- Email: [email protected]
The management of fevers in Kenyan children and adults in an area of seasonal malaria transmission
Transactions of the Royal Society of Tropical Medicine and Hygiene (2004) 98, 111—115
The management of fevers in Kenyan children and adults in an area of seasonal malaria transmission Helen L. Guyatt*, Robert W. Snow KEMRI-Wellcome Trust Collaborative Programme, GPO 00100, Nairobi, PO Box 43640, Kenya and Centre for Tropical Medicine, University of Oxford, John Radcliffe Hospital, Oxford OX3 9DU, UK Received 10 April 2003 ; received in revised form 31 July 2003; accepted 1 August 2003
KEYWORDS Malaria; Fevers; Treatment; Self-medication; Adults; Sulfadoxinepyrimethamine; Kenya
Summary This study investigates the source, timing and types of treatment for fevers across all ages in a low malaria-transmission area of Kenya. The period prevalence for fever, and subsequent treatment seeking behaviour, was similar across all ages. The use of the informal retail sector was common (47% of first actions), though most visits to shops and chemists (77%) resulted in treatment with an antipyretic not an antimalarial. The major source of the first line recommended drug, sulfadoxine-pyrimethamine (SP), was at the formal health sector, and 32% of fevers made at least one visit to a health care facility. Although only 7% of fevers received SP within 24 hours of fever onset, 27% ultimately received treatment with this antimalarial. It is estimated that of the total amount of SP consumed in this population, only 20% is administered to children less than 5 years old. In this area of Kenya disease risks decline with increasing age, however, adult populations consume over 40% of prescribed or purchased anti-malarial drugs. In light of the proposed new, more costly anti-malarial drug combinations these findings have major implications for the effective allocation of limited financial resources at household and government levels. © 2003 Royal Society of Tropical Medicine and Hygiene. Published by Elsevier Ltd. All rights reserved.
1. Introduction Despite renewed efforts aimed at disease prevention through insecticide-treated bednets, prompt, effective treatment of clinical cases will continue to be a major cornerstone of efforts to Roll Back Malaria (RBM) in Africa. In most settings this relies upon the presumptive treatment of fever (Chandramohan et al., 2002). Our current un*Corresponding author. Present address: c/o Dr. T.P. Robinson, FAO (AGAL), Viale delle Terme di Caracalla, 00100 Rome, Italy. Tel./fax: +39-0763-308-246. E-mail address: [email protected] (H.L. Guyatt).
derstanding of how communities manage fevers derives from investigations of treatment-seeking behaviour among children aged less than 5 years old. This seems reasonable as paediatric populations bear over 75% of the disease burden in Africa (Snow et al., 1999). As such the RBM partnership has proposed that a measurable and appropriate target in Africa should be that 60% of childhood fevers are treated with an appropriate drug within 24 hours of onset (WHO, 2000). Consequently very little is known about the management of fevers amongst older children and adult populations, who are at least risk of poor clinical outcomes following infection, but continue to presumptively seek
0035-9203/$ — see front matter © 2003 Royal Society of Tropical Medicine and Hygiene. Published by Elsevier Ltd. All rights reserved. doi:10.1016/S0035-9203(03)00016-6
112
H.L. Guyatt, R.W. Snow
treatment for malaria. This is of particular relevance to understanding the drug requirements in populations, against ever declining drug budgets and the assumed resistance pressure posed by drug use at community-levels. The current study assessed the treatment seeking behaviour for fevers across all ages in an area of low, seasonal malaria transmission in the western highlands of Kenya.
tion was obtained on the exact source, timing and treatment obtained. All fevers were based only on the patients’ description (no temperatures were measured), and the word for fever is distinct from malaria (Esosera in Kigusii). Sources of treatment were classified as (i) informal retail sector (self-treatment with western-style medications obtained from shops or chemists), (ii) formal health sector (which included private clinics, and the public government and mission dispensaries, health centres and hospitals), (iii) home remedies. Differences between age-groups in the proportions visiting the formal health sector, receiving SP and seeking treatment within the first day of fever were tested using χ2 (SPSS version 9.0 for Windows, Chicago, IL 60606, USA).
2. Methods The study area and population have been described in detail elsewhere (Guyatt et al., 2002a,b). Briefly, Gucha district is in the western highland belt of Kenya with a predominantly rural subsistence farming population. The clinical patterns of disease in this community reflect those of areas of low, stable endemic risks for malaria, with acute seasonal distributions coincidental with the rains (Hay et al., 2002). Infection prevalence of malaria among children (less than 15 years of age) residing in the study area was 12.5% (Guyatt et al., 2002b). Homesteads in the area were visited in July 2000 (normally the peak of the transmission season) as part of a larger survey evaluating the effectiveness of insecticide-treated bednets and indoor residual house-spraying. As part of this survey, information was collected on treatment seeking behaviour for fevers. An event analysis approach was employed, in which questions focused on a recent fever episode (Riberera in the Kigusii language) in a sample of 1770 individuals. These were selected at random across three age-groups (6 months— 4 years, 5—14 years and 15 years or over) from 590 homesteads. Individuals were asked to recall whether they had had a fever in the last 14 days, and the treatment-seeking behaviour associated with the most recent fever was recorded. Informa-
3. Results Of the 1770 individuals examined, 157 (8.9%) reported a fever in the last 14 days, 81 (52%) of which had resolved by the time of interview (Table 1). The period prevalence of reported fever was similar across the three age-groups. Seventy-four percent of those with a fever sought some treatment, with the majority first visiting the informal retail sector. Only 26% of persons with a fever first visited the formal health sector, compared to nearly half of those first visiting shops or chemists. A small percentage (13%) of individuals went on to seek another treatment source, with a third ultimately visiting a formal health care provider. Although children were more likely to have visited the formal health care system than adults (see Table 1), the differences were not significant. Table 2 summarises the drugs administered to the individuals who sought any treatment. Overall a third (33%) of persons with a fever received
Table 1 Period prevalence of fever and sources of treatment in a population of 1,770 in highland Kenya
Number Number Number Number Number Number Number Number a
interviewed about fever with fever in last 14 days (%) with fever on day of survey (%) of fevers accessing treatmenta (%) first visiting informal retail sector (%) first visiting formal health sector (%) with any visit to formal health sector (%) with two treatment sources (%)
Children <5 years
Children 5—14 years
Adults
All ages
499 61 24 46 25 18 19 5
601 42 16 34 22 12 18 9
670 54 36 36 26 10 13 7
1770 157 76 116 73 40 50 21
(12%) (5%) (75%) (41%) (30%) (31%) (8%)
(7%) (3%) (81%) (35%) (29%) (43%) (21%)
Only 3 individuals aged <5 years and with a current fever had sought home remedies.
(8%) (5%) (67%) (48%) (19%) (24%) (13%)
(9%) (4%) (74%) (47%) (26%) (32%) (13%)
Management of fevers in Kenyan children
113
Table 2 Drugs administered for the treatment of fever episodes in a highland Kenyan population Children <5 years
Children 5—14 years
Adults
All ages
By fever episode (% of all fevers) Anti-malarials SP SP within 24 hours Antipyretics onlya
61 19 17 3 23
fevers (31%) (28%) (5%) (38%)
42 18 15 5 15
fevers (43%) (36%) (12%) (36%)
54 15 11 3 19
fevers (36%) (20%) (6%) (35%)
157 fevers 52 (33%) 43 (27%) 11 (7%) 57 (36%)
By visit to health care facility (% of visits) Anti-malarials SP SP within 24 hours Anti-pyretics onlya
20 16 14 2 2
visits (80%) (70%) (10%) (10%)
18 13 12 2 4
visits (72%) (67%) (11%) (22%)
14 11 7 2 2
visits (79%) (50%) (14%) (14%)
52 40 33 6 8
visits (77%) (63%) (12%) (15%)
By visit to informal retail sector (% of visits) 28 visits 25 visits 29 visits 82 visits Anti-malarials 4 (14%) 7 (28%) 7 (24%) 18 (22%) SP 3 (11%) 3 (12%) 4 (14%) 10 (12%) SP within 24 hours 1 (4%) 3 (12%) 1 (3%) 5 (6%) Antipyretics onlya 24 (86%) 18 (72%) 21 (72%) 63 (77%) a These included paracetamol (mostly brand name Panadol® ), aspirin and ibuprofen. One child <5 years received paracetamol as a home remedy.
an anti-malarial drug and 27% received SP (the nationally recommended first line treatment for malaria). However, only 7% had received SP within 24 hours of onset of the fever. Although children tended to receive SP more often than adults the differences were not significant. Most visits to the formal health care system resulted in treatment with an anti-malarial (77%), primarily SP. In contrast only 12% of visits to shops or pharmacies resulted in treatment with this drug. Overall, more than a third of persons with a fever (36%) were only treated with an antipyretic (mostly Panadol® ), and most of these were taken after visiting the informal retail sector (Table 2).
The first action taken for a fever among all patients was completed within 96 hours, with 43% seeking treatment within the first 24 hours and 44% within the next 24 hours (Table 3). Most individuals with a fever waited until the second day of fever before visiting a formal health care facility or receiving SP. There were no significant differences in the timings of the first treatment by age.
4. Discussion This study confirms previous observations in Kenya that use of the informal retail sector for treating
Table 3 Timings of treatments for 116 fever episodes in a highland Kenyan population Children <5 years
Children 5—14 years
Adults
Days between onset of fever and first treatment (% of those seeking 0 17 (37%) 17 (50%) 16 (44%) 1 25 (54%) 11 (32%) 15 (42%) 2+ 4 (9%) 6 (18%) 5 (14%)
All ages treatment) 50 (43%) 51 (44%) 15 (13%)
Days between onset of fever and first visit to health facility (% of those visiting health facilities) 0 2 (11%) 4 (22%) 4 (31%) 10 (20%) 1 15 (79%) 9 (50%) 5 (38%) 29 (58%) 2+ 2 (10%) 5 (28%) 4 (31%) 11 (22%) Days between onset of fever and treatment with SP (% of those treated with SP) 0 3 (18%) 5 (33%) 3 (27%) 11 (26%) 1 13 (76%) 6 (40%) 4 (36%) 23 (53%) 2+ 1 (6%) 4 (27%) 4 (36%) 9 (21%)
114 fevers is common (Hamel et al., 2001; Molyneux et al., 1999; Mwabu, 1986; Mwenesi et al., 1995; Ruebush et al., 1995; Snow et al., 1992), and that few fevers are treated within 24 hours of onset with SP, the nationally recommended first line drug for the management of uncomplicated malaria (Amin et al., in press). However, unlike previous studies that have focused only on young children (aged less than 5 years), this study investigated actions for fevers across all ages. It was found that only 7% of persons with a fever received SP in the first 24 hours, though 27% received this first-line drug within the first four days of fever. However, the primary observation, with important practical implications for malaria case-management in this area of Kenya, was that the 14-day prevalence of reported ‘‘fever’’, and the actions taken to manage this event, were similar for all age groups. Most importantly, they were not significantly different for the most biologically at-risk groups (young children) compared to least at-risk populations (adults). Of those receiving SP, 60% are older children or adults. The net result is that the larger doses (and hence costs) are used to presumptively manage malaria in patient groups least likely to benefit from therapeutics. Assuming children aged 1—4 years receive a third of an adult dose, and those aged 5—14 years three-quarters, adults and older children are estimated to have consumed 80% of the SP in this study population. Fever is a vague term used by many communities to describe a generalised malaise and has a poor sensitivity and specificity when compared to subsequent parasitologically confirmed clinical diagnoses of malaria (McCombie, 2002). Nevertheless, fever serves as the prompt for malaria action at both the household level for self-medication and at clinic settings where diagnosis is made presumptively upon history of fever. A recent modelled approach to estimating the potential requirements for artemisinin-based combination therapy in Africa has highlighted the potential commodity cost implications of malaria over-diagnosis among older children and adults at least risk of poor clinical outcomes (Snow et al., 2003). However, our understanding of drug use among older children and adults continues to remain poorly defined, thus hampering efforts to quantify drug needs both for communities and health care facilities. Our findings confirm impressions of drug prescription practices among government formal health services that the diagnosis of malaria is as common in adult patient groups as in paediatric patient groups. The mismatch between clinical risk and drug use might be acceptable where drugs used to manage ‘‘fever’’
H.L. Guyatt, R.W. Snow are safe and cheap, and do not mask the effective management of other underlying and potentially fatal conditions. It seems likely that replacements for failing monotherapies in East Africa will have less favourable toxicity margins and will be by no means as cheap. The effective targeting of new malaria case-management will require commensurate efforts to improve upon community and service provider (informal and formal) diagnostic capabilities and awareness of clinical risk. Limited household and government financial resources to purchase drugs must not result in those most able to articulate demand receiving preferential access to better drugs over those most biologically vulnerable.
Acknowledgements The authors are grateful to Wellcome Trust, UK and KEMRI for financial support. Helen Guyatt and Bob Snow are supported by the Wellcome Trust Career Development Fellowship (055100) and Senior Fellowship (058992) programmes respectively. We would wish to thank the people of Nyamache division who participated in the study and the local Ministry of Health and Government Civil Administration staff in Gucha for their support in facilitating the surveys. This paper is published with the permission of the director for the Kenya Medical Research Institute.
References Amin, A.A., Marsh, V., Noor, A.M., Ochola, S.A., Snow, R.W., The use of formal and informal curative services in the management of paediatric fevers in four districts in Kenya. Trop. Med. Int. Health (in press). Chandramohan, D., Jaffar, S., Greenwood, B., 2002. Use of clinical algorithms for diagnosing malaria. Trop. Med. Int. Health 7, 45—52. Guyatt, H.L., Kinnear, J., Burini, M., Snow, R.W., 2002a. A cost analysis of an emergency relief agency programme of vector control in highland Kenya. Health Policy Plann. 17, 144— 153. Guyatt, H.L., Corelett, S., Robinson, T.P., Ochola, S.A., Snow, R.W., 2002b. Malaria prevention in highland Kenya. Indoor residual house-spraying vs. insecticide-treated bednets. Trop. Med. Int. Health 7, 298—303. Hamel, M.J., Odhacha, A., Roberts, J.M., Deming, M.S., 2001. Malaria control in Bungoma district, Kenya: a survey of home treatment of children with fever, bednet use and attendance at antenatal clinics. Bull. World Health Organ. 79, 1014— 1023. Hay, S.I., Noor, A.M., Simba, M., Busulo, M., Guyatt, H., Ochola, S.A., Snow, R.W., 2002. The clinical epidemiology of malaria in the highlands of western Kenya. Emerg. Infect. Dis. 8, 619—624.
Management of fevers in Kenyan children McCombie, S.C., 2002. Self-treatment for malaria: the evidence and methodological issues. Health Policy Plann. 17, 333— 344. Molyneux, C.S., Mung’ala-Odera, V., Harpham, T., Snow, R.W., 1999. Maternal responses to childhood fevers: a comparison of rural and urban residents in coastal Kenya. Trop. Med. Int. Health 4, 836—845. Mwabu, G.M., 1986. Health care decisions at the household level: results of a rural health survey in Kenya. Soc. Sci. Med. 22, 315—319. Mwenesi, H., Harpham, T., Snow, R.W., 1995. Child malaria treatment among mothers in Kenya. Soc. Sci. Med. 49, 1271— 1277. Ruebush, T.K., Kern, M.K., Campbell, C.C., Oloo, A.J., 1995. Self-treatment of malaria in a rural area of western Kenya. Bull. World Health Organ. 73, 229—236.
115 Snow, R.W., Peshu, N., Forster, D., Mwenesi, H., Marsh, K., 1992. The role of shops in the treatment and prevention of childhood malaria on the coast of Kenya. Trans. R. Soc. Trop. Med. Hyg. 86, 237—239. Snow, R.W., Craig, M., Deichmann, U., Marsh, K., 1999. Estimating mortality, morbidity and disability due to malaria among Africa’s non-pregnant population. Bull. World Health Organ. 77, 624—640. Snow, R.W., Eckert, E., Teklehaimanot, A., 2003. Estimating the needs for artesunate-based combination therapy for malaria case-management in Africa. Trends Parasitol. 19, 363—369. WHO, 2000. The Abuja Declaration on Roll Back Malaria in Africa by the African Heads of States and Governments, 25th April 2000, Abuja, Nigeria. Statement prepared by RBM, WHO, Geneva, April 2000.
The management of fevers in Kenyan children and adults in an area of seasonal malaria transmission Helen L. Guyatt*, Robert W. Snow KEMRI-Wellcome Trust Collaborative Programme, GPO 00100, Nairobi, PO Box 43640, Kenya and Centre for Tropical Medicine, University of Oxford, John Radcliffe Hospital, Oxford OX3 9DU, UK Received 10 April 2003 ; received in revised form 31 July 2003; accepted 1 August 2003
KEYWORDS Malaria; Fevers; Treatment; Self-medication; Adults; Sulfadoxinepyrimethamine; Kenya
Summary This study investigates the source, timing and types of treatment for fevers across all ages in a low malaria-transmission area of Kenya. The period prevalence for fever, and subsequent treatment seeking behaviour, was similar across all ages. The use of the informal retail sector was common (47% of first actions), though most visits to shops and chemists (77%) resulted in treatment with an antipyretic not an antimalarial. The major source of the first line recommended drug, sulfadoxine-pyrimethamine (SP), was at the formal health sector, and 32% of fevers made at least one visit to a health care facility. Although only 7% of fevers received SP within 24 hours of fever onset, 27% ultimately received treatment with this antimalarial. It is estimated that of the total amount of SP consumed in this population, only 20% is administered to children less than 5 years old. In this area of Kenya disease risks decline with increasing age, however, adult populations consume over 40% of prescribed or purchased anti-malarial drugs. In light of the proposed new, more costly anti-malarial drug combinations these findings have major implications for the effective allocation of limited financial resources at household and government levels. © 2003 Royal Society of Tropical Medicine and Hygiene. Published by Elsevier Ltd. All rights reserved.
1. Introduction Despite renewed efforts aimed at disease prevention through insecticide-treated bednets, prompt, effective treatment of clinical cases will continue to be a major cornerstone of efforts to Roll Back Malaria (RBM) in Africa. In most settings this relies upon the presumptive treatment of fever (Chandramohan et al., 2002). Our current un*Corresponding author. Present address: c/o Dr. T.P. Robinson, FAO (AGAL), Viale delle Terme di Caracalla, 00100 Rome, Italy. Tel./fax: +39-0763-308-246. E-mail address: [email protected] (H.L. Guyatt).
derstanding of how communities manage fevers derives from investigations of treatment-seeking behaviour among children aged less than 5 years old. This seems reasonable as paediatric populations bear over 75% of the disease burden in Africa (Snow et al., 1999). As such the RBM partnership has proposed that a measurable and appropriate target in Africa should be that 60% of childhood fevers are treated with an appropriate drug within 24 hours of onset (WHO, 2000). Consequently very little is known about the management of fevers amongst older children and adult populations, who are at least risk of poor clinical outcomes following infection, but continue to presumptively seek
0035-9203/$ — see front matter © 2003 Royal Society of Tropical Medicine and Hygiene. Published by Elsevier Ltd. All rights reserved. doi:10.1016/S0035-9203(03)00016-6
112
H.L. Guyatt, R.W. Snow
treatment for malaria. This is of particular relevance to understanding the drug requirements in populations, against ever declining drug budgets and the assumed resistance pressure posed by drug use at community-levels. The current study assessed the treatment seeking behaviour for fevers across all ages in an area of low, seasonal malaria transmission in the western highlands of Kenya.
tion was obtained on the exact source, timing and treatment obtained. All fevers were based only on the patients’ description (no temperatures were measured), and the word for fever is distinct from malaria (Esosera in Kigusii). Sources of treatment were classified as (i) informal retail sector (self-treatment with western-style medications obtained from shops or chemists), (ii) formal health sector (which included private clinics, and the public government and mission dispensaries, health centres and hospitals), (iii) home remedies. Differences between age-groups in the proportions visiting the formal health sector, receiving SP and seeking treatment within the first day of fever were tested using χ2 (SPSS version 9.0 for Windows, Chicago, IL 60606, USA).
2. Methods The study area and population have been described in detail elsewhere (Guyatt et al., 2002a,b). Briefly, Gucha district is in the western highland belt of Kenya with a predominantly rural subsistence farming population. The clinical patterns of disease in this community reflect those of areas of low, stable endemic risks for malaria, with acute seasonal distributions coincidental with the rains (Hay et al., 2002). Infection prevalence of malaria among children (less than 15 years of age) residing in the study area was 12.5% (Guyatt et al., 2002b). Homesteads in the area were visited in July 2000 (normally the peak of the transmission season) as part of a larger survey evaluating the effectiveness of insecticide-treated bednets and indoor residual house-spraying. As part of this survey, information was collected on treatment seeking behaviour for fevers. An event analysis approach was employed, in which questions focused on a recent fever episode (Riberera in the Kigusii language) in a sample of 1770 individuals. These were selected at random across three age-groups (6 months— 4 years, 5—14 years and 15 years or over) from 590 homesteads. Individuals were asked to recall whether they had had a fever in the last 14 days, and the treatment-seeking behaviour associated with the most recent fever was recorded. Informa-
3. Results Of the 1770 individuals examined, 157 (8.9%) reported a fever in the last 14 days, 81 (52%) of which had resolved by the time of interview (Table 1). The period prevalence of reported fever was similar across the three age-groups. Seventy-four percent of those with a fever sought some treatment, with the majority first visiting the informal retail sector. Only 26% of persons with a fever first visited the formal health sector, compared to nearly half of those first visiting shops or chemists. A small percentage (13%) of individuals went on to seek another treatment source, with a third ultimately visiting a formal health care provider. Although children were more likely to have visited the formal health care system than adults (see Table 1), the differences were not significant. Table 2 summarises the drugs administered to the individuals who sought any treatment. Overall a third (33%) of persons with a fever received
Table 1 Period prevalence of fever and sources of treatment in a population of 1,770 in highland Kenya
Number Number Number Number Number Number Number Number a
interviewed about fever with fever in last 14 days (%) with fever on day of survey (%) of fevers accessing treatmenta (%) first visiting informal retail sector (%) first visiting formal health sector (%) with any visit to formal health sector (%) with two treatment sources (%)
Children <5 years
Children 5—14 years
Adults
All ages
499 61 24 46 25 18 19 5
601 42 16 34 22 12 18 9
670 54 36 36 26 10 13 7
1770 157 76 116 73 40 50 21
(12%) (5%) (75%) (41%) (30%) (31%) (8%)
(7%) (3%) (81%) (35%) (29%) (43%) (21%)
Only 3 individuals aged <5 years and with a current fever had sought home remedies.
(8%) (5%) (67%) (48%) (19%) (24%) (13%)
(9%) (4%) (74%) (47%) (26%) (32%) (13%)
Management of fevers in Kenyan children
113
Table 2 Drugs administered for the treatment of fever episodes in a highland Kenyan population Children <5 years
Children 5—14 years
Adults
All ages
By fever episode (% of all fevers) Anti-malarials SP SP within 24 hours Antipyretics onlya
61 19 17 3 23
fevers (31%) (28%) (5%) (38%)
42 18 15 5 15
fevers (43%) (36%) (12%) (36%)
54 15 11 3 19
fevers (36%) (20%) (6%) (35%)
157 fevers 52 (33%) 43 (27%) 11 (7%) 57 (36%)
By visit to health care facility (% of visits) Anti-malarials SP SP within 24 hours Anti-pyretics onlya
20 16 14 2 2
visits (80%) (70%) (10%) (10%)
18 13 12 2 4
visits (72%) (67%) (11%) (22%)
14 11 7 2 2
visits (79%) (50%) (14%) (14%)
52 40 33 6 8
visits (77%) (63%) (12%) (15%)
By visit to informal retail sector (% of visits) 28 visits 25 visits 29 visits 82 visits Anti-malarials 4 (14%) 7 (28%) 7 (24%) 18 (22%) SP 3 (11%) 3 (12%) 4 (14%) 10 (12%) SP within 24 hours 1 (4%) 3 (12%) 1 (3%) 5 (6%) Antipyretics onlya 24 (86%) 18 (72%) 21 (72%) 63 (77%) a These included paracetamol (mostly brand name Panadol® ), aspirin and ibuprofen. One child <5 years received paracetamol as a home remedy.
an anti-malarial drug and 27% received SP (the nationally recommended first line treatment for malaria). However, only 7% had received SP within 24 hours of onset of the fever. Although children tended to receive SP more often than adults the differences were not significant. Most visits to the formal health care system resulted in treatment with an anti-malarial (77%), primarily SP. In contrast only 12% of visits to shops or pharmacies resulted in treatment with this drug. Overall, more than a third of persons with a fever (36%) were only treated with an antipyretic (mostly Panadol® ), and most of these were taken after visiting the informal retail sector (Table 2).
The first action taken for a fever among all patients was completed within 96 hours, with 43% seeking treatment within the first 24 hours and 44% within the next 24 hours (Table 3). Most individuals with a fever waited until the second day of fever before visiting a formal health care facility or receiving SP. There were no significant differences in the timings of the first treatment by age.
4. Discussion This study confirms previous observations in Kenya that use of the informal retail sector for treating
Table 3 Timings of treatments for 116 fever episodes in a highland Kenyan population Children <5 years
Children 5—14 years
Adults
Days between onset of fever and first treatment (% of those seeking 0 17 (37%) 17 (50%) 16 (44%) 1 25 (54%) 11 (32%) 15 (42%) 2+ 4 (9%) 6 (18%) 5 (14%)
All ages treatment) 50 (43%) 51 (44%) 15 (13%)
Days between onset of fever and first visit to health facility (% of those visiting health facilities) 0 2 (11%) 4 (22%) 4 (31%) 10 (20%) 1 15 (79%) 9 (50%) 5 (38%) 29 (58%) 2+ 2 (10%) 5 (28%) 4 (31%) 11 (22%) Days between onset of fever and treatment with SP (% of those treated with SP) 0 3 (18%) 5 (33%) 3 (27%) 11 (26%) 1 13 (76%) 6 (40%) 4 (36%) 23 (53%) 2+ 1 (6%) 4 (27%) 4 (36%) 9 (21%)
114 fevers is common (Hamel et al., 2001; Molyneux et al., 1999; Mwabu, 1986; Mwenesi et al., 1995; Ruebush et al., 1995; Snow et al., 1992), and that few fevers are treated within 24 hours of onset with SP, the nationally recommended first line drug for the management of uncomplicated malaria (Amin et al., in press). However, unlike previous studies that have focused only on young children (aged less than 5 years), this study investigated actions for fevers across all ages. It was found that only 7% of persons with a fever received SP in the first 24 hours, though 27% received this first-line drug within the first four days of fever. However, the primary observation, with important practical implications for malaria case-management in this area of Kenya, was that the 14-day prevalence of reported ‘‘fever’’, and the actions taken to manage this event, were similar for all age groups. Most importantly, they were not significantly different for the most biologically at-risk groups (young children) compared to least at-risk populations (adults). Of those receiving SP, 60% are older children or adults. The net result is that the larger doses (and hence costs) are used to presumptively manage malaria in patient groups least likely to benefit from therapeutics. Assuming children aged 1—4 years receive a third of an adult dose, and those aged 5—14 years three-quarters, adults and older children are estimated to have consumed 80% of the SP in this study population. Fever is a vague term used by many communities to describe a generalised malaise and has a poor sensitivity and specificity when compared to subsequent parasitologically confirmed clinical diagnoses of malaria (McCombie, 2002). Nevertheless, fever serves as the prompt for malaria action at both the household level for self-medication and at clinic settings where diagnosis is made presumptively upon history of fever. A recent modelled approach to estimating the potential requirements for artemisinin-based combination therapy in Africa has highlighted the potential commodity cost implications of malaria over-diagnosis among older children and adults at least risk of poor clinical outcomes (Snow et al., 2003). However, our understanding of drug use among older children and adults continues to remain poorly defined, thus hampering efforts to quantify drug needs both for communities and health care facilities. Our findings confirm impressions of drug prescription practices among government formal health services that the diagnosis of malaria is as common in adult patient groups as in paediatric patient groups. The mismatch between clinical risk and drug use might be acceptable where drugs used to manage ‘‘fever’’
H.L. Guyatt, R.W. Snow are safe and cheap, and do not mask the effective management of other underlying and potentially fatal conditions. It seems likely that replacements for failing monotherapies in East Africa will have less favourable toxicity margins and will be by no means as cheap. The effective targeting of new malaria case-management will require commensurate efforts to improve upon community and service provider (informal and formal) diagnostic capabilities and awareness of clinical risk. Limited household and government financial resources to purchase drugs must not result in those most able to articulate demand receiving preferential access to better drugs over those most biologically vulnerable.
Acknowledgements The authors are grateful to Wellcome Trust, UK and KEMRI for financial support. Helen Guyatt and Bob Snow are supported by the Wellcome Trust Career Development Fellowship (055100) and Senior Fellowship (058992) programmes respectively. We would wish to thank the people of Nyamache division who participated in the study and the local Ministry of Health and Government Civil Administration staff in Gucha for their support in facilitating the surveys. This paper is published with the permission of the director for the Kenya Medical Research Institute.
References Amin, A.A., Marsh, V., Noor, A.M., Ochola, S.A., Snow, R.W., The use of formal and informal curative services in the management of paediatric fevers in four districts in Kenya. Trop. Med. Int. Health (in press). Chandramohan, D., Jaffar, S., Greenwood, B., 2002. Use of clinical algorithms for diagnosing malaria. Trop. Med. Int. Health 7, 45—52. Guyatt, H.L., Kinnear, J., Burini, M., Snow, R.W., 2002a. A cost analysis of an emergency relief agency programme of vector control in highland Kenya. Health Policy Plann. 17, 144— 153. Guyatt, H.L., Corelett, S., Robinson, T.P., Ochola, S.A., Snow, R.W., 2002b. Malaria prevention in highland Kenya. Indoor residual house-spraying vs. insecticide-treated bednets. Trop. Med. Int. Health 7, 298—303. Hamel, M.J., Odhacha, A., Roberts, J.M., Deming, M.S., 2001. Malaria control in Bungoma district, Kenya: a survey of home treatment of children with fever, bednet use and attendance at antenatal clinics. Bull. World Health Organ. 79, 1014— 1023. Hay, S.I., Noor, A.M., Simba, M., Busulo, M., Guyatt, H., Ochola, S.A., Snow, R.W., 2002. The clinical epidemiology of malaria in the highlands of western Kenya. Emerg. Infect. Dis. 8, 619—624.
Management of fevers in Kenyan children McCombie, S.C., 2002. Self-treatment for malaria: the evidence and methodological issues. Health Policy Plann. 17, 333— 344. Molyneux, C.S., Mung’ala-Odera, V., Harpham, T., Snow, R.W., 1999. Maternal responses to childhood fevers: a comparison of rural and urban residents in coastal Kenya. Trop. Med. Int. Health 4, 836—845. Mwabu, G.M., 1986. Health care decisions at the household level: results of a rural health survey in Kenya. Soc. Sci. Med. 22, 315—319. Mwenesi, H., Harpham, T., Snow, R.W., 1995. Child malaria treatment among mothers in Kenya. Soc. Sci. Med. 49, 1271— 1277. Ruebush, T.K., Kern, M.K., Campbell, C.C., Oloo, A.J., 1995. Self-treatment of malaria in a rural area of western Kenya. Bull. World Health Organ. 73, 229—236.
115 Snow, R.W., Peshu, N., Forster, D., Mwenesi, H., Marsh, K., 1992. The role of shops in the treatment and prevention of childhood malaria on the coast of Kenya. Trans. R. Soc. Trop. Med. Hyg. 86, 237—239. Snow, R.W., Craig, M., Deichmann, U., Marsh, K., 1999. Estimating mortality, morbidity and disability due to malaria among Africa’s non-pregnant population. Bull. World Health Organ. 77, 624—640. Snow, R.W., Eckert, E., Teklehaimanot, A., 2003. Estimating the needs for artesunate-based combination therapy for malaria case-management in Africa. Trends Parasitol. 19, 363—369. WHO, 2000. The Abuja Declaration on Roll Back Malaria in Africa by the African Heads of States and Governments, 25th April 2000, Abuja, Nigeria. Statement prepared by RBM, WHO, Geneva, April 2000.