- Email: [email protected]
guardians and health workers on childhood malaria in a Tanzanian district—implications for malaria control
Acta Tropica 87 (2003) 305 /313 www.elsevier.com/locate/actatropica
Local understanding, perceptions and reported practices of mothers/guardians and health workers on childhood malaria in a Tanzanian district* implications for malaria control /
C. Comoro a, S.E.D. Nsimba b,c,*, M. Warsame c, G. Tomson c,d a
Department of Sociology, University of Dar-es-Salaam, Dar-es-Salaam, Tanzania Department of Clinical Pharmacology, Muhimbili University College of Health Sciences, Dar-es-Salaam, Tanzania c Division of International Health (IHCAR), Department of Public Health Sciences, Karolinska Institute, S-171 76, Stockholm, Sweden d Medical Management Centre, Karolinska Institute, Stockholm, Sweden b
Received 21 December 2001; received in revised form 4 December 2002; accepted 20 January 2003
Abstract Knowledge on local understanding, perceptions and practices of care providers regarding management of childhood malaria are needed for better malaria control in urban, peri-urban and rural communities. Mothers of under five children attending five purposively selected public health facilities in the Kibaha district, Tanzania, were invited to participate in 10 focus group discussions (FGDs). The health workers of these facilities were included in six other FGDs to elicit their professional views. Analysis was done using interpretative and qualitative approaches. Both health workers and all mothers were clear about the signs and symptoms of homa ya malaria , a description consistent with the biomedical definition of mild malaria. Although most of the mothers related this to mosquito bites, some did not. Mothers also described a severe childhood illness called degedege , consistent with convulsions. Most of the mothers failed to associate this condition with malaria, believing it is caused by evil spirits. Urinating on or fuming the child suffering from degedege with elephant dung were perceived to be effective remedies while injections were considered fatal for such condition. Traditional healers were seen as the primary source of treatment outside homes for this condition and grandmothers and mother in-laws are the key decision makers in the management. Our findings revealed major gaps in managing severe malaria in the study communities. Interventions addressing these gaps and targeting mothers/guardians, mother in-laws, grandmothers and traditional healers are needed. # 2003 Published by Elsevier B.V. Keywords: Focus groups discussions; Mothers; Under-fives; Malaria; Tanzania; Traditional healers; Degedege
1. Introduction
* Corresponding author. Tel.: /46-8-517-79895; fax: /46-8311590. E-mail address: [email protected] (S.E.D. Nsimba). 0001-706X/03/$ - see front matter # 2003 Published by Elsevier B.V. doi:10.1016/S0001-706X(03)00113-X
Malaria is the leading cause of morbidity and mortality in children in Tanzania (Kilama and Kihamia, 1991). Adequate case management*/ early recognition and prompt treatment */has
306
C. Comoro et al. / Acta Tropica 87 (2003) 305 /313
been and still continues to be the basis for malaria control here as in most of Sub-Saharan Africa (National Malaria Control Programme, 1990; WHO, 1993). Formal health service has been the primary focus for the National Malaria Control Programme in the country. However, surveys carried out in some African countries revealed that malaria in children is first recognised and responded to at home (Mwenesi et al., 1995a; McCombie, 1996). This is further supported by the reports of frequent self-treatment with antimalarials stocked at home and/or bought at drug shops (Massele et al., 1993; Mnyika et al., 1995; Nsimba et al., 1999). Understanding local knowledge, perceptions and practices of malaria management has become the focus of research during the last decade. Studies from Kenya and Ghana reported that mothers had good knowledge in recognising symptoms suggestive of malaria (Ruebush et al., 1995; Ahorlu et al., 1997). However, surveys from Tanzania and Nigeria reported the contrary (Rooth and Bjo¨rkman, 1992; Molineaux and Gramiccia, 1980). Other studies revealed that convulsions, a prominent condition in severe malaria, were not associated with malaria by caretakers but rather to supernatural cause (Mwenesi et al., 1995b; Winch et al., 1995; Ahorlu et al., 1997). Traditional healers were reported to be the focal point for consultation. As part of a malaria project aiming to develop context specific interventions to improve the quality of malaria case management, we investigated the local understanding, perceptions and practices of caretakers on management of childhood malaria in urban, peri-urban and rural settings.
2. Material and methods 2.1. Study area and population The study was conducted between July and September 1999 in the Kibaha district located 40 kms north west of Dar-es-Salaam. The area is highly endemic for malaria (Nsimba et al., 1999)
and is populated by the Zaramo, Doe, Kwavi and Ndengeleko tribes. In order to assess the knowledge, perceptions and practices about childhood malaria, mothers with sick children attending five purposively selected health facilities were targeted. The district had in total 10 public health facilities serving urban (n /3), peri-urban (n /1) and rural (n /6) communities. We wanted to include facilities from all the different geographical areas. Thus we selected two facilities in the urban (Mkoani health centre and Mwendapole dispensary), the periurban (Mlandizi health centre) and two in the rural (Ruvu Stesheni and Magindu dispensaries). We chose the two most interior of the rural health facilities. In each health facility, mothers with sick underfives were randomly selected by asking each of them to pick a number from a basket and those who picked even numbers were again purposively selected. Those mothers who were finally selected were informed about the study and were asked if they had time to participate in the focus group discussions (FGDs). Everybody consented verbally. Mothers with severely ill children were excluded. Discussions were also held with the health workers in the selected five facilities in order to elicit a professional viewpoint of the problem. 2.2. Data collection We used FGDs as a method and its thematic guides were translated into Swahili language and were pre-tested in similar facilities in Dar-esSalaam. Our selection to use this method was based on its comparative advantage over individual interviews. These include group interaction of participants which is lacking in interviews, it is relatively cheap as it can be conducted in a short span of time by small number of staff with limited financial resources, it provides an opportunity for researchers to meet and talk to many people in one session. Furthermore, the method allows researchers to get ideas about peoples experience, opinions, beliefs and participants’ responses can be presented with actual quotations that helps the reader to get the main ideas or messages.
C. Comoro et al. / Acta Tropica 87 (2003) 305 /313
The focus of the inquiry was on mothers’ knowledge and perceptions on major health problem in their community, malaria seasonality, signs and symptoms of childhood malaria and treatment practices. The FGD sessions were moderated by the principal investigator (first author) and assisted by the second author. Each session lasted between 1 h and 15 min to 1 h and 45 min. Note taking was done by two graduate social scientists in their final year master of arts in Sociology at the University of Dar-es-Salaam. The notes were translated into the English language by the first and second authors, respectively. Discussions with health workers at each facility were done either prior to or after their working hours. A total of 16 FGDs were held, 10 (two per facility) consisting of 7 /10 participants per group session were held with mothers, the remaining six FGDs were held with the health workers (5 /9 participants per group). One FGD was held at each of the three dispensaries as planned, three FGDs with health care providers instead of four were held at health centres. In one of the health centres, Mkoani, part of the health staff were out of station for outreach activities and thus we managed to get only one FGD group. The FGDs for mothers and health care providers were conducted separately at each facility. The FGDs with mothers were done before those of health workers, and prior to the consultations with the health workers. 2.3. Data analysis Data analysis was done using the interpretative understanding approach (Moser and Kalton, 1971). This analysis involves making sense of the collected information through eliciting meanings (actual words used by participants) or responses that people gave to different situations which are important (e.g. what the word ‘degedeg e’ means to them). A code sheet was created following the focus group guide and data were coded. Then a master sheet was done giving listing of all the responses from focus groups. These responses were interpreted by looking at their patterns. Sorting and sifting through information, looking for patterns, consensus, differences, variations or
307
contradictions, and weighing the relative importance of information complimented the interpretative understanding. Ranking and tabulating of the data was also done.
3. Ethical approval The study was approved by the Muhimbili University College of Health Sciences (MUCHS) Human Ethics Committee, Tanzania. Permission were obtained from the regional and district administrative authorities. Informed consent was also obtained from the mothers and health workers in the studied facilities. Mothers with severely ill children were not included in the study. Discussions with the health workers were held after their consultation hours.
4. Results Demographic characteristics of the participants are presented in Table 1. All but one of participants accompanying children under 5 years were mothers. The remaining one was a female guardian of a child. The majority of the mothers aged between 21 /30 years. Most described their occupation either as housewives (32%) or peasants (49%). The rest were petty traders or civil servants. The age profile of the health workers ranged from 20s to 40s. They were 4 Clinical Officers, 6 Table 1 Socio-demographic profile of participants in the 16 FGDs Characteristics
Mothers Health care providers
No. participants Females (%)
85 100
43 58
20 74 3.5 2.3
19 81 / /
8.2 87.1 4.7
7 84 9
Marital status Single (%) Married (%) Cohabiting (%) Widowed (%) Education No formal education (%) Primary education (%) Secondary education (%)
308
C. Comoro et al. / Acta Tropica 87 (2003) 305 /313
Assistant Clinical Officers, 6 Trained Nurses, 11 Nurse Assistants, 12 MCH Aides and 4 Traditional Birth Attendants. Most health workers had primary education with some basic medical training. All but three of the 16 FGDs were unanimous that malaria attacks were the main health problems in their communities. The remaining three who were from mothers (two groups from Mlandizi, one from Ruvu Stesheni) considered cholera and gastrointestinal problems (vomiting and diarrhoea) as the number one health problem and ranked malaria second. Children were recognised as being the most vulnerable group to malaria. Pregnant women, visitors and immigrants were also considered by some of the participants to be at special risk. When participants were asked to identify season(s) associated with peak of malaria in their communities, nominal seasons as opposed to lunar calendar were much more familiar to both categories (mothers and health workers). Most of the participants associated rainy and cold seasons with peak of malaria. The two categories of participants (mothers and health workers) provided similar descriptions of malaria fever ‘homa ya malaria ’ in children. All associated symptoms of malaria with hot body/ high temperature, vomiting, loss of appetite, diarrhoea, yellow eyes and crying. The health workers, in addition to the above, also mentioned anaemia and cough. Seven of the ten mothers’ FGDs related malaria to mosquito bites as all the FGDs with health workers did. The cause of malaria was unknown to the remaining mother groups. The Mlandizi mothers blamed population growth and multiplication of pit latrines as attractive breeding grounds for mosquitoes. The mothers had a special explanation for conditions compatible with severe malaria, which is a febrile condition that attacks many children locally called degedege. Some of the mothers avoided even mentioning it because it was believed to be a bad omen. Thus, they simply referred it as ‘childhood disease’ and gave a very elaborative description of the condition. High body temperature and the child’s body behaviour changes (eye rolling, jolting, quavering, and stiffening, foaming in the mouth) featured prominently in the descrip-
Table 2 Mothers’ description of degedege according to the participants in the 10 FGDs with mothers FGD groups
Description of degedege
Mkoani (urban)
FGD 1 FGD 2
Excessive high temperature, body jolts, rolling of the eyes High fever, spasm of body jolt or fits, mouth foaming
Mwendapole (urban)
FGD 3 FGD 4
Turning of eyes, weakening of the limbs Fits or body jolts, shouting, heavy vibration of head veins
Mlandizi (periurban)
FGD 5 FGD 6
Body jolts, fits, shouting, head veins vibrates and high temperature Popping and twisting of the eyes, stiffening of legs, high temperature, crying continuously
Ruvu Stesheni (rural)
FGD 7
Eyes turning, shivering of the body and inability to walk, breathing through the ribs High stomach temperature, body jolts and stiffening
FGD 8 Magindu (rural)
FGD 9 FGD 10
High fever, heavy breathing, cold feet, warm upper body, eye twisting High fever, heavy breathing, cold feet, warm body, stiffening of the limbs
tion of signs and symptoms of degedege (Table 2). The health care providers gave similar description of the condition and participants in one group stressed on the notion of excessive temperature in the brain and convulsion. Anaemia was noted as an important severe childhood illness by health workers, a notion that none of the mothers’ group had. With regards to mothers’ perceived causes of degedege , three views emerged from the discussion, the dominating one being that of evil spirits ‘shetan i’, as stated by participants in six of the mothers’ FGDs. Only two FGDs from the mothers group, one from urban and one from peri-urban settings, mentioned malaria as the cause. In contrast, all the health care providers (six FGDs) associated convulsions with malaria (Table 3). The first line of action in responding to malaria fever across the 16 FGDs was giving antipyretics
C. Comoro et al. / Acta Tropica 87 (2003) 305 /313
Table 3 Health workers’ description of degedege according to participants in the 6 FGDs with health workers FGD groups
Description of degedege
Mkoani (urban) FGD 11 Mwendapole FGD (urban) 12 Mlandizi (peri- FGD urban) 13 FGD 14
Excessive high temperature in the brain, convulsion High temperature or no temperature High temperature, vomiting yellowish Very high temperature, Anaemic child
Ruvu Stesheni (rural)
Severe fever of the stomach, severe fever of the head, diarrhoea, vomiting and loss of appetite Asymptomatic, stiffening, popping eyes, foaming in the mouth
FGD 15
Magindu (rural) FGD 16
and sponging to lower the body temperature. Observing the condition of the child before deciding to take to hospital was frequently mentioned and was commented as follows: ‘If the child brightens/resumes playing she is no longer taken to hospital. If the fever persists or escalates the child is sent to hospital’. All mothers said they kept stock of aspirin or panadol at home. Some said they had chloroquine syrup in stock. When the child gets better, some mothers finish the dose while others discontinue administration and stock the remaining medicine for the ‘anticipated’ next attack. Some mothers expressed their dissatisfaction about the lack of testing for malaria parasites at the primary health care facilities. The health workers of Magindu rural dispensary (FGD 16) raised similar concerns about the lack of diagnostic facilities at primary health care facilities. With regard to children with convulsions, mothers of six FGDs stated that they would take them to a traditional healer (Table 4), in local language fundi or mganga, a Kiswahili word for an expert. It is a local title for a traditional healer along the coastal line and nearby regions in various realms of medicine, which involves divination by a healer to find the causes of illness. Mothers in one FGD from the rural, one from the peri-urban and one from the urban presented conflicting views: some would take their children to hospital, some to the traditional healers and
309
some to both, first to traditional healer then to hospital (Table 4). The later group stated that traditional remedies cool down the problem and the hospital subsequently treats the child completely. Some mothers, who related degedege to malaria as its cause, stated they would seek care from traditional practitioners for such condition. According to the health workers, there are two malaria-related behaviours cutting across the ethnic population. When a child gets the severe malaria attack, the indigenous people (i.e. Zaramo, Kwere, Doe etc.) normally start at the fundi . On contrary terms the immigrants, mostly comprising of the Wakwavi and other up-country people go straight to the hospital with their sick children. According to the mothers, urinating on or fuming the child with elephant dung were the most common traditional remedies for degedege . Furthermore, other treatment options mentioned by mothers include: making the child inhale garlic vapour or smell from a pit latrine. During the treatment of the condition there are several restrictions in force in a household with a suffering child. These include (i) no knock at the door where the sick child is and in particular the formal traditional request of ‘hodi wenyewe’ (i.e. may I come in); (ii) taking fire out of such a house is totally prohibited. (iii) Entering in such a house with a lemon fruit, raw fish or pepper is absolutely disallowed. Concerns about giving injectables to a child with degedege were raised. Mothers reported that, people say and themselves had witnessed several incidences whereby a child with the condition died after getting an injection. However, according to health workers on the other hand, the first response towards the condition was to give antipyretics, anti-convulsants, sponging and chloroquine injections. These health workers, confirmed the prevalence of the traditional practices stated by mothers in the following response ‘A child suffering from degedege must be urinated on and before getting to the hospital s/he has been sent to several traditional healers’. As a whole, degedege is considered by mothers as fatal hence decision making is exercised by important people in the community such as
C. Comoro et al. / Acta Tropica 87 (2003) 305 /313
310
Table 4 Mothers’ perception of the causes of degedege and how they respond to it FGDs Mkoani (urban)
Perceived causes of degedege FGD 1 FGD 2
Actions taken
Excessively high temperature in the brain Take the child to hospital Severe fever related to rapid multiplication of malaria parasites Take the child to traditional healers
Mwendapole (urban) FGD 3 FGD 4
Believed to be due to ‘Shetani’ (evil spirit) Cause unknown but may be due to high temperature
Take the child to traditional healer Give aspirin, chloroquine Take the child traditional healer
Mlandizi (peri-urban)
FGD 5
Due to malaria which has gone into the head
Take the child to hospital
FGD 6
Due to ‘Shetani’ (evil spirit)
Take the child to traditional healer Take the child to traditional healer
Ruvu Stesheni (rural) FGD 7 FGD 8
Cause unknown but may be due to ‘Shetani’ Due to ‘Shetani’
Take the child to traditional healer Take the child to traditional healer
Magindu (rural)
Cause is related to ‘Shetani’
Take the child to traditional healer Few will visit hospital as first resort Take the child to traditional healer
FGD 9
FGD 10 Cause is related to ‘Shetani’
traditional healers, grandmothers and mother-in laws. In this study, these three categories were the immediate referrals for this childhood condition, the same advisors who were strong in forbidding young mothers from taking children with degedege to hospital. For example a common dialogue between young mothers and mothers in-law or grandmothers regarding taking the sick child to hospital reported as follows: Verbal translation: ‘Habali komga huko imwana? Kolonda kumkoma’ ? Translation for the question from grandmothers/mothers-in-law: Why are you taking him/her there? Do you want to kill him/her? Mothers’ response to this: verbal translation; ‘Simkoma bule mie. Nomgala akatibiwe ’ meaning ‘I am not killing him/her. I am taking him/her to hospital for treatment’. The mothers reported that they were discourteously warned by mother-inlaws quoting ‘You are taking the child to the hospital, you are going to cause the child’s death and be responsible for it’. Causing death in this manner is a stigma that no young mother wants to live with. However, if a mother was to take a convulsing child to hospital, the mothers-in-law/ grandmothers simply concludes: ‘Look here, this child is not a woman who you can easily replace with another one’.
5. Discussion Focus groups as qualitative research methods can be used successfully to assess needs, develop interventions, test new ideas or programmes, improve existing programmes and generate a range of ideas on a particular subject. Although we used FGDs in a more interpretative and evaluative way (Moser and Kalton, 1971), FGDs have been used as an interactive method in behavioural interventions (Hadiyono et al., 1996), and also as a follow up in assessing effects after interventions (Kachur et al., 1999). Furthermore, FGDs yield detailed descriptions of experiences, opinions, perceptions and beliefs about people’s health behaviour. As in our study, FGDs alone have been used as a self-contained method in studies which serve as the only principal source of data (Asbury, 1995; McDaniel and Bach, 1996). This study documents mothers’ good knowledge in recognising mild malaria. However, they perceived degedege (convulsions) as caused by evil spirits and practiced traditional home treatment. Grandmothers and mothers in-laws were the key decision-makers in the management for this condition including where to seek care (traditional
C. Comoro et al. / Acta Tropica 87 (2003) 305 /313
healers). Grandmothers and mothers-in laws are powerful tools to preserve deep-rooted cultural and family beliefs. They were strong in forbidding young mothers from taking sick children to hospital and gave powerful condemnations and/ or directives, which few young mothers would dare to ignore. Urinating on and fuming the child suffering from degedege with elephant dung were the dominating traditional remedies reported by most mothers and were perceived to be effective while injections were considered fatal. Urinating on the child itself may not pose any risk to the child’s health but fuming the child with smoke from elephant dung can cause health problems. Use of local herbs was reported from the coastal region of Tanzania (Makemba et al., 1996), Kenya (Mwenesi et al., 1995b), Uganda (KengeyaKayondo et al., 1994). In addition to home treatment the indigenous people (i.e. Zaramo, Kwere, Doe) in our study area when consulting outside home normally start at the traditional healer. Such practices are potential causes for delaying the child from getting proper hospital treatment for this life threatening condition. Most of the respondents perceive malaria as the major health problem in the community and recognised that young children, pregnant women and immigrants were the most vulnerable groups, compatible with high malaria transmission in the area (Kilama and Kihamia, 1991). The mothers’ description of malaria symptoms was consistent with the biomedical definition of mild malaria, an indication of some knowledge in recognising malaria fever. This is in contrast to the poor recognition rates of malaria symptoms reported earlier from Tanzania (Rooth and Bjo¨rkman, 1992). However, other studies from Uganda, Kenya and Ghana reported good recognition rate (Kengeya-Kayondo et al., 1994; Ruebush et al., 1995; Ahorlu et al., 1997). Primary health care facilities in the area are lacking laboratory support and malaria is diagnosed on a clinical basis and treated presumptively similar to what is being done at home by mothers. The latter expressed their dissatisfaction with the situation complaining about the lack of diagnostic facilities at these health units. The poor quality of
311
care in terms of both malaria diagnosis and treatment reported from public health facilities in the same district (Nsimba et al., 2002) might justify their dissatisfaction. Despite the fact that all the participants in the mothers FGDs were knowledgeable on the signs and symptoms of malaria fever, yet participants in three of the ten FGDs were not as clear on its causes. This is in consistent with other reports of perceptions regarding malaria aetiology (KengeyaKayondo et al., 1994; Ahorlu et al., 1997). This will have implication on preventive measures like use of bed nets. Home treatment mainly with antipyretics and sponging appears to be the common initial response to malaria fever in the study communities, as has also been reported from Uganda and Kenya (Kengeya-Kayondo et al., 1994; Mwenesi et al., 1995a,b). The practice of waiting for fever to subside for a day or so after administering only antipyretics as reported by the mothers may imply a delay in prompt malaria treatment. Participants in the mothers’ groups had a locally and culturally based explanation for degedege as an illness that attacks many children, mostly under-fives. Although the mothers provided descriptions of the condition which is compatible with convulsions, there were conflicting views when it comes to its cause and treatment. Most (six FGDs) believed that the condition was caused by evil spirits (shetani ), a common perception in other African communities (Aikins et al., 1993; Makemba et al., 1996; Ahorlu et al., 1997). However, most of the mothers from the urban settings associated the illness with malaria. This suggests that the cultural perception on degedege was most pronounced in rural areas, indicating that proper knowledge on the relationship between malaria and convulsions already begun to diffuse in the urban communities. The health workers seem to be aware of the traditional treatment practices carried out in the communities. Interestingly, some of the mothers from the urban settings prefer the hospital as first resort for degedege , while others would combine the two treatment seeking practices. This might indicate a transition from old practices to modern. The fact that mothers who associated degedege
312
C. Comoro et al. / Acta Tropica 87 (2003) 305 /313
with malaria yet would take their children to traditional healers is worrisome. Such discrepancy between knowledge and practice may have implications for a successful educational intervention. The perception and the fear that an injection would be fatal for children suffering from such condition might have influenced the preference to traditional healers. It would have been interesting if our study had also assessed the knowledge and perceptions of the traditional practitioners which unfortunately was not the case. None of the other studies on degedege in Tanzania (Winch et al., 1995; Makemba et al., 1996), Uganda (Kengeya-Kayondo et al., 1994) and Ghana (Ahorlu et al., 1997) had shown explicitly the role of grandmothers and mothers in-law in decision making for seeking care for a child with convulsions. In addition to the findings in these reports, we also found treatment practices such as, urinating on the child, fuming him/her with elephant dung and inhalation of garlic vapour, which the study groups commonly perceived as effective treatment for convulsions. Furthermore, we found that treatment practice for convulsions (degedege ) were accompanied with a lot of restrictions in homes. Our findings also reflect the views of mothers from urban, periurban and rural settings on the management of degedege . Our study sample being drawn from those deciding to seek care at public health facilities does not represent the total population in the community. This study rather illuminates an understanding of illness perceptions leading to patterns of care seeking. The primary goal of the National Malaria Control Programme is to reduce morbidity and mortality through prompt diagnosis and adequate treatment. The focus of delivery of such measures has been through the formal health facilities. However, such settings, where trained personnel provide treatment, represent only part of the providers of care for children with malaria. Since early recognition of febrile illnesses suspected to be malaria begins in the home, efforts should be directed at improving quality of malaria management in homes. It has been documented that malaria mortality can be reduced by about 40%
through effectively training mothers to recognise uncomplicated malaria in their under five children and also promptly give appropriate treatment (Kidane and Morrow, 2000). It is time to assess community interventions on improving community understanding and treatment seeking practices for severe childhood malaria. Such interventions should address the gaps in local perceptions and practices regarding management of convulsions and impaired consciousness, prominent features in severe malaria.
Acknowledgements We thank mothers and health workers for participating in the study and the district authorities in Kibaha for their co-operation. Special thanks go to our research assistants, Mr Adiel Mushi and Mr Huruma Kisaka, for note taking during the FGDs. Professor Amos Massele (Dept of Clinical Pharmacology-MUCHS, co-ordinator of the malaria research group) and Dr Edmund. J. Kayombo-Institute of Traditional Medicine (MUCHS) gave valuable comments during the revision of this manuscript. Financial support was provided by SAREC-the research division of the Swedish International Development Co-operation Agency (Sida).
References Ahorlu, C.K., Dunyo, S.K., Afari, E.A., Koram, K.A., Nkurumah, F.K., 1997. Malaria related beliefs and behaviour in southern Ghana. Implications for treatment, prevention and control. Trop. Med. Int. Health 2, 488 /499. Aikins, M.K., Pickering, H., Alonso, P.L., D’Alessandro, U., Lindsay, S.W., Todd, J., Greenwood, B.M., 1993. A malaria control trial using insecticide treated bed-nets and targeted chemoprophylaxis in a rural area of the Gambia, West Africa-Perceptions of the causes of malaria and its treatment and prevention in the study area. Trans. R. Soc. Trop. Med. Hyg. 87, 25 /30. Asbury, J., 1995. Overview of focus group research. Qual. Health Res. 5, 414 /420. Hadiyono, J.E.P., Suryawati, S., Danu, S.S., Sunartono, Santoso, B., 1996. Interactional group discussion: results of a controlled trial using a behavioural intervention to reduce the use of injection in public health facilities. Soc. Sci. Med. 42, 1177 /1183.
C. Comoro et al. / Acta Tropica 87 (2003) 305 /313 Kachur, S.P., Phillip-Howard, P.A., Odhacha, A.M., Ruebush, T.K., Oloo, A.J., Nahlen, B.L., 1999. Maintenance and sustained use of insecticide-treated bed nots and curtains three years after a controlled trial in western Kenya. Trop. Med. Int. Health 4, 728 /735. Kengeya-Kayondo, J.F., Seely, J.A., Kajura-Bajenja, E., Kabunga, E., Mubiru, E., Sembaija, F., Mulder, W.D., 1994. Recognition, treatment seeking behaviour and perception of cause of malaria among rural women in Uganda. Acta Trop. 58, 267 /273. Kidane, G., Morrow, R.H., 2000. Teaching mothers to provide home treatment of malaria in Tigray, Ethiopia: a randomised trial. Lancet 356, 550 /555. Kilama, W.L., Kihamia, C.M., 1991. Malaria. In: Heath and Disease in Tanzania. Harper Collins, pp. 117 /132. Makemba, A.M., Winch, P.J., Makame, V.M., Mehl, G.L., Premji, Z., Minjas, J.N., Shiff, C.J., 1996. Treatment practices for degedege, a locally recognised febrile illness, and implications for strategies to decrease mortality from severe malaria in Bagamoyo district. Tanzania Trop. Med. Int. Health 1, 305 /313. Massele, A.Y., Sayi, J., Nsimba, S.E.D., Ofori-Adjei, D., Laing, R.O., 1993. Knowledge and management of malaria in Dares-Salaam. Tanzania East Afr. Med. J. 70, 639 /642. McCombie, S.C., 1996. Treatment seeking behaviour for malaria: a review of recent research. Soc. Sci. Med. 43, 933 /945. McDaniel, R.W., Bach, C.A., 1996. Focus group research: the question of scientific rigor. Rehab. Nurs. Res. 3, 53 /59. Mnyika, S.K., Killewo, J.Z.J., Kabalimu, T., 1995. Selfmedication with antimalarial drugs in Dar-es-Salaam. Tanzania Trop. Geogr. Med. 47, 32 /34. Moser, C.A., Kalton, G., 1971. Survey Methods in Social Investigations, second ed.. Heinemann. Mwenesi, H.A., Harphan, T., Snow, R.W., 1995a. Child malaria treatment practices among mothers in Kenya. Soc. Sci. Med. 40, 1271 /1277.
313
Mwenesi, H.A., Harphan, T., Marsh, K., Snow, R.W., 1995b. Perceptions of symptoms of severe childhood malaria among Mujikenda and Luo residents of coastal Kenya. J. Biosoc. Sci. 27, 235 /244. Molineaux, L., Gramiccia, G., 1980. The Garki Project: Research on the Epidemiology and Control of Malaria in the Sudan Savanna of West Africa. World Health Organisation, Geneva, p. 39. National Malaria Control Programme. Handbook for Health Workers. Plan of operation 1990 /1995. Communicable Disease Control Series No. 2. Epidemiology and communicable control Unit, Ministry of Health-Tanzania Mainland, 1990, pp. 9 /10. Nsimba, S.E.D., Massele, A.Y., Mbatiya, Z.A., Warsame, M.Y., Tomson, G., 1999. A household survey of sources, availability and use of antimalarials in a rural area of Tanzania. Drug Info. J. 33, 1032 /1035. Nsimba, S.E.D., Massele, A.Y., Eriksen, J., Gustafsson, L., Tomson, G., Warsame, M.Y., 2002. Case management of malaria in under-fives in primary health care facilities in a Tanzanian District. Trop. Med. Int. Health 7, 201 /209. Rooth, I., Bjo¨rkman, A., 1992. Fever episodes in a holoendemic malaria area of Tanzania: parasitological and clinical findings and diagnosis aspects related to malaria. Trans. R. Soc. Trop. Med. Hyg. 86, 479 /482. Ruebush, T.K., Kern, M.K., Campbell, C.C., Aloo, A.J., 1995. Self treatment of malaria in a rural area of Western Kenya. Bull. World Health Organ. 73, 229 /236. Winch, J.P., Makemba, A.M., Kamazima, R.S., Lurie, M., Lwihula, G.K., Premji, Z., Minjas, N.J., Shiff, J.C., 1995. Local terminology for febrile illness in Bagamoyo district, Tanzania and its impact on the design of a communitybased malaria control programme. Soc. Sci. Med. 42, 1057 / 1067. World Health Organisation, 1993. Implementation of the Global Malaria Strategy. World Health Organisation. Tech Report Series No. 839; WHO, Geneva.
Local understanding, perceptions and reported practices of mothers/guardians and health workers on childhood malaria in a Tanzanian district* implications for malaria control /
C. Comoro a, S.E.D. Nsimba b,c,*, M. Warsame c, G. Tomson c,d a
Department of Sociology, University of Dar-es-Salaam, Dar-es-Salaam, Tanzania Department of Clinical Pharmacology, Muhimbili University College of Health Sciences, Dar-es-Salaam, Tanzania c Division of International Health (IHCAR), Department of Public Health Sciences, Karolinska Institute, S-171 76, Stockholm, Sweden d Medical Management Centre, Karolinska Institute, Stockholm, Sweden b
Received 21 December 2001; received in revised form 4 December 2002; accepted 20 January 2003
Abstract Knowledge on local understanding, perceptions and practices of care providers regarding management of childhood malaria are needed for better malaria control in urban, peri-urban and rural communities. Mothers of under five children attending five purposively selected public health facilities in the Kibaha district, Tanzania, were invited to participate in 10 focus group discussions (FGDs). The health workers of these facilities were included in six other FGDs to elicit their professional views. Analysis was done using interpretative and qualitative approaches. Both health workers and all mothers were clear about the signs and symptoms of homa ya malaria , a description consistent with the biomedical definition of mild malaria. Although most of the mothers related this to mosquito bites, some did not. Mothers also described a severe childhood illness called degedege , consistent with convulsions. Most of the mothers failed to associate this condition with malaria, believing it is caused by evil spirits. Urinating on or fuming the child suffering from degedege with elephant dung were perceived to be effective remedies while injections were considered fatal for such condition. Traditional healers were seen as the primary source of treatment outside homes for this condition and grandmothers and mother in-laws are the key decision makers in the management. Our findings revealed major gaps in managing severe malaria in the study communities. Interventions addressing these gaps and targeting mothers/guardians, mother in-laws, grandmothers and traditional healers are needed. # 2003 Published by Elsevier B.V. Keywords: Focus groups discussions; Mothers; Under-fives; Malaria; Tanzania; Traditional healers; Degedege
1. Introduction
* Corresponding author. Tel.: /46-8-517-79895; fax: /46-8311590. E-mail address: [email protected] (S.E.D. Nsimba). 0001-706X/03/$ - see front matter # 2003 Published by Elsevier B.V. doi:10.1016/S0001-706X(03)00113-X
Malaria is the leading cause of morbidity and mortality in children in Tanzania (Kilama and Kihamia, 1991). Adequate case management*/ early recognition and prompt treatment */has
306
C. Comoro et al. / Acta Tropica 87 (2003) 305 /313
been and still continues to be the basis for malaria control here as in most of Sub-Saharan Africa (National Malaria Control Programme, 1990; WHO, 1993). Formal health service has been the primary focus for the National Malaria Control Programme in the country. However, surveys carried out in some African countries revealed that malaria in children is first recognised and responded to at home (Mwenesi et al., 1995a; McCombie, 1996). This is further supported by the reports of frequent self-treatment with antimalarials stocked at home and/or bought at drug shops (Massele et al., 1993; Mnyika et al., 1995; Nsimba et al., 1999). Understanding local knowledge, perceptions and practices of malaria management has become the focus of research during the last decade. Studies from Kenya and Ghana reported that mothers had good knowledge in recognising symptoms suggestive of malaria (Ruebush et al., 1995; Ahorlu et al., 1997). However, surveys from Tanzania and Nigeria reported the contrary (Rooth and Bjo¨rkman, 1992; Molineaux and Gramiccia, 1980). Other studies revealed that convulsions, a prominent condition in severe malaria, were not associated with malaria by caretakers but rather to supernatural cause (Mwenesi et al., 1995b; Winch et al., 1995; Ahorlu et al., 1997). Traditional healers were reported to be the focal point for consultation. As part of a malaria project aiming to develop context specific interventions to improve the quality of malaria case management, we investigated the local understanding, perceptions and practices of caretakers on management of childhood malaria in urban, peri-urban and rural settings.
2. Material and methods 2.1. Study area and population The study was conducted between July and September 1999 in the Kibaha district located 40 kms north west of Dar-es-Salaam. The area is highly endemic for malaria (Nsimba et al., 1999)
and is populated by the Zaramo, Doe, Kwavi and Ndengeleko tribes. In order to assess the knowledge, perceptions and practices about childhood malaria, mothers with sick children attending five purposively selected health facilities were targeted. The district had in total 10 public health facilities serving urban (n /3), peri-urban (n /1) and rural (n /6) communities. We wanted to include facilities from all the different geographical areas. Thus we selected two facilities in the urban (Mkoani health centre and Mwendapole dispensary), the periurban (Mlandizi health centre) and two in the rural (Ruvu Stesheni and Magindu dispensaries). We chose the two most interior of the rural health facilities. In each health facility, mothers with sick underfives were randomly selected by asking each of them to pick a number from a basket and those who picked even numbers were again purposively selected. Those mothers who were finally selected were informed about the study and were asked if they had time to participate in the focus group discussions (FGDs). Everybody consented verbally. Mothers with severely ill children were excluded. Discussions were also held with the health workers in the selected five facilities in order to elicit a professional viewpoint of the problem. 2.2. Data collection We used FGDs as a method and its thematic guides were translated into Swahili language and were pre-tested in similar facilities in Dar-esSalaam. Our selection to use this method was based on its comparative advantage over individual interviews. These include group interaction of participants which is lacking in interviews, it is relatively cheap as it can be conducted in a short span of time by small number of staff with limited financial resources, it provides an opportunity for researchers to meet and talk to many people in one session. Furthermore, the method allows researchers to get ideas about peoples experience, opinions, beliefs and participants’ responses can be presented with actual quotations that helps the reader to get the main ideas or messages.
C. Comoro et al. / Acta Tropica 87 (2003) 305 /313
The focus of the inquiry was on mothers’ knowledge and perceptions on major health problem in their community, malaria seasonality, signs and symptoms of childhood malaria and treatment practices. The FGD sessions were moderated by the principal investigator (first author) and assisted by the second author. Each session lasted between 1 h and 15 min to 1 h and 45 min. Note taking was done by two graduate social scientists in their final year master of arts in Sociology at the University of Dar-es-Salaam. The notes were translated into the English language by the first and second authors, respectively. Discussions with health workers at each facility were done either prior to or after their working hours. A total of 16 FGDs were held, 10 (two per facility) consisting of 7 /10 participants per group session were held with mothers, the remaining six FGDs were held with the health workers (5 /9 participants per group). One FGD was held at each of the three dispensaries as planned, three FGDs with health care providers instead of four were held at health centres. In one of the health centres, Mkoani, part of the health staff were out of station for outreach activities and thus we managed to get only one FGD group. The FGDs for mothers and health care providers were conducted separately at each facility. The FGDs with mothers were done before those of health workers, and prior to the consultations with the health workers. 2.3. Data analysis Data analysis was done using the interpretative understanding approach (Moser and Kalton, 1971). This analysis involves making sense of the collected information through eliciting meanings (actual words used by participants) or responses that people gave to different situations which are important (e.g. what the word ‘degedeg e’ means to them). A code sheet was created following the focus group guide and data were coded. Then a master sheet was done giving listing of all the responses from focus groups. These responses were interpreted by looking at their patterns. Sorting and sifting through information, looking for patterns, consensus, differences, variations or
307
contradictions, and weighing the relative importance of information complimented the interpretative understanding. Ranking and tabulating of the data was also done.
3. Ethical approval The study was approved by the Muhimbili University College of Health Sciences (MUCHS) Human Ethics Committee, Tanzania. Permission were obtained from the regional and district administrative authorities. Informed consent was also obtained from the mothers and health workers in the studied facilities. Mothers with severely ill children were not included in the study. Discussions with the health workers were held after their consultation hours.
4. Results Demographic characteristics of the participants are presented in Table 1. All but one of participants accompanying children under 5 years were mothers. The remaining one was a female guardian of a child. The majority of the mothers aged between 21 /30 years. Most described their occupation either as housewives (32%) or peasants (49%). The rest were petty traders or civil servants. The age profile of the health workers ranged from 20s to 40s. They were 4 Clinical Officers, 6 Table 1 Socio-demographic profile of participants in the 16 FGDs Characteristics
Mothers Health care providers
No. participants Females (%)
85 100
43 58
20 74 3.5 2.3
19 81 / /
8.2 87.1 4.7
7 84 9
Marital status Single (%) Married (%) Cohabiting (%) Widowed (%) Education No formal education (%) Primary education (%) Secondary education (%)
308
C. Comoro et al. / Acta Tropica 87 (2003) 305 /313
Assistant Clinical Officers, 6 Trained Nurses, 11 Nurse Assistants, 12 MCH Aides and 4 Traditional Birth Attendants. Most health workers had primary education with some basic medical training. All but three of the 16 FGDs were unanimous that malaria attacks were the main health problems in their communities. The remaining three who were from mothers (two groups from Mlandizi, one from Ruvu Stesheni) considered cholera and gastrointestinal problems (vomiting and diarrhoea) as the number one health problem and ranked malaria second. Children were recognised as being the most vulnerable group to malaria. Pregnant women, visitors and immigrants were also considered by some of the participants to be at special risk. When participants were asked to identify season(s) associated with peak of malaria in their communities, nominal seasons as opposed to lunar calendar were much more familiar to both categories (mothers and health workers). Most of the participants associated rainy and cold seasons with peak of malaria. The two categories of participants (mothers and health workers) provided similar descriptions of malaria fever ‘homa ya malaria ’ in children. All associated symptoms of malaria with hot body/ high temperature, vomiting, loss of appetite, diarrhoea, yellow eyes and crying. The health workers, in addition to the above, also mentioned anaemia and cough. Seven of the ten mothers’ FGDs related malaria to mosquito bites as all the FGDs with health workers did. The cause of malaria was unknown to the remaining mother groups. The Mlandizi mothers blamed population growth and multiplication of pit latrines as attractive breeding grounds for mosquitoes. The mothers had a special explanation for conditions compatible with severe malaria, which is a febrile condition that attacks many children locally called degedege. Some of the mothers avoided even mentioning it because it was believed to be a bad omen. Thus, they simply referred it as ‘childhood disease’ and gave a very elaborative description of the condition. High body temperature and the child’s body behaviour changes (eye rolling, jolting, quavering, and stiffening, foaming in the mouth) featured prominently in the descrip-
Table 2 Mothers’ description of degedege according to the participants in the 10 FGDs with mothers FGD groups
Description of degedege
Mkoani (urban)
FGD 1 FGD 2
Excessive high temperature, body jolts, rolling of the eyes High fever, spasm of body jolt or fits, mouth foaming
Mwendapole (urban)
FGD 3 FGD 4
Turning of eyes, weakening of the limbs Fits or body jolts, shouting, heavy vibration of head veins
Mlandizi (periurban)
FGD 5 FGD 6
Body jolts, fits, shouting, head veins vibrates and high temperature Popping and twisting of the eyes, stiffening of legs, high temperature, crying continuously
Ruvu Stesheni (rural)
FGD 7
Eyes turning, shivering of the body and inability to walk, breathing through the ribs High stomach temperature, body jolts and stiffening
FGD 8 Magindu (rural)
FGD 9 FGD 10
High fever, heavy breathing, cold feet, warm upper body, eye twisting High fever, heavy breathing, cold feet, warm body, stiffening of the limbs
tion of signs and symptoms of degedege (Table 2). The health care providers gave similar description of the condition and participants in one group stressed on the notion of excessive temperature in the brain and convulsion. Anaemia was noted as an important severe childhood illness by health workers, a notion that none of the mothers’ group had. With regards to mothers’ perceived causes of degedege , three views emerged from the discussion, the dominating one being that of evil spirits ‘shetan i’, as stated by participants in six of the mothers’ FGDs. Only two FGDs from the mothers group, one from urban and one from peri-urban settings, mentioned malaria as the cause. In contrast, all the health care providers (six FGDs) associated convulsions with malaria (Table 3). The first line of action in responding to malaria fever across the 16 FGDs was giving antipyretics
C. Comoro et al. / Acta Tropica 87 (2003) 305 /313
Table 3 Health workers’ description of degedege according to participants in the 6 FGDs with health workers FGD groups
Description of degedege
Mkoani (urban) FGD 11 Mwendapole FGD (urban) 12 Mlandizi (peri- FGD urban) 13 FGD 14
Excessive high temperature in the brain, convulsion High temperature or no temperature High temperature, vomiting yellowish Very high temperature, Anaemic child
Ruvu Stesheni (rural)
Severe fever of the stomach, severe fever of the head, diarrhoea, vomiting and loss of appetite Asymptomatic, stiffening, popping eyes, foaming in the mouth
FGD 15
Magindu (rural) FGD 16
and sponging to lower the body temperature. Observing the condition of the child before deciding to take to hospital was frequently mentioned and was commented as follows: ‘If the child brightens/resumes playing she is no longer taken to hospital. If the fever persists or escalates the child is sent to hospital’. All mothers said they kept stock of aspirin or panadol at home. Some said they had chloroquine syrup in stock. When the child gets better, some mothers finish the dose while others discontinue administration and stock the remaining medicine for the ‘anticipated’ next attack. Some mothers expressed their dissatisfaction about the lack of testing for malaria parasites at the primary health care facilities. The health workers of Magindu rural dispensary (FGD 16) raised similar concerns about the lack of diagnostic facilities at primary health care facilities. With regard to children with convulsions, mothers of six FGDs stated that they would take them to a traditional healer (Table 4), in local language fundi or mganga, a Kiswahili word for an expert. It is a local title for a traditional healer along the coastal line and nearby regions in various realms of medicine, which involves divination by a healer to find the causes of illness. Mothers in one FGD from the rural, one from the peri-urban and one from the urban presented conflicting views: some would take their children to hospital, some to the traditional healers and
309
some to both, first to traditional healer then to hospital (Table 4). The later group stated that traditional remedies cool down the problem and the hospital subsequently treats the child completely. Some mothers, who related degedege to malaria as its cause, stated they would seek care from traditional practitioners for such condition. According to the health workers, there are two malaria-related behaviours cutting across the ethnic population. When a child gets the severe malaria attack, the indigenous people (i.e. Zaramo, Kwere, Doe etc.) normally start at the fundi . On contrary terms the immigrants, mostly comprising of the Wakwavi and other up-country people go straight to the hospital with their sick children. According to the mothers, urinating on or fuming the child with elephant dung were the most common traditional remedies for degedege . Furthermore, other treatment options mentioned by mothers include: making the child inhale garlic vapour or smell from a pit latrine. During the treatment of the condition there are several restrictions in force in a household with a suffering child. These include (i) no knock at the door where the sick child is and in particular the formal traditional request of ‘hodi wenyewe’ (i.e. may I come in); (ii) taking fire out of such a house is totally prohibited. (iii) Entering in such a house with a lemon fruit, raw fish or pepper is absolutely disallowed. Concerns about giving injectables to a child with degedege were raised. Mothers reported that, people say and themselves had witnessed several incidences whereby a child with the condition died after getting an injection. However, according to health workers on the other hand, the first response towards the condition was to give antipyretics, anti-convulsants, sponging and chloroquine injections. These health workers, confirmed the prevalence of the traditional practices stated by mothers in the following response ‘A child suffering from degedege must be urinated on and before getting to the hospital s/he has been sent to several traditional healers’. As a whole, degedege is considered by mothers as fatal hence decision making is exercised by important people in the community such as
C. Comoro et al. / Acta Tropica 87 (2003) 305 /313
310
Table 4 Mothers’ perception of the causes of degedege and how they respond to it FGDs Mkoani (urban)
Perceived causes of degedege FGD 1 FGD 2
Actions taken
Excessively high temperature in the brain Take the child to hospital Severe fever related to rapid multiplication of malaria parasites Take the child to traditional healers
Mwendapole (urban) FGD 3 FGD 4
Believed to be due to ‘Shetani’ (evil spirit) Cause unknown but may be due to high temperature
Take the child to traditional healer Give aspirin, chloroquine Take the child traditional healer
Mlandizi (peri-urban)
FGD 5
Due to malaria which has gone into the head
Take the child to hospital
FGD 6
Due to ‘Shetani’ (evil spirit)
Take the child to traditional healer Take the child to traditional healer
Ruvu Stesheni (rural) FGD 7 FGD 8
Cause unknown but may be due to ‘Shetani’ Due to ‘Shetani’
Take the child to traditional healer Take the child to traditional healer
Magindu (rural)
Cause is related to ‘Shetani’
Take the child to traditional healer Few will visit hospital as first resort Take the child to traditional healer
FGD 9
FGD 10 Cause is related to ‘Shetani’
traditional healers, grandmothers and mother-in laws. In this study, these three categories were the immediate referrals for this childhood condition, the same advisors who were strong in forbidding young mothers from taking children with degedege to hospital. For example a common dialogue between young mothers and mothers in-law or grandmothers regarding taking the sick child to hospital reported as follows: Verbal translation: ‘Habali komga huko imwana? Kolonda kumkoma’ ? Translation for the question from grandmothers/mothers-in-law: Why are you taking him/her there? Do you want to kill him/her? Mothers’ response to this: verbal translation; ‘Simkoma bule mie. Nomgala akatibiwe ’ meaning ‘I am not killing him/her. I am taking him/her to hospital for treatment’. The mothers reported that they were discourteously warned by mother-inlaws quoting ‘You are taking the child to the hospital, you are going to cause the child’s death and be responsible for it’. Causing death in this manner is a stigma that no young mother wants to live with. However, if a mother was to take a convulsing child to hospital, the mothers-in-law/ grandmothers simply concludes: ‘Look here, this child is not a woman who you can easily replace with another one’.
5. Discussion Focus groups as qualitative research methods can be used successfully to assess needs, develop interventions, test new ideas or programmes, improve existing programmes and generate a range of ideas on a particular subject. Although we used FGDs in a more interpretative and evaluative way (Moser and Kalton, 1971), FGDs have been used as an interactive method in behavioural interventions (Hadiyono et al., 1996), and also as a follow up in assessing effects after interventions (Kachur et al., 1999). Furthermore, FGDs yield detailed descriptions of experiences, opinions, perceptions and beliefs about people’s health behaviour. As in our study, FGDs alone have been used as a self-contained method in studies which serve as the only principal source of data (Asbury, 1995; McDaniel and Bach, 1996). This study documents mothers’ good knowledge in recognising mild malaria. However, they perceived degedege (convulsions) as caused by evil spirits and practiced traditional home treatment. Grandmothers and mothers in-laws were the key decision-makers in the management for this condition including where to seek care (traditional
C. Comoro et al. / Acta Tropica 87 (2003) 305 /313
healers). Grandmothers and mothers-in laws are powerful tools to preserve deep-rooted cultural and family beliefs. They were strong in forbidding young mothers from taking sick children to hospital and gave powerful condemnations and/ or directives, which few young mothers would dare to ignore. Urinating on and fuming the child suffering from degedege with elephant dung were the dominating traditional remedies reported by most mothers and were perceived to be effective while injections were considered fatal. Urinating on the child itself may not pose any risk to the child’s health but fuming the child with smoke from elephant dung can cause health problems. Use of local herbs was reported from the coastal region of Tanzania (Makemba et al., 1996), Kenya (Mwenesi et al., 1995b), Uganda (KengeyaKayondo et al., 1994). In addition to home treatment the indigenous people (i.e. Zaramo, Kwere, Doe) in our study area when consulting outside home normally start at the traditional healer. Such practices are potential causes for delaying the child from getting proper hospital treatment for this life threatening condition. Most of the respondents perceive malaria as the major health problem in the community and recognised that young children, pregnant women and immigrants were the most vulnerable groups, compatible with high malaria transmission in the area (Kilama and Kihamia, 1991). The mothers’ description of malaria symptoms was consistent with the biomedical definition of mild malaria, an indication of some knowledge in recognising malaria fever. This is in contrast to the poor recognition rates of malaria symptoms reported earlier from Tanzania (Rooth and Bjo¨rkman, 1992). However, other studies from Uganda, Kenya and Ghana reported good recognition rate (Kengeya-Kayondo et al., 1994; Ruebush et al., 1995; Ahorlu et al., 1997). Primary health care facilities in the area are lacking laboratory support and malaria is diagnosed on a clinical basis and treated presumptively similar to what is being done at home by mothers. The latter expressed their dissatisfaction with the situation complaining about the lack of diagnostic facilities at these health units. The poor quality of
311
care in terms of both malaria diagnosis and treatment reported from public health facilities in the same district (Nsimba et al., 2002) might justify their dissatisfaction. Despite the fact that all the participants in the mothers FGDs were knowledgeable on the signs and symptoms of malaria fever, yet participants in three of the ten FGDs were not as clear on its causes. This is in consistent with other reports of perceptions regarding malaria aetiology (KengeyaKayondo et al., 1994; Ahorlu et al., 1997). This will have implication on preventive measures like use of bed nets. Home treatment mainly with antipyretics and sponging appears to be the common initial response to malaria fever in the study communities, as has also been reported from Uganda and Kenya (Kengeya-Kayondo et al., 1994; Mwenesi et al., 1995a,b). The practice of waiting for fever to subside for a day or so after administering only antipyretics as reported by the mothers may imply a delay in prompt malaria treatment. Participants in the mothers’ groups had a locally and culturally based explanation for degedege as an illness that attacks many children, mostly under-fives. Although the mothers provided descriptions of the condition which is compatible with convulsions, there were conflicting views when it comes to its cause and treatment. Most (six FGDs) believed that the condition was caused by evil spirits (shetani ), a common perception in other African communities (Aikins et al., 1993; Makemba et al., 1996; Ahorlu et al., 1997). However, most of the mothers from the urban settings associated the illness with malaria. This suggests that the cultural perception on degedege was most pronounced in rural areas, indicating that proper knowledge on the relationship between malaria and convulsions already begun to diffuse in the urban communities. The health workers seem to be aware of the traditional treatment practices carried out in the communities. Interestingly, some of the mothers from the urban settings prefer the hospital as first resort for degedege , while others would combine the two treatment seeking practices. This might indicate a transition from old practices to modern. The fact that mothers who associated degedege
312
C. Comoro et al. / Acta Tropica 87 (2003) 305 /313
with malaria yet would take their children to traditional healers is worrisome. Such discrepancy between knowledge and practice may have implications for a successful educational intervention. The perception and the fear that an injection would be fatal for children suffering from such condition might have influenced the preference to traditional healers. It would have been interesting if our study had also assessed the knowledge and perceptions of the traditional practitioners which unfortunately was not the case. None of the other studies on degedege in Tanzania (Winch et al., 1995; Makemba et al., 1996), Uganda (Kengeya-Kayondo et al., 1994) and Ghana (Ahorlu et al., 1997) had shown explicitly the role of grandmothers and mothers in-law in decision making for seeking care for a child with convulsions. In addition to the findings in these reports, we also found treatment practices such as, urinating on the child, fuming him/her with elephant dung and inhalation of garlic vapour, which the study groups commonly perceived as effective treatment for convulsions. Furthermore, we found that treatment practice for convulsions (degedege ) were accompanied with a lot of restrictions in homes. Our findings also reflect the views of mothers from urban, periurban and rural settings on the management of degedege . Our study sample being drawn from those deciding to seek care at public health facilities does not represent the total population in the community. This study rather illuminates an understanding of illness perceptions leading to patterns of care seeking. The primary goal of the National Malaria Control Programme is to reduce morbidity and mortality through prompt diagnosis and adequate treatment. The focus of delivery of such measures has been through the formal health facilities. However, such settings, where trained personnel provide treatment, represent only part of the providers of care for children with malaria. Since early recognition of febrile illnesses suspected to be malaria begins in the home, efforts should be directed at improving quality of malaria management in homes. It has been documented that malaria mortality can be reduced by about 40%
through effectively training mothers to recognise uncomplicated malaria in their under five children and also promptly give appropriate treatment (Kidane and Morrow, 2000). It is time to assess community interventions on improving community understanding and treatment seeking practices for severe childhood malaria. Such interventions should address the gaps in local perceptions and practices regarding management of convulsions and impaired consciousness, prominent features in severe malaria.
Acknowledgements We thank mothers and health workers for participating in the study and the district authorities in Kibaha for their co-operation. Special thanks go to our research assistants, Mr Adiel Mushi and Mr Huruma Kisaka, for note taking during the FGDs. Professor Amos Massele (Dept of Clinical Pharmacology-MUCHS, co-ordinator of the malaria research group) and Dr Edmund. J. Kayombo-Institute of Traditional Medicine (MUCHS) gave valuable comments during the revision of this manuscript. Financial support was provided by SAREC-the research division of the Swedish International Development Co-operation Agency (Sida).
References Ahorlu, C.K., Dunyo, S.K., Afari, E.A., Koram, K.A., Nkurumah, F.K., 1997. Malaria related beliefs and behaviour in southern Ghana. Implications for treatment, prevention and control. Trop. Med. Int. Health 2, 488 /499. Aikins, M.K., Pickering, H., Alonso, P.L., D’Alessandro, U., Lindsay, S.W., Todd, J., Greenwood, B.M., 1993. A malaria control trial using insecticide treated bed-nets and targeted chemoprophylaxis in a rural area of the Gambia, West Africa-Perceptions of the causes of malaria and its treatment and prevention in the study area. Trans. R. Soc. Trop. Med. Hyg. 87, 25 /30. Asbury, J., 1995. Overview of focus group research. Qual. Health Res. 5, 414 /420. Hadiyono, J.E.P., Suryawati, S., Danu, S.S., Sunartono, Santoso, B., 1996. Interactional group discussion: results of a controlled trial using a behavioural intervention to reduce the use of injection in public health facilities. Soc. Sci. Med. 42, 1177 /1183.
C. Comoro et al. / Acta Tropica 87 (2003) 305 /313 Kachur, S.P., Phillip-Howard, P.A., Odhacha, A.M., Ruebush, T.K., Oloo, A.J., Nahlen, B.L., 1999. Maintenance and sustained use of insecticide-treated bed nots and curtains three years after a controlled trial in western Kenya. Trop. Med. Int. Health 4, 728 /735. Kengeya-Kayondo, J.F., Seely, J.A., Kajura-Bajenja, E., Kabunga, E., Mubiru, E., Sembaija, F., Mulder, W.D., 1994. Recognition, treatment seeking behaviour and perception of cause of malaria among rural women in Uganda. Acta Trop. 58, 267 /273. Kidane, G., Morrow, R.H., 2000. Teaching mothers to provide home treatment of malaria in Tigray, Ethiopia: a randomised trial. Lancet 356, 550 /555. Kilama, W.L., Kihamia, C.M., 1991. Malaria. In: Heath and Disease in Tanzania. Harper Collins, pp. 117 /132. Makemba, A.M., Winch, P.J., Makame, V.M., Mehl, G.L., Premji, Z., Minjas, J.N., Shiff, C.J., 1996. Treatment practices for degedege, a locally recognised febrile illness, and implications for strategies to decrease mortality from severe malaria in Bagamoyo district. Tanzania Trop. Med. Int. Health 1, 305 /313. Massele, A.Y., Sayi, J., Nsimba, S.E.D., Ofori-Adjei, D., Laing, R.O., 1993. Knowledge and management of malaria in Dares-Salaam. Tanzania East Afr. Med. J. 70, 639 /642. McCombie, S.C., 1996. Treatment seeking behaviour for malaria: a review of recent research. Soc. Sci. Med. 43, 933 /945. McDaniel, R.W., Bach, C.A., 1996. Focus group research: the question of scientific rigor. Rehab. Nurs. Res. 3, 53 /59. Mnyika, S.K., Killewo, J.Z.J., Kabalimu, T., 1995. Selfmedication with antimalarial drugs in Dar-es-Salaam. Tanzania Trop. Geogr. Med. 47, 32 /34. Moser, C.A., Kalton, G., 1971. Survey Methods in Social Investigations, second ed.. Heinemann. Mwenesi, H.A., Harphan, T., Snow, R.W., 1995a. Child malaria treatment practices among mothers in Kenya. Soc. Sci. Med. 40, 1271 /1277.
313
Mwenesi, H.A., Harphan, T., Marsh, K., Snow, R.W., 1995b. Perceptions of symptoms of severe childhood malaria among Mujikenda and Luo residents of coastal Kenya. J. Biosoc. Sci. 27, 235 /244. Molineaux, L., Gramiccia, G., 1980. The Garki Project: Research on the Epidemiology and Control of Malaria in the Sudan Savanna of West Africa. World Health Organisation, Geneva, p. 39. National Malaria Control Programme. Handbook for Health Workers. Plan of operation 1990 /1995. Communicable Disease Control Series No. 2. Epidemiology and communicable control Unit, Ministry of Health-Tanzania Mainland, 1990, pp. 9 /10. Nsimba, S.E.D., Massele, A.Y., Mbatiya, Z.A., Warsame, M.Y., Tomson, G., 1999. A household survey of sources, availability and use of antimalarials in a rural area of Tanzania. Drug Info. J. 33, 1032 /1035. Nsimba, S.E.D., Massele, A.Y., Eriksen, J., Gustafsson, L., Tomson, G., Warsame, M.Y., 2002. Case management of malaria in under-fives in primary health care facilities in a Tanzanian District. Trop. Med. Int. Health 7, 201 /209. Rooth, I., Bjo¨rkman, A., 1992. Fever episodes in a holoendemic malaria area of Tanzania: parasitological and clinical findings and diagnosis aspects related to malaria. Trans. R. Soc. Trop. Med. Hyg. 86, 479 /482. Ruebush, T.K., Kern, M.K., Campbell, C.C., Aloo, A.J., 1995. Self treatment of malaria in a rural area of Western Kenya. Bull. World Health Organ. 73, 229 /236. Winch, J.P., Makemba, A.M., Kamazima, R.S., Lurie, M., Lwihula, G.K., Premji, Z., Minjas, N.J., Shiff, J.C., 1995. Local terminology for febrile illness in Bagamoyo district, Tanzania and its impact on the design of a communitybased malaria control programme. Soc. Sci. Med. 42, 1057 / 1067. World Health Organisation, 1993. Implementation of the Global Malaria Strategy. World Health Organisation. Tech Report Series No. 839; WHO, Geneva.