of year. The long rains begin usually come from March to June and the short rains from October to December. During the period of study and the preceding years, the rainfall patterns followed the expected pattern described above (
This pattern of climate is very typical in this area and is seen every year.
A cross-sectional study was conducted to capture the knowledge, attitude and practices of the community in Mwea division on integrated malaria management (IMM). A structured questionnaire directly addressing IMM usage by the community was administered to 389 households. Study villages and households were selected using a randomized cluster sampling method from a list of households in Mwea division provided by the Central Bureau of Statistics, Ministry of Planning and National Development. These households were physically identified by locally recruited field workers and their locations mapped by a geographical positioning system and the information geo-referenced into the geographic information system (GIS) at Eastern and Southern Africa Centre for International Parasite Control (ESACIPAC) offices in KEMRI. Sensitization and public awareness meetings were held through public meetings (or
The densities of daytime indoor resting anopheline mosquitoes were estimated in 90 households from a sample of 6 villages in Mwea division (
The malaria transmission potential of mosquitoes was determined by estimating the entomological inoculation rate (EIR) in the study villages. The EIR was estimated by multiplying the human biting rate (HBR) and sporozoite rate (SR). The HBR is the mosquito density per house per person multiplied by the proportion of those with human blood in their midgut (Human Blood Index). The SR is the proportion of mosquitoes with sporoziotes in their salivary glands.
Malaria prevalence in the community was assessed by reviewing all the patients' records from the outpatient data at the Kimbimbi sub-District hospital. Consent to view this data was obtained from the District Medical Officer of Health. The Kimbimbi sub-district hospital is the health centre where the majority of the Mwea community seeks treatment. The outpatient records of these patients were collected regardless of age or malaria infection status. From all the outpatient records we only collected laboratory confirmed malaria cases for every month for eight years. The patient details were kept by the health centre although the ESACIPAC malaria team was given access to the data with the permission of the District Medical Officer of Health. The malaria prevalence in the school children was also examined by a one time cross sectional surveillance in public primary schools. Consent to screen the pupils was obtained from parents, guardians and the teachers through Parent Teacher Association meetings that were held prior to the screening. Finger prick blood samples from children were collected and thick blood films made on glass slides. These were later Giemsa stained and examined under a 100× light microscope for malaria parasite identification. For malaria parasite positive samples the species of malaria parasite was determined and the person treated according to the Government policy, through the existing channel at the health centre. The examination of school children and patients records for malaria parasite infection was approved by the KEMRI ethical review committee.
A randomized cluster sampling method was used to select households. Cluster sampling was used because an updated sampling frame was not available. Administrative units called sub-locations within the Mwea division formed the clusters while the primary sampling units were households. To ensure that all the sampled clusters were proportionately represented in the sample, a probability proportional to size of the individual clusters was used to select the number of households. The first household was randomly sampled within the cluster and subsequent households were sampled until the required number of households within that specific cluster was achieved. Consequently, the total number of households required for the calculated sample size was achieved after visiting all the clusters. The sample size (N) of households for the KAP study was determined using the sample size determination equation, where N =
A total of 389 households were selected to participate in the KAP study. Ninety eight percent of all the interviewed households reported malaria as the most common disease in the community. Similarly, 98% of the households also knew malaria and that malaria is transmitted by mosquitoes but could not identify which mosquito species was involved. Sixty eight percent identified malaria symptoms including chills (67%), headache (67%) and body pains (66%). Fifty eight percent reported fever (58%), body weakness (53%) and vomiting (37%) as major symptoms of malaria. Thus the community was knowledgeable about the symptoms of malaria. Three months preceding this study, 76% of households reported at least two family members suffering from malaria and within the same time frame, those households experienced malaria between two to four times. Malaria during pregnancy was reported in 4% of households. Malaria related fatalities were reported in 4% of the households over the last five years prior to this study.
To control malaria stricken members, households sought treatment from local health centers, either government or privately funded. Ninety seven percent of households reported seeking treatment from the nearest health facility while 3% used medications from local chemists, retail shops and other places. A high proportion of households (81%) used generic drugs before seeking treatment in health centers. Of the households that sought treatment, 52% preferred government health centers over privately funded health centers (22%). Eighty six percent reported using sulphadoxine-pyrimethamine (SP) antimalarial drugs including trade names like Fansidar®, Metakelfin®, Orodor® and Falcidin®; 8% used Chloroquine® and Malaratab® and 6% used pain relievers and other unknown medications. Households reported spending approximately USD 12.00 (equivalent to Kes 840.00 at exchange rate of Kes 70.00 to the dollar) for treatment including diagnosis per single malaria case. The maximum amount of money spent on malaria treatment including associated costs of transport, diagnosis and hospital ward admission was USD 600.00 (∼Kes. 42,000.00 per person). This is a clear illustration of the financial burden imposed by malaria on households. Households that did not get treatment for malaria could not afford to pay for treatment while others were negligent. The average distance a malaria victim travelled to a health center was 2.5 kilometers (approximately 1.55 miles). The nearest health facility was 50 meters away while the farthest was 15 kilometers (9.32 miles) away.
About 90% of households interviewed in Mwea division have bed nets, 65% of which are treated with insecticides. This is a very high percentage of bed net and insecticide treated net (ITN) ownership as compared to other areas in the Kenya
Households reported that they used a variety of measures to reduce mosquito populations in their homes. Thirty five percent reported that they used environmental management that included removal of stagnant water bodies and removal of brush in the household compound. In addition, the respondents used several methods to lower the numbers of mosquitoes in their homes including mosquito repellants (31%), insecticides canisters (11%) bought from the shops, window and door screens (6%), traditional plants (3%) and “other methods” (3%). These “other methods” include burning herbs in their homes and avoiding construction of homes in swampy areas. Other mosquito reduction measures that were reported in Mwea Irrigation Scheme include anti-larval measures in rice paddies where rice farming households planted a special plant, a water fern that fixes nitrogen in the soil to improve soil fertility of the paddies. Scientifically known as
A univariate analysis of variance was used to analyze the association of environmental management activities such as clearing bushes around the homestead and removing stagnant water bodies that provide ideal mosquito breeding sites on the indoor density of mosquito vectors, the sporozoite rates and entomological inoculation rates in the study villages. In households with stagnant water, a significant association was seen between bush clearing around homesteads and sporozoite rates (P = 0.013) and EIR (P = 0.019) while in households without stagnant water bodies, there was no significant association between clearing the bushes and sporozoite rates (P = 0.124) or EIR (P = 0.448). This suggests that presence of stagnant water was a risk factor for malaria transmission. Overall, there was a significant association between clearing bushes around homesteads and the indoor density of resting mosquitoes regardless of the presence (P = 0.030) or absence (P = 0.038) of stagnant water bodies
At the Kimbimbi sub-district hospital, over the last 8 years (2000–2007), 46,842 individuals representing 29.3% of the population in Mwea division visited this health centre for malaria treatment. Of those examined, 21% (9,900) were positive for either
The r-square value of the trend line is indicated. The stop gaps in the figure are spaces that separate the years so that the yearly changes in the malaria cases are clearly seen.
A total of 90 households, 15 homesteads from each of the 6 villages were selected for mosquito collection. The average number of persons sleeping in the households the previous night of sampling was 4.3 in Nguka village, 7.5 in Rukanga village, 5.8 in Kirwara village, 3.2 in Kimuri village, 6.0 in Komboini village and 5.5 in Kiumbui village. A total of 13,620 female
Villages sampled in Mwea division | ||||||
Nguka | Rukanga | Kirwara | Kimuri | Kiumbui | Kombuini | |
No of mosquitoes caught | 8,073 | 970 | 1,319 | 1,542 | 1,042 | 596 |
No with sporozoites in salivary gland | 11 | 1 | 2 | 3 | 0 | 0 |
Sporozoite rates | 0.14 | 0.1 | 0.15 | 0.19 | 0 | 0 |
All other anopheline mosquitoes did not test positive for sporoziotes and are not shown in this table.
A total of 8073 female
The pink part represents the proportion that is infected per village while the blue part represents the total number of sporozoite positive mosquitoes caught in the division.
The size of the circles is proportional to the number of infectious bites received. The blue pie in the circle is the proportion of infectious bites contributed by each village to the total EIR.
Blood feeding stage | ||||||||||
E | BF | HG | G | T | E | BF | HG | G | T | |
Number of mosquito | 1,760 | 5,095 | 2,236 | 4,451 | 13,542 | 17 | 22 | 12 | 27 | 78 |
Sporozoite positive | 5 | 7 | 2 | 3 | 17 | 0 | 0 | 0 | 0 | 0 |
Sporozoite rate | 0.28 | 0.14 | 0.09 | 0.07 | 0.13 | 0.00 | 0.00 | 0.00 | 0.00 | 0.00 |
This study illustrates that the use of integrated malaria management can effectively reduce malaria transmission in a malaria endemic community Africa. The community members in Mwea division demonstrated good knowledge about malaria disease and how to control it. Through interviews, the community members revealed that they have known and implemented integrated malaria management for several years by using methods that target the malaria parasite in the human host while at the same time implementing methods that shorten the lifespan of the malaria vectors and those that also reduce the malaria vector population thus preventing malaria transmission. These methods as used by the Mwea community constitute integrated malaria management (IMM) and potentially may have led to the reduction in malaria cases at the referral hospital. The effectiveness of IMM in malaria control has a long history and is being advocated as the best way forward because of its success in many tropical environments
Government health centers are a backbone in case management of malaria and need to be strengthened for effective diagnosis and treatment of malaria. Mwea division has a total of 23 health centers funded both by the government and the private sector. Most of the privately funded health centers are missionary funded. These health centers all together serve a population of 160,000 people in the Mwea division. The drug of choice in the treatment of malaria in these health centers has been sulphadoxine-pyrimethamine (SP) until 2004 when this was changed to by the government to Coartem, an artemisinin based combination therapy, owing to increasing resistance levels to SP in the area of close to 30%. However, the change of antimalarial drug policy from SP to arthemether-lumefanthrine only came into effect in health centers across the country in 2006. By the year 2005, the community in Mwea still used SP drugs because Coartem® was not freely provided by the government and some patients could not afford the nominal fee. In the absence of SP the health centre provided amodiaquine® in its place. Although SP is increasingly becoming ineffective due to resistance by malaria parasites in the Mwea area
Even as the usage of ITNs and effective antimalarial medicines is encouraged, the additive value of other control measures on malaria transmission in the community needs to be recognized. The community members in Mwea made use of their knowledge about the benefits of ITNs and antimalarial medicines in the prevention of malaria transmission and treatment of malaria, respectively. The demand for ITNs and free medication provided by the government increased in the community while at the household level it was clear that other malaria control tools were applied to prevent mosquito bites and mosquito breeding. Homesteads that had stagnant water bodies in their compounds were at greater risk of malaria transmission than homes that did not. In addition, removal of bushes around the homesteads may have helped reduce mosquito population in the home. Homesteads reported covering up puddles of water in their homesteads that would provide ideal breeding grounds for mosquitoes and also clearing bushes around the homesteads that provide cover for adult mosquitoes that emerge from the breeding sites. The removal of potential mosquito breeding sites through environmental management could have been motivated by the proximity of this community to the rice irrigation areas which provide breeding grounds for mosquitoes during planting season. Although stagnant water bodies in the households influenced malaria transmission risk, it is also likely that this association may well be attributable to other numerous confounding factors that we were not able to adjust for. The removal of stagnant water was encouraged by public health officials who were targeting bilharzia in the area. This action may have brought collateral benefits by mitigating malaria mosquito breeding.
The use of environmental management has not been advocated strongly as part of an integrated malaria management package. A major advantage of mosquito breeding site removal is that it is cheaper when you consider the cost per unit head of person protected from malaria mosquitoes. A study in Malindi at the coast of Kenya demonstrated the willingness of the community members to use EM to control mosquitoes
The reduction of malaria cases in the referral hospital in Mwea division over successive years is a reflection of the effectiveness of malaria control measures implemented by this community. Between 1990 and 1995 malaria accounted for 13% of the total deaths reported in Mwea division
In conclusion, it is possible that the free distribution of ITNs resulted in increased usage and the increased availability of antimalarial medicines played a big role in the malaria decline in this community. In addition, the high knowledge base of the community as it pertains to malaria mosquito control measures may have had a bigger role in the observed low malaria. The introduction of a nitrogen fixing weed into the rice paddies in the early 2000 to boost rice production may also have played a role. However the specific impact of this weed on mosquito population reduction needs to be investigated further. Similarly, a thorough study needs to be initiated to investigate the potential use of community driven larval control activities in the area. The combined usage of a variety of malaria control tools that are widely available together with environmental management measures in Mwea division is likely to have resulted in reduced mosquito populations and malaria transmission in the area.
We acknowledge the community members in Mwea division, particularly those households who participated in this study. We are very much indebted to the then District Officer of Mwea division for the efforts he put forward in mobilizing the community to support this study. We also acknowledge the support of the district medical officer of Health and district education officer for supporting this study. We also thank the school teachers and heads of the schools who participated in the study. Many thanks to the local mosquito collection field workers, Mr. Charles Asiago of Walter Reed for his assistance with sporozoite ELISA. We also thank Mr. Kobayashi and Jane Wangui for administrative support. We thank the drivers who endured long and tiring days to the several villages in Mwea division during the collection of this data. This paper has been published with the permission of the Director Kenya Medical Research Institute.