- Email: [email protected]
Stability of malaria in a community in Bataan, The Philippines: prospects for control
Acta Tropica 63 (1997) 267 – 273
Stability of malaria in a community in Bataan, The Philippines: prospects for control A. Saula,*, V.Y Belizariob, M.D.G. Bustosc, F. Espinoc, M.A. Lansangc, N.P. Salazard, E. Torresc a Australian Centre for International and Tropical Health and Nutrition, The Queensland Institute of Medical Research, P.O. Royal Brisbane Hospital, Queensland 4029, Australia b College of Public Health, Uni6ersity of The Philippines Manila, Ermita, Manila, Philippines c Research Institute for Tropical Medicine, Alabang, Muntinlupa, Metro Manila, Philippines d SEAMEO-TROPMED, 420 /6 Raj6ithi Road, Bangkok 10400, Thailand
Received 22 May 1996; revised 30 September 1996; accepted 7 October 1996
Abstract Malaria in Morong, Bataan, The Philippines, a municipality with relatively low level, but stable malaria is associated with small foci of relatively high endemicity. Although there is little association between age and symptomatic malaria, there is a reservoir of asymptomatic cases which are present throughout the year. Risk analysis suggests that the greatest risk factor in acquiring malaria depends on place of residence and not on occupation, including those associated with forest activities such as charcoal making. Foci of infection and the timing of symptomatic cases is closely correlated with breeding sites and abundance of adult Anopheles fla6irostris. In spite of this close association, widely held views in the community that malaria is not related to mosquito transmission are likely to make better malaria control based on vector control difficult to sustain. Observation of treatment practices in the community and estimates of the number of apparently asymptomatic carriers from active case detection illustrate the importance of delayed treatment in providing a continuing reservoir of infection. These results highlight the need for improved early case detection and treatment. © 1997 Elsevier Science B.V. All rights reserved Keywords: Malaria; Health-seeking behaviour; Case detection; Anopheles fla6irostris; Plasmodium falciparum; Plasmodium 6i6ax * Corresponding author. Tel.: +61 7 33620402; fax: + 61 7 33620104; e-mail: [email protected] 0001-706X/97/$17.00 © 1997 Elsevier Science B.V. All rights reserved PII S 0 0 0 1 - 7 0 6 X ( 9 6 ) 0 0 6 2 6 - 2
268
A. Saul et al. / Acta Tropica 63 (1997) 267–273
1. Introduction The studies reported in the previous papers represent one of the largest surveys of malaria carried out in an area of low endemic malaria and one of the few studies where entomological, parasitological and behavioural data have been collected (Belizario et al., 1997; Bustos et al., 1997; Espino et al., 1997; Lansang et al., 1997; Torres et al., 1997). In this paper, we present an integrated summary of the results and explore the implications for control.
2. Stability of malaria Studies over the three year period September 1989–1992 reinforced the view that the number of malaria cases coming from Morong was reasonably stable. Although the transmission rates varied within Morong, there was no evidence of large scale epidemics. As such, the pattern of malaria in this area conforms to the original description of ‘stable malaria’ by Macdonald (1957). Unfortunately, the term ‘stable malaria’ has been used by later workers to refer to areas of high prevalence with a markedly skewed age distribution with most cases in children and by inference, ‘unstable malaria’ has become synonymous with areas of low endemicity, regardless of the actual stability or otherwise of the malaria. This study illustrates that care is needed in applying such terms.
2.1. High IFAT area The area with the highest parasite prevalence, the highest prevalence of high IFAT titres and the highest frequency of people coming to the clinics, accounted for approximately 50% of the malaria in Morong, although the population was only about 900, or 4.8% of the total. In this area prevalence was 10–15% at the height of the transmission season and was still relatively high during the wet or low transmission season of September–December. Although the prevalence was relatively high, because the population was quite small, the actual number of people infected at any time was also quite small. For example, during the first active case detection (ACD) survey in 1992, we project that if all the population had been surveyed, that there would have been about 66 patent cases of Plasmodium 6i6ax and about 28 of P. falciparum. (Belizario et al., 1997). Several lines of evidence suggest that many of these people detected by ACD were chronically infected. Most had either no or mild symptoms. A high proportion of the P. falciparum infections were gametocytaemic indicating that the infection must have been patent for at least a week before detected. A survey in September – October 1989 (Bustos et al., 1997), the period with relatively few clinical cases, still found significant numbers of people infected and amongst these, the proportion of gametocyte positive was high. An investigation of health-seeking behaviour (Espino et al., 1997) indicated that many would not seek adequate treatment when malaria was suspected.
A. Saul et al. / Acta Tropica 63 (1997) 267–273
269
Within this population, it appears that malaria is maintained by transmission in and around the houses. In the high IFAT area, clusters of houses, or sitios (Bustos et al., 1997) were located near breeding sites of the major incriminated vector, Anopheles fla6irostris (Torres et al., 1997). Risk analysis suggested that activity such as forest work played a relatively minor role in the risk of being parasiaemic as detected by ACD (Lansang et al., 1997). A long term human reservoir of infection presumably contributes to the stability of malaria in this area. This would be critical for P. falciparum to enable it to survive from one transmission season to another. It may also be important for P. 6i6ax, but for this parasite, hypnozoite forms may contribute to a long term reservoir.
2.2. Medium IFAT area Approximately 2500 (13%) people lived in the area where the prevalence of high IFAT titres was between 25 and 50% and this area contributed about 40% of the cases of malaria detected. The overall number of cases as judged by clinical records, ACD and IFAT titres suggests that although the prevalence is lower, the number of cases is similar from year to year. However, as may be expected from an area with relatively low prevalence, the available data suggests more fluctuation in individual sitios from year to year. For example, in the adjoining sitios of Anahao and Pag-asa, situated about 3 km northeast of the main centre of the Poblacion (Bustos et al., 1997; Torres et al., 1997), more than 20 cases were detected in a preliminary ACD survey in 1991 (Cheng et al., 1993) whereas only 5 and 2 cases were detected in 1992 and 1993, respectively. The relative importance of localized transmission and importation of cases from the high IFAT area could not be determined. The ACD survey indicated that like the high IFAT area, there were people with apparently long term chronic infections who could act as an infectious reservoir. On the other hand, entomology surveys only found larval An. fla6irostris adjacent to some sitios in this area. While the higher ratio of people to mosquitoes may be a factor in the lower endemicity of this area, other factors in the mosquito ecology may also be important, e.g. the peak man landing rates in Anahao were recorded in September and October 1992, a period where the overall abundance in Morong of An. fla6irostris is low. From these data, we speculate that localized transmission is important, but that importation from the high IFAT sitios plays a role in long term maintenance of the infection.
2.3. Low IFAT area Malaria in the low IFAT area appears both quantitatively and qualitatively different to the pattern observed in the higher prevalence areas of Morong. Unlike these areas, there was a highly significant association with the risk of getting malaria and occupation (Lansang et al., 1997). Most of the cases of malaria in this area came from one sitio, Ibayo, where there were a number of people who had plots of land situated in the high IFAT area (Espino et al., 1997). It appears that
270
A. Saul et al. / Acta Tropica 63 (1997) 267–273
the ongoing presence of malaria in this area is due to the stability of malaria in the nearby medium and high IFAT area and the lack of IFAT titres or parasites in children and the inability to find vector breeding sites would suggest that little or no transmission occurs in this area.
3. Implications for better control programs
3.1. Potential for impro6ement Snow and Marsh (1995) proposed that in areas of stable malaria in Africa, there was no correlation between disease from malaria and the entomological inoculation rates (EIR) and therefore little improvement could be expected in the disease without major efforts in control programs. Others have pointed out that in improving control, there may be a risk of increasing disease due to waning clinical immunity in the population (Sauerwein and Meuwissen, 1995). Our assessment suggests that neither of these concerns are valid in the area we studied. Although actual EIR rates were not measured in this study, as detailed in Belizario et al. (1997) indirect evidence from seropositivity rates, parasite prevalence and frequency of passive case detection (PCD) cases suggests that most people in the high IFAT area (i.e. the area with \ 50% of the population with a high IFAT score) will be infected each year. Since the EIR must be greater than the actual infection rates, the EIR in this area would be at least one infectious bite per person per year. The frequency of symptomatic malaria does vary within Morong and in a manner which suggests that the frequency is closely linked with vector abundance both in space and time and thus, with the EIR (Belizario et al., 1997; Torres et al., 1997). This is a low endemic area and it may be expected that there would be at least a sizable proportion of the population with little clinical immunity. Perhaps surprisingly, even in the sitios with the highest endemicity, as detailed in Belizario et al. (1997) there was no marked effect of age on the frequency of symptomatic and asymptomatic cases suggesting that the probability of developing symptoms was not simply a function of past exposure nor does it support the hypothesis by Baird (1995) who suggested that adults are able to acquire clinical immunity faster than children. It is conceivable that the lack of observed age dependency is an artifact with the acquisition of clinical immunity balanced by an increased adult exposure. However, there is no evidence that children have a lower inoculation rate, and to explain the lack of age dependence on the basis of simultaneous changes in inoculation rate and clinical immunity in both the medium and high IFAT areas would require an unlikely set of coincidences. The frequency of severe malaria as defined using standard World Health Organization criteria (WHO, 1994) was too low to estimate. We estimate that there were more than 2000 infections per year occurring in the study population (Belizario et al., 1997), yet review of the clinical records indicate that over the 3 year period
A. Saul et al. / Acta Tropica 63 (1997) 267–273
271
covering the preliminary field work and the surveys that only two deaths attributable to malaria occurred. Both were young adults: a pregnant woman and a male. This situation appears different to that reported from low endemic areas in Africa (Mbogo et al., 1995; Trape et al., 1987; Trape et al., 1993) where severe disease is relatively frequent. Two of the locations in the Kilifi district studied by Mbogo et al. (1995) had EIR comparable to that found in the higher transmission areas of Morong, but had a significant number of children (B 5 years old) with severe malaria. The reasons for this difference are not clear. It is possible that the ready availability of treatment at the sitio level with access to clinical facilities if the first level of self treatment fails together with a high awareness of malaria in the community (Espino et al., 1997) provides sufficient cover to prevent most serious disease, even if insufficient to prevent the continuing transmission. These two observations give reassurance that in the Philippine context, control programs are worthwhile. They are likely to give a direct decrease in the number of symptomatic cases. Importantly, the data also suggests that the fear of instigating a program aimed at a major reduction of parasite prevalence will give rise to a large number of cases of severe malaria as natural immunity wanes is unfounded. If that were the case, we would have expected to see many cases of severe malaria in the medium IFAT area where the prevalence of high IFAT suggests that most people are relatively infrequently infected.
3.2. Possible control strategies Continuing malaria transmission depends on two factors, a human reservoir of infection and a suitable vector. The program at Morong identified that both were important. This suggests that two, not necessarily exclusive, strategies could be used to improve malaria control in Morong: better vector control and reduction of the human reservoir by better treatment. Vector control measures have been a major part of the existing control program, with an emphasis on house spraying. More recently, impregnated bed nets are being distributed and a program of introducing larvicidal fish into streams has been revitalized. To be successful, these programs require a long term commitment and a high degree of local cooperation, e.g. bed nets will only work if the residents use them regularly. There is surprisingly little data available in the Philippines on the efficacy of these programs. For example, despite many years of house spraying there is little published information on the effects on the vectorial capacity of the principal vector, An. fla6irostris. On theoretical grounds, bed nets if used regularly, should be quite effective since the An. fla6irostris predominantly bites late at night when most people would be protected (Torres et al., 1997). However, the study into perceptions of malaria indicated that a high proportion of the community did not believe that malaria was primarily carried by mosquitoes (Espino et al., 1997). In spite of the ready availability of posters and other educational material in Morong, most people believed that the primary route of transmission was through contaminated water. Clearly, without a major education program or incentives unrelated to malaria control, long term enthusiasm for and compliance with an intensified vector control program remains problematic.
272
A. Saul et al. / Acta Tropica 63 (1997) 267–273
A second approach concentrates on better detection and treatment of malaria carriers within the community. Studies of community perceptions and health-seeking behaviour (Espino et al., 1997) indicated that treatment for malaria was widely understood, but the practices of self treatment often led to inadequate doses and the availability of treatment through official channels, e.g. the Rural Health Units, was not optimal, especially due to delays in the ability to provide positive identification of malaria through blood films. In this area, drug resistance is not a significant factor. A study of treatment failure following chloroquine treatment carried out during the first ACD showed that treatment failure was associated with a failure to complete the course of treatment (Burton et al., unpublished M.Sc. thesis, University of Queensland, 1992) Studies done over the period 1989–1993 showed that relatively asymptomatic people could be detected through ACD and these were often gametocytaemic (Bustos et al., 1997; Belizario et al., 1997). Furthermore, many of the people who were symptomatic with P. falciparum were also gametocytaemic, indicating that they also had been patent for some time. Although such carriers could be found, the number at any time was not likely to be high and so extra effort in finding and rapidly treating this reservoir of infection should make a substantial impact. Hopefully, the credibility gained in the community through successful intervention based on rapid case detection and treatment may lead to more acceptance of other measures such as better vector control which will help the sustainability of such campaigns.
4. Postscript Based on these considerations, a community based volunteer scheme was introduced in the later part of 1993. The full details of this scheme will be reported elsewhere (Espino et al., in preparation), but briefly, 30 volunteers from the sitios identified as the medium and high IFAT areas were trained to make malaria thick and thin films from people with symptoms suggestive of malaria. The municipal government provided transport to enable the films to be rapidly forwarded to the Rural Health Unit, where they were read by a microscopist. Anti-malaria drugs were issued and their administration supervised by the community volunteer. Most patients were treated within 24 h of consulting the community volunteer. In the first year of operation (1994) the number of cases of malaria detected approximately doubled, indicating that the surveillance was working well, but apparently having little effect on the number of new cases. However, the number of cases detected in the next transmission season (January–May, 1995) fell by over 90%, and at the time of writing (March, 1996), remained at very low levels through the start of the 1996 transmission season. The results from this study are now being applied to the development of a control program to cover the 450 000 people in the Province of Agusan del Sur on the Island of Mindinao.
A. Saul et al. / Acta Tropica 63 (1997) 267–273
273
Acknowledgements Our heartfelt thanks and acknowledgements are given to all those who made this study possible: special mention is due to the people of the Municipality of Morong, Bataan, to the staff of the Socio-Behavioural Group, the Data Management Unit and the Department of Parasitology and Medical Entomology of the Research Institute for Tropical Medicine; to the personnel of the Malaria Control Service Central Office in Manila and of the Balanga Regional Office, to Dr Rolando Banson to the Philippine Refugee Processing Center Health Services Group and to the Rural Health Unit Morong staff, Dr Benjie Dacula, the nurses, midwives and barangay health workers. This study was funded by a NIH TMRC grant (No. SRC (55) 5 P50 AI030601-02, grants from the World Health Organization/ World Bank Special Program for Research and Training in Tropical Diseases (Social and Economic Research Programme), the Australian National Health and Medical Research Council and the Research Institute for Tropical Medicine, Department of Health, Philippines.
References Baird, J.K. (1995) Host age as a determinant of naturally acquired immunity to Plasmodium falciparum. Parasitol. Today 11, 105–111. Belizario, V.Y. Saul, A., Bustos, M.D.G. et al. (1997) Field epidemiological studies on malaria in a low endemic area in The Philippines. Acta Trop. 63, 241 – 256. Bustos, M.D.G., Saul A., Lansang, M.A., Salazar, N.P. and Gomes, M. (1997) Profile of Morong, Bataan, an area of low malaria endemicity in The Philippines. Acta Trop. 63, 195 – 207. Cheng, Q., Stowers, A., Huang, T-Y. et al. (1993) Polymorphism in Plasmodium 6i6ax MSA1 gene — the result of intragenic recombinations? Parasitology 106, 335 – 345. Espino, F., Manderson, L., Aciun, C., Domingo, F. and Ventura, E. (1997) Perceptions of malaria in a low endemic area in The Philippines: transmission and prevention of disease. Acta Trop. 63, 221–239. Lansang, M.A., Belizario, V.Y., Bustos, M.D.G., Saul, A. and Aguirre, A.(1997) Risk factors for infection with malaria in a low endemic community in Bataan, The Philippines. Acta Trop. 63, 257–265. Macdonald, G. (1957) The epidemiology and control of malaria. Oxford University Press, London. Mbogo, C.N., Snow, R.W., Khamala, C.P.M. et al. (1995) Relationship between Plasmodium falciparum transmission by vector populations and the incidence of severe disease at nine sites on the Kenyan coast. Am. J. Trop. Med. Hyg. 52, 201 – 206. Sauerwein, R. and Meuwissen, J. (1995) Is reduction of transmission desirable for malaria control? Parasitol. Today 11, 425–526. Snow, R.W. and Marsh, K. (1995) Will reducing Plasmodium falciparum transmission alter malaria mortality among African children? Parasitol. Today 11, 188 – 190. Torres, E.P., Salazar, N.P., Belilzario, V.Y. and Saul, A. (1997) Vector abundance and behaviour in an area of low malaria endemicity in Bataan, The Philippines. Acta Trop. 63, 209 – 220. Trape, J.F., Quinet, M.C., Nzingoula, S. et al. (1987) Malaria and urbanization in Central Africa: the example of Brazzaville. Part V: pernicious attacks and mortality. Trans. R. Soc. Trop. Med. Hyg. 81 (2), 34–42. Trape, J.F., Lefebre-Zante, E., Legros, F. et al. (1993) Malaria morbidity among children exposed to low seasonal malaria in Dakar, Senegal and its implications for malaria control in tropical Africa. Am. J. Trop. Med. Hyg. 48, 748 – 756.
Stability of malaria in a community in Bataan, The Philippines: prospects for control A. Saula,*, V.Y Belizariob, M.D.G. Bustosc, F. Espinoc, M.A. Lansangc, N.P. Salazard, E. Torresc a Australian Centre for International and Tropical Health and Nutrition, The Queensland Institute of Medical Research, P.O. Royal Brisbane Hospital, Queensland 4029, Australia b College of Public Health, Uni6ersity of The Philippines Manila, Ermita, Manila, Philippines c Research Institute for Tropical Medicine, Alabang, Muntinlupa, Metro Manila, Philippines d SEAMEO-TROPMED, 420 /6 Raj6ithi Road, Bangkok 10400, Thailand
Received 22 May 1996; revised 30 September 1996; accepted 7 October 1996
Abstract Malaria in Morong, Bataan, The Philippines, a municipality with relatively low level, but stable malaria is associated with small foci of relatively high endemicity. Although there is little association between age and symptomatic malaria, there is a reservoir of asymptomatic cases which are present throughout the year. Risk analysis suggests that the greatest risk factor in acquiring malaria depends on place of residence and not on occupation, including those associated with forest activities such as charcoal making. Foci of infection and the timing of symptomatic cases is closely correlated with breeding sites and abundance of adult Anopheles fla6irostris. In spite of this close association, widely held views in the community that malaria is not related to mosquito transmission are likely to make better malaria control based on vector control difficult to sustain. Observation of treatment practices in the community and estimates of the number of apparently asymptomatic carriers from active case detection illustrate the importance of delayed treatment in providing a continuing reservoir of infection. These results highlight the need for improved early case detection and treatment. © 1997 Elsevier Science B.V. All rights reserved Keywords: Malaria; Health-seeking behaviour; Case detection; Anopheles fla6irostris; Plasmodium falciparum; Plasmodium 6i6ax * Corresponding author. Tel.: +61 7 33620402; fax: + 61 7 33620104; e-mail: [email protected] 0001-706X/97/$17.00 © 1997 Elsevier Science B.V. All rights reserved PII S 0 0 0 1 - 7 0 6 X ( 9 6 ) 0 0 6 2 6 - 2
268
A. Saul et al. / Acta Tropica 63 (1997) 267–273
1. Introduction The studies reported in the previous papers represent one of the largest surveys of malaria carried out in an area of low endemic malaria and one of the few studies where entomological, parasitological and behavioural data have been collected (Belizario et al., 1997; Bustos et al., 1997; Espino et al., 1997; Lansang et al., 1997; Torres et al., 1997). In this paper, we present an integrated summary of the results and explore the implications for control.
2. Stability of malaria Studies over the three year period September 1989–1992 reinforced the view that the number of malaria cases coming from Morong was reasonably stable. Although the transmission rates varied within Morong, there was no evidence of large scale epidemics. As such, the pattern of malaria in this area conforms to the original description of ‘stable malaria’ by Macdonald (1957). Unfortunately, the term ‘stable malaria’ has been used by later workers to refer to areas of high prevalence with a markedly skewed age distribution with most cases in children and by inference, ‘unstable malaria’ has become synonymous with areas of low endemicity, regardless of the actual stability or otherwise of the malaria. This study illustrates that care is needed in applying such terms.
2.1. High IFAT area The area with the highest parasite prevalence, the highest prevalence of high IFAT titres and the highest frequency of people coming to the clinics, accounted for approximately 50% of the malaria in Morong, although the population was only about 900, or 4.8% of the total. In this area prevalence was 10–15% at the height of the transmission season and was still relatively high during the wet or low transmission season of September–December. Although the prevalence was relatively high, because the population was quite small, the actual number of people infected at any time was also quite small. For example, during the first active case detection (ACD) survey in 1992, we project that if all the population had been surveyed, that there would have been about 66 patent cases of Plasmodium 6i6ax and about 28 of P. falciparum. (Belizario et al., 1997). Several lines of evidence suggest that many of these people detected by ACD were chronically infected. Most had either no or mild symptoms. A high proportion of the P. falciparum infections were gametocytaemic indicating that the infection must have been patent for at least a week before detected. A survey in September – October 1989 (Bustos et al., 1997), the period with relatively few clinical cases, still found significant numbers of people infected and amongst these, the proportion of gametocyte positive was high. An investigation of health-seeking behaviour (Espino et al., 1997) indicated that many would not seek adequate treatment when malaria was suspected.
A. Saul et al. / Acta Tropica 63 (1997) 267–273
269
Within this population, it appears that malaria is maintained by transmission in and around the houses. In the high IFAT area, clusters of houses, or sitios (Bustos et al., 1997) were located near breeding sites of the major incriminated vector, Anopheles fla6irostris (Torres et al., 1997). Risk analysis suggested that activity such as forest work played a relatively minor role in the risk of being parasiaemic as detected by ACD (Lansang et al., 1997). A long term human reservoir of infection presumably contributes to the stability of malaria in this area. This would be critical for P. falciparum to enable it to survive from one transmission season to another. It may also be important for P. 6i6ax, but for this parasite, hypnozoite forms may contribute to a long term reservoir.
2.2. Medium IFAT area Approximately 2500 (13%) people lived in the area where the prevalence of high IFAT titres was between 25 and 50% and this area contributed about 40% of the cases of malaria detected. The overall number of cases as judged by clinical records, ACD and IFAT titres suggests that although the prevalence is lower, the number of cases is similar from year to year. However, as may be expected from an area with relatively low prevalence, the available data suggests more fluctuation in individual sitios from year to year. For example, in the adjoining sitios of Anahao and Pag-asa, situated about 3 km northeast of the main centre of the Poblacion (Bustos et al., 1997; Torres et al., 1997), more than 20 cases were detected in a preliminary ACD survey in 1991 (Cheng et al., 1993) whereas only 5 and 2 cases were detected in 1992 and 1993, respectively. The relative importance of localized transmission and importation of cases from the high IFAT area could not be determined. The ACD survey indicated that like the high IFAT area, there were people with apparently long term chronic infections who could act as an infectious reservoir. On the other hand, entomology surveys only found larval An. fla6irostris adjacent to some sitios in this area. While the higher ratio of people to mosquitoes may be a factor in the lower endemicity of this area, other factors in the mosquito ecology may also be important, e.g. the peak man landing rates in Anahao were recorded in September and October 1992, a period where the overall abundance in Morong of An. fla6irostris is low. From these data, we speculate that localized transmission is important, but that importation from the high IFAT sitios plays a role in long term maintenance of the infection.
2.3. Low IFAT area Malaria in the low IFAT area appears both quantitatively and qualitatively different to the pattern observed in the higher prevalence areas of Morong. Unlike these areas, there was a highly significant association with the risk of getting malaria and occupation (Lansang et al., 1997). Most of the cases of malaria in this area came from one sitio, Ibayo, where there were a number of people who had plots of land situated in the high IFAT area (Espino et al., 1997). It appears that
270
A. Saul et al. / Acta Tropica 63 (1997) 267–273
the ongoing presence of malaria in this area is due to the stability of malaria in the nearby medium and high IFAT area and the lack of IFAT titres or parasites in children and the inability to find vector breeding sites would suggest that little or no transmission occurs in this area.
3. Implications for better control programs
3.1. Potential for impro6ement Snow and Marsh (1995) proposed that in areas of stable malaria in Africa, there was no correlation between disease from malaria and the entomological inoculation rates (EIR) and therefore little improvement could be expected in the disease without major efforts in control programs. Others have pointed out that in improving control, there may be a risk of increasing disease due to waning clinical immunity in the population (Sauerwein and Meuwissen, 1995). Our assessment suggests that neither of these concerns are valid in the area we studied. Although actual EIR rates were not measured in this study, as detailed in Belizario et al. (1997) indirect evidence from seropositivity rates, parasite prevalence and frequency of passive case detection (PCD) cases suggests that most people in the high IFAT area (i.e. the area with \ 50% of the population with a high IFAT score) will be infected each year. Since the EIR must be greater than the actual infection rates, the EIR in this area would be at least one infectious bite per person per year. The frequency of symptomatic malaria does vary within Morong and in a manner which suggests that the frequency is closely linked with vector abundance both in space and time and thus, with the EIR (Belizario et al., 1997; Torres et al., 1997). This is a low endemic area and it may be expected that there would be at least a sizable proportion of the population with little clinical immunity. Perhaps surprisingly, even in the sitios with the highest endemicity, as detailed in Belizario et al. (1997) there was no marked effect of age on the frequency of symptomatic and asymptomatic cases suggesting that the probability of developing symptoms was not simply a function of past exposure nor does it support the hypothesis by Baird (1995) who suggested that adults are able to acquire clinical immunity faster than children. It is conceivable that the lack of observed age dependency is an artifact with the acquisition of clinical immunity balanced by an increased adult exposure. However, there is no evidence that children have a lower inoculation rate, and to explain the lack of age dependence on the basis of simultaneous changes in inoculation rate and clinical immunity in both the medium and high IFAT areas would require an unlikely set of coincidences. The frequency of severe malaria as defined using standard World Health Organization criteria (WHO, 1994) was too low to estimate. We estimate that there were more than 2000 infections per year occurring in the study population (Belizario et al., 1997), yet review of the clinical records indicate that over the 3 year period
A. Saul et al. / Acta Tropica 63 (1997) 267–273
271
covering the preliminary field work and the surveys that only two deaths attributable to malaria occurred. Both were young adults: a pregnant woman and a male. This situation appears different to that reported from low endemic areas in Africa (Mbogo et al., 1995; Trape et al., 1987; Trape et al., 1993) where severe disease is relatively frequent. Two of the locations in the Kilifi district studied by Mbogo et al. (1995) had EIR comparable to that found in the higher transmission areas of Morong, but had a significant number of children (B 5 years old) with severe malaria. The reasons for this difference are not clear. It is possible that the ready availability of treatment at the sitio level with access to clinical facilities if the first level of self treatment fails together with a high awareness of malaria in the community (Espino et al., 1997) provides sufficient cover to prevent most serious disease, even if insufficient to prevent the continuing transmission. These two observations give reassurance that in the Philippine context, control programs are worthwhile. They are likely to give a direct decrease in the number of symptomatic cases. Importantly, the data also suggests that the fear of instigating a program aimed at a major reduction of parasite prevalence will give rise to a large number of cases of severe malaria as natural immunity wanes is unfounded. If that were the case, we would have expected to see many cases of severe malaria in the medium IFAT area where the prevalence of high IFAT suggests that most people are relatively infrequently infected.
3.2. Possible control strategies Continuing malaria transmission depends on two factors, a human reservoir of infection and a suitable vector. The program at Morong identified that both were important. This suggests that two, not necessarily exclusive, strategies could be used to improve malaria control in Morong: better vector control and reduction of the human reservoir by better treatment. Vector control measures have been a major part of the existing control program, with an emphasis on house spraying. More recently, impregnated bed nets are being distributed and a program of introducing larvicidal fish into streams has been revitalized. To be successful, these programs require a long term commitment and a high degree of local cooperation, e.g. bed nets will only work if the residents use them regularly. There is surprisingly little data available in the Philippines on the efficacy of these programs. For example, despite many years of house spraying there is little published information on the effects on the vectorial capacity of the principal vector, An. fla6irostris. On theoretical grounds, bed nets if used regularly, should be quite effective since the An. fla6irostris predominantly bites late at night when most people would be protected (Torres et al., 1997). However, the study into perceptions of malaria indicated that a high proportion of the community did not believe that malaria was primarily carried by mosquitoes (Espino et al., 1997). In spite of the ready availability of posters and other educational material in Morong, most people believed that the primary route of transmission was through contaminated water. Clearly, without a major education program or incentives unrelated to malaria control, long term enthusiasm for and compliance with an intensified vector control program remains problematic.
272
A. Saul et al. / Acta Tropica 63 (1997) 267–273
A second approach concentrates on better detection and treatment of malaria carriers within the community. Studies of community perceptions and health-seeking behaviour (Espino et al., 1997) indicated that treatment for malaria was widely understood, but the practices of self treatment often led to inadequate doses and the availability of treatment through official channels, e.g. the Rural Health Units, was not optimal, especially due to delays in the ability to provide positive identification of malaria through blood films. In this area, drug resistance is not a significant factor. A study of treatment failure following chloroquine treatment carried out during the first ACD showed that treatment failure was associated with a failure to complete the course of treatment (Burton et al., unpublished M.Sc. thesis, University of Queensland, 1992) Studies done over the period 1989–1993 showed that relatively asymptomatic people could be detected through ACD and these were often gametocytaemic (Bustos et al., 1997; Belizario et al., 1997). Furthermore, many of the people who were symptomatic with P. falciparum were also gametocytaemic, indicating that they also had been patent for some time. Although such carriers could be found, the number at any time was not likely to be high and so extra effort in finding and rapidly treating this reservoir of infection should make a substantial impact. Hopefully, the credibility gained in the community through successful intervention based on rapid case detection and treatment may lead to more acceptance of other measures such as better vector control which will help the sustainability of such campaigns.
4. Postscript Based on these considerations, a community based volunteer scheme was introduced in the later part of 1993. The full details of this scheme will be reported elsewhere (Espino et al., in preparation), but briefly, 30 volunteers from the sitios identified as the medium and high IFAT areas were trained to make malaria thick and thin films from people with symptoms suggestive of malaria. The municipal government provided transport to enable the films to be rapidly forwarded to the Rural Health Unit, where they were read by a microscopist. Anti-malaria drugs were issued and their administration supervised by the community volunteer. Most patients were treated within 24 h of consulting the community volunteer. In the first year of operation (1994) the number of cases of malaria detected approximately doubled, indicating that the surveillance was working well, but apparently having little effect on the number of new cases. However, the number of cases detected in the next transmission season (January–May, 1995) fell by over 90%, and at the time of writing (March, 1996), remained at very low levels through the start of the 1996 transmission season. The results from this study are now being applied to the development of a control program to cover the 450 000 people in the Province of Agusan del Sur on the Island of Mindinao.
A. Saul et al. / Acta Tropica 63 (1997) 267–273
273
Acknowledgements Our heartfelt thanks and acknowledgements are given to all those who made this study possible: special mention is due to the people of the Municipality of Morong, Bataan, to the staff of the Socio-Behavioural Group, the Data Management Unit and the Department of Parasitology and Medical Entomology of the Research Institute for Tropical Medicine; to the personnel of the Malaria Control Service Central Office in Manila and of the Balanga Regional Office, to Dr Rolando Banson to the Philippine Refugee Processing Center Health Services Group and to the Rural Health Unit Morong staff, Dr Benjie Dacula, the nurses, midwives and barangay health workers. This study was funded by a NIH TMRC grant (No. SRC (55) 5 P50 AI030601-02, grants from the World Health Organization/ World Bank Special Program for Research and Training in Tropical Diseases (Social and Economic Research Programme), the Australian National Health and Medical Research Council and the Research Institute for Tropical Medicine, Department of Health, Philippines.
References Baird, J.K. (1995) Host age as a determinant of naturally acquired immunity to Plasmodium falciparum. Parasitol. Today 11, 105–111. Belizario, V.Y. Saul, A., Bustos, M.D.G. et al. (1997) Field epidemiological studies on malaria in a low endemic area in The Philippines. Acta Trop. 63, 241 – 256. Bustos, M.D.G., Saul A., Lansang, M.A., Salazar, N.P. and Gomes, M. (1997) Profile of Morong, Bataan, an area of low malaria endemicity in The Philippines. Acta Trop. 63, 195 – 207. Cheng, Q., Stowers, A., Huang, T-Y. et al. (1993) Polymorphism in Plasmodium 6i6ax MSA1 gene — the result of intragenic recombinations? Parasitology 106, 335 – 345. Espino, F., Manderson, L., Aciun, C., Domingo, F. and Ventura, E. (1997) Perceptions of malaria in a low endemic area in The Philippines: transmission and prevention of disease. Acta Trop. 63, 221–239. Lansang, M.A., Belizario, V.Y., Bustos, M.D.G., Saul, A. and Aguirre, A.(1997) Risk factors for infection with malaria in a low endemic community in Bataan, The Philippines. Acta Trop. 63, 257–265. Macdonald, G. (1957) The epidemiology and control of malaria. Oxford University Press, London. Mbogo, C.N., Snow, R.W., Khamala, C.P.M. et al. (1995) Relationship between Plasmodium falciparum transmission by vector populations and the incidence of severe disease at nine sites on the Kenyan coast. Am. J. Trop. Med. Hyg. 52, 201 – 206. Sauerwein, R. and Meuwissen, J. (1995) Is reduction of transmission desirable for malaria control? Parasitol. Today 11, 425–526. Snow, R.W. and Marsh, K. (1995) Will reducing Plasmodium falciparum transmission alter malaria mortality among African children? Parasitol. Today 11, 188 – 190. Torres, E.P., Salazar, N.P., Belilzario, V.Y. and Saul, A. (1997) Vector abundance and behaviour in an area of low malaria endemicity in Bataan, The Philippines. Acta Trop. 63, 209 – 220. Trape, J.F., Quinet, M.C., Nzingoula, S. et al. (1987) Malaria and urbanization in Central Africa: the example of Brazzaville. Part V: pernicious attacks and mortality. Trans. R. Soc. Trop. Med. Hyg. 81 (2), 34–42. Trape, J.F., Lefebre-Zante, E., Legros, F. et al. (1993) Malaria morbidity among children exposed to low seasonal malaria in Dakar, Senegal and its implications for malaria control in tropical Africa. Am. J. Trop. Med. Hyg. 48, 748 – 756.