University of Ghana http://ugspace.ug.edu.gh RC159. A5 D 99 3 0692 1007 3529 7 bite C.l G371259 University of Ghana http://ugspace.ug.edu.gh MPH DISSERTATION TOPIC: TREATMENT PRACTICES FOR MALARIA IN SMALL CHILDREN BY CAREGIVERS IN THE WASSA WEST DISTRICT OF GHANA THIS DISSERT AT ATION IS SUBMITTED TO THE SCHOOL OF PUBLIC HEALTH, UNIVERSITY OF GHANA, LEGON IN PARTIAL FULFILLMENT OF THE REQUIREMENT FOR THE AWARD OF PUBLIC HEALTH DEGREE. BY EMMANUEL KOFI DZOTSI DATE OF STUDY: JUNE - AUGUST 2003 University of Ghana http://ugspace.ug.edu.gh D E C L A R A T IO N I, Dr Emmanuel Kofi Dzotsi, do hereby declare that this dissertation is my own work being the product of my research while a resident of the Ghana School of Public Health and that the same work has not been submitted anywhere for the same purpose. Signed.... ..... Dr. Emmanuel Kofi Dzotsi (M.P.H. Resident, 2002/2003) Academic Supervisors: 1 ■ Signed..................................................... DR. IRENE AGYEPONG (District Director of Health Services, Dangme-West District) 2. Signed. Professor Fred Binka (School of Public Health University of Ghana) University of Ghana http://ugspace.ug.edu.gh DEDICATION This work is dedicated to my children Selasi Vincent Dzotsi, Senam Samuel Dzotsi and Dzifa Peace Dzotsi. University of Ghana http://ugspace.ug.edu.gh ABSTRACT The study was conducted in 5 randomly selected communities in the Wassa West district in the Western region of Ghana using both qualitative and quantitative methods to obtain mothers or caregiver’s knowledge on symptoms of childhood malaria (uncomplicated and complicated) in children under five years of age in relation to its management. A total of 300 caregivers were interviewed using a structured questionnaire. Majority of the respondents were mothers constituting 79%. Although the study population was made up of different ethnic groups, the local term used to represent a cluster of symptoms synonymous with the clinical diagnosis of malaria was “ebun" or "fever” which were used interchangeably. All mothers or caregivers named one or more signs by which they recognized the disease. The percentage of mothers or caregivers who perceived symptoms of malaria in small children to be fever was found to be 91%. Fever (hot body), vomiting and loss of appetite were the three most common signs mentioned. Although 89% of mothers or caregivers had access to a health facility, this does not appear to influence their treatment seeking behaviour. This is indicated by the fact that as many as 60% managed their child’s disease at home and used both traditional and modem treatment. The commonest anti-malaria drug was chloroquine. This is a significant finding considering the upsurge of chloroquine resistance, although chloroquine is still currently Ghana’s official first line drug for treating malaria. 74.8% of mothers or caregivers treated children under-5 years with malaria at home for 3 days. Poverty was a major factor that encouraged home treatment as confirmed by the fact that 55% of them treated their ill children at home because they had no money to send the child to hospital. The delay in seeking early appropriate treatment may be the cause of the high under-5 mortality. Higher level of education of mothers or caregivers was significantly associated with promptness of sending their sick children to clinic/hospital at the onset of the illness. However, in the event child did not recover with self­ medication, 85% of mothers or caregivers would send their children with malaria to clinic or hospital. The results of the study call for prompt educational action targeting mothers for the correct treatment of both complicated and uncomplicated malaria in children under-5 years. University of Ghana http://ugspace.ug.edu.gh Acknowledgements I would like to acknowledge the ideas and support from a number ofpersons: all staff o f Nsuaem, Aboso and Dompim Health Centres especially Mr. Osman Ibrahim Moro, Mr Robert Playe, Mary Yebuah, Gladys Dadziwa. The most important acknowledgement, by far, is to my supervisors Dr. T.S.M Avotri, Dr. Irene Agyepong and Prof. Fred Binka who have worked with me on this desertation. They have made considerable contribution, to which these few lines can hardly do justice; their intellectual vigour and generously given ideas and support have been invaluable. I also acknowledge the moral support given me by my colleague Dr. Yakubu Mahama. My sincere thanks also go to the chiefs and people o f Wassa West District who willingly provided the necessary information and ideas, especially the chiefs, assemblymen and people o f Aboso, Nsuaem, Dompim, and Pieso . And finally thanks to Mr. J. K. Arthur, Margaret Amponsah and all staff o f District Health Administration, Tarkwa for their continued support, patience and tolerance during the course o f writing this dissertation. University of Ghana http://ugspace.ug.edu.gh TABLE OF CONTENTS CONTENTS PAGES D eclaration z D edication H A bstract m Acknowledgem ents zv List o f tables and fig u re s viii L ist o f A bbrevia tions ix CHAPTER ONE 1.0 INTRODUCTION 1 1.1 Global m alaria burden 1 1.2 M alaria burden in Ghana 1 1.3 W hat is m alaria? 1 2.0 STATEMENT OF THE PROBLEM 2 3.0 RELEVANCE OF STUDY 2 CHAPTER TWO 4.0 LITERATURE REVIEW 3 4.1 Term inology related to m alaria 3 4.2 Local diagnostic criteria for m alaria in under-5 4 4.3 H ealth seeking behaviour for uncom plicated m alaria 5 4.4 Home rem edies for uncom plicated m alaria 6 4.5 H ealth seeking behaviour for childhood convulsion 7 5.0 HYPOTHESIS 7 6.0 STUDY OBJECTIVES 7 v University of Ghana http://ugspace.ug.edu.gh CHAPTER THREE 8 7.0 METHODOLOGY 8 7.1 Background o f study area 8 7.2 V ariables 9 7.3 Study design 12 7.4 E thical consideration 12 7.5 Data co llection 12 7.6 Pretesting o f research tools in a com m unity 13 7.7 Sample size calcu lation 13 7.8 Sam pling 13 7.9 Data analysis 14 CHAPTER FOUR 15 8.0 RESULTS 15 8.1 C haracteristics o f study com m unities and population 15 8.1.1 The study com m unity characteristics 15 8.1.2 Access to health facility 16 8.1.3 C haracteristics o f participants in FGD 17 8.1.4 Age sex d istribu tion o f respondents 17 8.1.5 Ethnic d istribu tion o f respondents 18 8.1.6 M arital status o f respondents 19 8.1.7 Educational status o f respondents 19 8.1.8 O ccupational status o f respondents 19 8.1.9 R elationship o f respondents with child 19 8.2 Term inology related to m alaria 20 8.3 M others/caregivers diagnostic criteria for m alaria in ch ildren 21 8.4 Treatm ent seeking practices for m alaria 23 8.4.1 Treatm ent seeking practices for uncom plicated m alaria 23 8.4.2 T reatm ent seeking practices o f childhood convulsions 29 8.4.3 Perceived causes o f childhood convulsions 30 8.4.4 Home rem edies for childhood m alaria 33 vi University of Ghana http://ugspace.ug.edu.gh CHAPTER FIVE 34 9.0 DISCUSSIONS 34 9.1 C haracteristics o f respondents 34 9.2 Term inology related to m alaria 35 9.3 M others/caregivers diagnostic criteria for m alaria 35 9.4 Treatm ent practices o f uncom plicated m alaria 36 9.5 Perceived causes and treatm ent practices o f m alaria 40 10.0 CONCLUSIONS 41 11.0 RECOM M ENDATIONS 42 12.0 REFERENCES 43 13.0 ANNEX (DATA COLLECTION TOOLS) 45 13.1 Focus group discussion guide 49 13.2 C ross-sectional survey questionnaire 50 vii University of Ghana http://ugspace.ug.edu.gh LIST OF TABLES Page Table 0: The proportion of population interviewed in the communities 14 Table 1: The communities and respondents interviewed 15 Table 2: Characteristics of the mothers group in focus group discussion 17 Table 3: Ethnic Distribution of respondents 18 Table 4: Relationship of respondents with child 19 Table 5: Local names for malaria 20 Table 6: Perceived symptoms of malaria in small children 22 Table 7: Number of signs mentioned as indicating malaria in small children 23 Table 8: First treatment/actions for small children with malaria 25 Table 9: Comparison of first treatment/actions for malaria in a child and educational level of respondents 25 Table 10: Comparison of first treatment /action for child with malaria and duration/time the treatment/action was taken 26 Table 11: Comparison of duration/time first line of treatment or action was taken and education of mothers/caregivers 27 Table 12: Subsequent treatments/actions taken when child not recovered 28 Table 13: Comparison of second treatment/actions if child not recovered and access to clinic 28 Table 14: Perception and knowledge on management on management childhood convulsions 29 Table 15: Mothers or caregivers perceived causes of childhood convulsions 30 Table 16: Reasons mothers/caregivers send sick children to hospital/clinic 31 University of Ghana http://ugspace.ug.edu.gh Table 17: Characteristics of the men’s group in the FGD 47 Table 18: Characteristics of grandfathers’ group in the FGD 47 Table 19: Characteristics of grandmothers’ group in the FGD 48 LIST OF FIGURES__________________________________ PAGE Fig 1: Respondents access to health facility 16 Fig 2: The main reasons mothers/caregivers treat sick children at home 33 Fig 3: Map of Wassa West District showing the main Subdistricts 54 LIST OF ABBREVIATIONS DA - District Assembly FGD - Focus Group Discussion GHS - Ghana Health Service KAP- Knowledge, attitude and practice MAP - Malaria action plan MOH - Ministry of Health SPH - School of Public Health TC - Traditional Council ix University of Ghana http://ugspace.ug.edu.gh CHAPTER ONE 1. INTRODUCTION 1.0 BACKGROUND OF THE STUDY 1.1 Global Malaria Burden Malaria causes an estimated 300 to 500 million acute cases per year. Malaria is a disease of young and the poor, many of them children who live with no easy access to health service. 80% of the cases occur in Africa resulting in one million deaths per year. Over 95% of deaths occur among under-fives in Africa (Binka 2001). One in five of the world’s population is at risk of malaria. Each year, there are up to 500 million episodes of malaria illness and over 1 million malaria deaths. More than 90% of these deaths are in Sub-Saharan Africa. A third of malaria cases and two thirds of all deaths are in young children. Malaria costs US$12 billion every year in lost productivity reduced household income, and expenditure on treatment (Allan 2000). Over a million of malaria deaths are in children aged under-five (Trigg et al 1998). 1.2 Malaria Burden in Ghana Malaria is the single most important cause of mortality. Malaria is the second major killer after HIV/AIDS. Malaria forms 40% of all outpatient visit and it causes 30% of under-five mortality in Ghana. Only 22% of malaria patients receive prompt and accurate treatment (MOH, 1999). In the Wassa West District of Ghana malaria forms 26% causes of admission and 37% of outpatient attendance (Wassa West District Annual Report 2002). 1.3 What is Malaria? Malaria is a parasitic disease caused by Plasmodium parasites, and is transmitted by the bite of the Female Anopheles mosquito. The malaria parasite is injected into humans in the saliva of the mosquito when it feeds. In the human, the parasites undergo cyclical changes in the liver and red blood cells causing the symptoms of Malaria. The clinical symptoms are fever, vomiting and diarrhea, headache, cough, chills and muscle pain. Malaria is treated with antimalarials like chloroquine, sulfadoxine-pyrimethamine(SP), artesunate and others (Marsh et al 1998). 1 University of Ghana http://ugspace.ug.edu.gh 2.0 STATEMENT OF THE PROBLEM Studies conducted in Ghana indicated that 74-97% of people use home treatment for acute malaria. Home treatment most commonly involved the use of herbs. Self-medication for malaria is so common that it is estimated that the unofficial drug sellers in markets, streets and village shops account for as much as half of the antimalarials distributed (Foster 1991). 75% of mothers manage child's disease at home. Although, most caregivers of children under-five manage malaria at home, their treatment practices are inappropriate. In Ghana current studies show that, 92.6% treatment of children with fever at home is inappropriate. Furthermore, mothers or caregivers may not be able to recognize fever in their children and start early adequate and appropriate treatment. More than 50% of mothers are unaware of the correct dosage of antimalarial drugs like chloroquine. Studies (Dunyo et al 2000) have also proven that early and appropriate treatment of malaria detected in children by caretakers may prevent complications and lower mortality risk. The proposed research would enable us know the appropriateness of caregivers treatment practices. 3.0 RELEVANCE OF STUDY Malaria still remains an important cause of mortality in children under-five. The main objective of WHO adopted Roll Back Malaria strategy is early detection and rapid treatment of malaria. The National Malaria Control Program's objective is to reduce under-five mortality. The study would help know the therapeutic knowledge and treatment practices of malaria by caregivers of children under- five. On the bases of the results caregivers can be trained and educated to recognize the symptoms of malaria early and administer appropriate antimalaria drugs. If caregivers are able to detect early and appropriately treat malaria in children under-five, mortality, which is high in this age group, can be reduced. Findings from the study will enable the District Directorate of health services develop capacity at both institutional and community levels to control malaria. 2 University of Ghana http://ugspace.ug.edu.gh CHAPTER TWO 4.0 LITERATURE REVIEW We know that, caretakers behaviour in response to signs of disease are influenced by a wide range of factors other than accessibility and availability of services, including social networks and socio­ economic factors as well as perceptions of severity of illness. Also, local diagnostic categories and local understandings of the etiology of disease affect timing of diagnosis, methods of home treatment, patterns of health seeking behaviour outside of the domestic domain (Kendall 1990). However, we now know that caretakers may recognize, interpret and act on signs and symptoms of illness in ways that are not necessarily concordant with biomedical understanding of the disease or consistent with health education designed to encourage early treatment (Nicter 1993). This suggests that, caretakers may diagnose malaria based on their own knowledge of the signs and symptoms of the disease and adopt certain treatment methods, which may or may not be in accordance with the scientific way of diagnoses and treatment of malaria. A study conducted in 2000 in southern Ghana to compare the accuracy of presumptive diagnosis of malaria in children aged 1-9 years by caretakers of the children with health center staff concluded that early and appropriate treatment of malaria detected in children by caretakers may prevent complications that arise as a result of persistence of symptoms and attainment of high parasitemic levels (Dunyo et al 2000). This means that, if caretakers are able to diagnose malaria early and accurately give correct dose of antimalarials then the parasite level in the blood will be reduced. These will prevent complications that can lead to child’s death. 4.1 Terminology related to malaria Agyepong et al (1992) described among Ga Adangme speakers in Dangme West that, fever and 'asra1 are used interchangeably to refer to a number of symptoms including hot body and chills, headache, yellow urine, yellow vomit and yellow eyes, bitter taste in the mouth, bodily pains, and loss of energy (child does not play) and refusal to play. Also in a study conducted in rural and urban subdistricts of the Greater Accra Region in 1994, (Agyepong et al 1994) it was found that, among the Ga Adangmes the local term used to approximate the biomedical definition of malaria is ‘ asra’. Another study in southern Ghana by Ahorlu et al 1997 confirmed that Fever and malaria, which are locally called Asra or Atridi, were found to represent the same thing and are used interchangeably. 3 University of Ghana http://ugspace.ug.edu.gh A similar study conducted in Tanzania found that although the local term homa ya malaria or malaria fever appeared on the surface to correspond closely with the biomedical term malaria, significant and often subtle differences were found between the two terms. It is demonstrated that the position of the local term used to denote malaria in the local taxonomy of febrile illness has important implications for the design of health education interventions (Winch et al 1996). Also a study in Mali indicated that the term soumaya was previously associated with uncomplicated malaria (Thera 2000) 4.2 Local diagnostic criteria for malaria in children (Agyepong et el 1994) it was found that, virtually all mothers used change in body temperature (‘hot body’) as one symptom of fever, rural mothers looked for other physical signs of illness (vomiting, yellow urine, yellow eyes etc), whilst urban mothers assessed the child on the basis of behaviour (refusal to eat, won’t play). These findings suggest that, mothers diagnose malaria in their children on the basis of changes in body temperature and physical and behavioural signs of illness. In another finding (Ahorlu et al 1997), caretakers were well informed about major symptoms of malaria, which corresponds to the current clinical case definition of malaria. However, on the contrary a study conducted in Guinea (Bailo Dialo et al 2001) found that, mothers often failed to identify fever in their children and to consult or to provide antimalaria treatment. Studying mothers' perceptions and knowledge on childhood malaria in Tanzania, (Tarimo et al 1998), found that the perceived symptoms of childhood malaria included fever and gastrointestinal manifestations (loss of appetite, vomiting and diarrhea) which featured as the most important symptoms with frequencies of 93.5% and 73.8%, respectively. This was followed by other physiological and behavioural symptoms such as prostration, lethargy, inactivity, coldness, shivering and sweating as well as restlessness and excessive cries, at frequencies of 49.4%, 16.2% and 15.0% respectively. Respiratory symptoms and convulsions featured as the least important symptoms of malaria, being mentioned by only 7.3% and 5.8% of respondents, respectively. Mentioning fever alone or in combination with other symptoms was significantly associated with having primary education and above. Furthermore, a study conducted in Mali found that, the three most common signs by which mothers recognized malaria in their children were vomiting, fever and dark urine/yellow eyes/jaundice (Thera et al 2000). 4 University of Ghana http://ugspace.ug.edu.gh Also in another study in Zambia, 80% of caregivers giving narratives- reported noticing, crying or irritability, diminished activity and/or decreased appetite (Baume et al 2000). 4.3 Health seeking behaviour for uncomplicated malaria Agyepong et al (1992), found that 60% of children with presumed fever among rural respondents and 76.5% among urban respondents were treated at home using either paracetamol alone or paracetamol with chloroquine or herbal medicine. Among urban respondents, 52.7% of all respondents (76.5% of those administering paracetamol) also gave the sick child chloroquine. In the rural areas, of those who reported giving a child with fever paracetamol, 65.7% also said they would administer chloroquine; 13.8% would give the sick child herbs and an additional 9.2% would give herbs in addition to other medication. Also, their findings revealed that, in the event of a child with fever not improving from self-medication, 87.4% of urban and 90% of rural respondents said they would take the child to either a clinic or hospital. Furthermore, a study conducted in Mali found that, the health seeking behaviour (whom the mother consulted first at the beginning of the illness) revealed that 75.8% of women managed their child’s disease at home and used both traditional and modem treatment, only 7.6% sought advise and treatment from private nurses or health centres. The specificity and sensitivity of mothers' diagnoses was poor. The most common antimalarial drug used was chlorquine, often given at inappropriate dosage (Mahamadou et al 1998). Similarly, recent studies conducted in eastern Uganda Nashakira et al 2002 found that before attending the study sites, 72% of children had already been given some biomedical drugs, and 40% had received the recommended drug, chloroquine. Although, chloroquine remains the first- line treatment for malaria, studies show that since the late 1980s convincing evidence of a major public health impact of the spread of chloroquine resistance has been available. Hospital studies in various African countries have documented a 2- or 3-fold increase in malaria deaths and admissions for severe malaria, an increase temporarily related to the emergence of chloroquine resistance (Trape 2001). Also a study in Ibadan, Nigeria documenting the knowledge and home management practices of 376 mothers and caregivers of under five children on malaria fever found that as regards practices, self­ medication with modern drugs was common, these drugs had been given in the home by 265(70.5%) mothers while "Agbo", had been used by 95(25.5%) mothers before presenting at the clinic. 5 University of Ghana http://ugspace.ug.edu.gh Paracetamol was the modern drug often used (217 or 81.8%) followed by chloroquine (57 or 21.5%). However, drug treatment practices were often incorrect. Chloroquine was prescribed correctly by 15(26.3%) mothers, while 109(50.2%) gave the correct dose of paracetamol. Only 16(4.3%) of the children received anti-malarial on the day the illness began. The results revealed that both the knowledge and case management practices were poor. There is the need for educational programmes on malaria for mothers, especially for young, illiterate and unskilled mothers, including the family elders (Fawole et al 2001). Glick et al 1997 found that, 33% of mothers in rural areas of Guinea reported taking their sick children to a health care worker during the last episode of fever compared with 69% of mothers in urban areas and also mothers living closer to health care facilities were more likely to consult and to give chloroquine early than mothers living farther away. However, higher proportions were found in a recent study in Uganda, where caregivers reported that if their child had fever, 63% would go to a clinic or hospital as their first action and 97% as their first or second action. (Njama et al 2003). 4.4 Home remedies for uncomplicated malaria The preferred home treatments are antipyretic and analgesic drugs and herbal preparations. Generally, antimalarial drugs are given to less than 30% of febrile children and they are mostly bought in shops (Glick et al 1997). Home treatment of malaria combining herbs and over-the-counter drugs and inadequate doses of chloroquine was widespread. There is a need for a strong educational component to be incorporated into the MAP to correct misconceptions about malaria transmission, appropriate treatment and protection of households. Malaria control policies should recognize the role of home treatment and drug shops in the management of malaria and incorporate them into existing control strategies. (Ahorlu et al 1997) Further studies (Hamel et al 2001) found that, 47% of recently febrile children received home treatment with an antimalarial and that chloroquine was included in 92% of the home treatment. They also noted that the proportion of children receiving appropriate home treatment could be increased if carers could be redirected to administer an antimalarial to children with fever. Home treatment enhances the promptness of antimalaria treatment. 6 University of Ghana http://ugspace.ug.edu.gh 4.5 Health seeking behaviour 'for complicated malaria (childhood convulsion) In the Tanzanian study (Tarimo et al 1998), it was also found that home remedy for childhood convulsions included bathing and aspirin/paracetamol and antimalaria in (17.5%), and traditional treatment (local herbs, smoking etc) in 53.2%. Subsequent action after home remedy for convulsions was to wait till fits cease before the next action in 44.2% of cases, to go to a dispensary/health centre in 86.8% and to go to a traditional healer in only 9.4% of cases. Also a study in Zambia found that, herbal treatments are more commonly used when signs of convulsions appear than when they do not and traditional healers are more likely to be consulted (Baume et al 2000). 5.0 HYPOTHESIS Reviewing the literature, it can be hypothesized that more than half of mothers or caregivers manage malaria in children under- five years at home with paracetamol, chloroquine or herbal medicine. 6.0 STUDY OBJECTIVES 6.1. General Objectives: To describe mothers or caregivers diagnoses and treatment of malaria in children under-five. 6.2. Specific Objectives: 1. To describe terminology related to malaria in under-fives. 2. To identify the basis on which mothers or caregivers diagnose malaria in children under- five. 3. To describe the treatment seeking practices for complicated and uncomplicated malaria in under- five. 7 University of Ghana http://ugspace.ug.edu.gh CHAPTER THREE 7.0 METHODOLOGY 7.1 Background of Study Area: The study area was located in the Wassa-West District in the Western Region of Ghana. The district covers a total estimated area of 3500 sq km. The population has been estimated at 233,016(year 2000 census), with a mean population of 67 per sq. km. This is higher in the central parts of the district and in the mining communities. The male:female ratio is estimated at 1:1.12. The people in the district are mainly Wassaws. They form the indigenous ethnic group, however, other tribes, the Fantes, Ewes, Ahantas, Nzemas, Asantes, Krobos, Dwiras, Sefwis and the Northerners who have migrated from other parts of Ghana are represented in small proportions. The district is currently divided into 7 operational health subdistricts. There are 4 hospitals, 7 health centers and 4 rural health centers. The study was conducted in 5 randomly selected communities from 3 of the subdistricts. They are Aboso, Dompim, Fanti Mines, Nsuaem and Pieso (see district map fig 4). Pieso is a typically rural and deprived area, whilst the others are comparatively bigger towns and have access to clinic. The main economic activities are mining, lumbering, rubber, cocoa and coffee farming. Malaria is the first among the five top causes of outpatient attendance (29.34%). Malaria is the second cause of admission cases (26%). The annual incidence of malaria for the year 2001 was 158.9/1000. Institutional data (annual report 2002 for Wassa District) indicates that the incidence of malaria in under-five for 2002 was 175.8/1000. Malaria remains a disease of public health importance in the district. This may be due to the rainfall pattern in the district. There is prolonged rainy season, starting from March to September with a very short period of dry season from October to February. Even this dry period is also interspersed with rainfall. The almost all year rainfall making most dwelling areas marshy and swampy serves as breeding sites for mosquitoes. 8 University of Ghana http://ugspace.ug.edu.gh The insanitary conditions in most towns and villages in the district is a major factor. The poor drainage system results in the creation of stagnant pools of water that breed mosquitoes transmitting malaria. Even more critical is the numerous galamsey activities in the towns and villages. The ‘dig and wash’ method used in the extraction of gold creates numerous artificial ponds, which also serve as breeding sites for mosquitoes. 7.2 Variables 7.2.1 Background Variables: Variable Name Variable Definition Variable Measurement Age of caregiver Age in completed years Continuous. Completed years Relation with the Mother, father, guardian Mother, father, grandmother child Grandfather, auntie, sister, Brother, house care, others Sex of care giver Female male Dichotomous Female Male Marital status Current marital status Single Married Cohabiting/consensual Divorced/ separated/widowed Education Highest educational level No education achieved Primary Middle school J.S.S S.S.S Technical/vocational training college Teacher training Polytechnic University The above variables are potential confounding variables in analyzing caregiver’s knowledge on signs and symptoms of malaria in under-fives and their health seeking behaviours. 9 University of Ghana http://ugspace.ug.edu.gh 7.2.2. Key Variables: Variable Name Variable Definition Variable Measurement Local term for malaria in The local term commonly The local term that children used to refer to malaria- includes the following set associated symptoms. of symptoms; hot body, loss of appetite, diarrhea & vomiting, yellow urine. Local diagnostic criteria for The signs and symptoms by Fever (hot body) malaria which mothers or caregivers Vomiting identify malaria. Diarrhea Loss of appetite Child not playing Convulsion Chills Yellowish urine/eyes Others Health seeking behaviour for Whom the mother/care giver Nothing/wait and see uncomplicated and complicated consulted first at the Go to hospital malaria (convulsions) beginning of the illness Go to clinic Go to the chemical seller Chloroquine Paracetamol Go to herbalist Herbal treatment Enema Others Key Variables cont’d 10 University of Ghana http://ugspace.ug.edu.gh Variable Name Variable Definition Variable Measurement Home remedies for Treatment given at home by Sponging/bathing Malaria Mothers/caregivers Chloroquine Paracetamol Herbal treatment Enema Duration of first action How long the mother or Same day caregiver does the first 2 days action 3 days 1 week Others Next action taken The subsequent action taken Nothing/wait and see By mother/caregiver if child Go to clinic does not recover. Go to hospital Go to herbalist Go to chemical seller Chloroquine Paracetamol Herbalist Enema Others Causes of convulsion What mothers/caregivers Fever think is the cause of Phlegm childhood convulsion. Malaria Others 11 University of Ghana http://ugspace.ug.edu.gh 7.3 Study Design The study was Cross-sectional and descriptive. The data was collected by both qualitative and quantitative methods. Qualitative Design: Four Focus Group Discussions were conducted in Dompim, which is one of the five randomly selected communities. Grandparents and parents of children under-five were selected for the FGD. The focus group discussions were grouped into grandmothers, grandfathers, mothers and fathers groups. There were 8 participants in each group. The background of the participants in each FGD namely, ages, marital status, educational level, number of children, and occupations were collected. Quantitative Design: Structured questionnaire interviews of 300 mothers or caregivers of children under-five were conducted in Dompim, Nsuaem, Aboso, Pieso and Fanti Mines communities proportional to their various populations to supplement the qualitative data, i.e. to validate and check for the representativeness of data obtained from the qualitative research. 7.4 Ethical consideration The ethical committee of MOH was consulted for approval. Also, approval was obtained from DHMT, District Assembly, and the Traditional Council. Verbal consent was obtained from the heads of the households and individual study units before interviews. 7.5 Data collection A one-day training session was held for the field assistants. In all, 6 persons were trained as field assistants who assisted in data collection. 4 community health nurses and 2 workers from NADMO were trained. The FGD was first conducted in the selected communities with the help of 2 trained field assistants and the results used to back-up the structured questionnaires. The researcher was the moderator and the trained person the note taker. A FGD Guide was used and the responses of participants recorded with a tape recorder and note taking by the note taker. The trained community health nurses administered the structured questionnaires. The researcher also took part in the administering of the questionnaires. 12 University of Ghana http://ugspace.ug.edu.gh 7.6 Pretesting of Instruments in a Community The FGD guide was translated to the local language and pre-tested for feasibility and clarity and the results used to update the guide and survey questionnaires before the main study. The participants of the chosen community for pre-testing and that of the main study communities had similar characteristics like age, sex, occupation, and tribe/language. 7.7 Sample Size Calculation: From review of literature, the proportion of children under-five with malaria managed at home by caregivers = 75%. n = sample size = z2p(l-p)p2 z = 1.96 at 95% confident interval P= Desired difference between observed proportion and true proportion =5%(95%CI) n= Sample size = 288(300) 7.8 Sampling The target population was caregivers of children under-five. A caregiver is defined as the parent/guardian who provides the daily essential needs of the child such as bathing, feeding, clothing, sending to school or hospital when sick. As already described above Wassa West District has been divided into 7 operational health subdistricts. 3 of the subdistricts were selected by simple random sampling method. The names of the communities in the subdistrict were written on pieces of papers, folded and put in a closed container and shaken vigorously. 5 communities; Dompim, Nsuaem, Aboso, Pieso and Fanti Mines were then selected randomly. In the selected communities, the first household was selected by locating the approximate center of the community and choosing the nearest compound to identify the household. Then every third household was selected from the first. In each household, one caregiver with at least a child under- five years was interviewed with a structured questionnaire. The number of respondents interviewed in each community was determined based on a proportion of the population of the various communities. 13 University of Ghana http://ugspace.ug.edu.gh The table below illustrates the proportion of the population interviewed in each of the communities selected. Table 0: The proportion of population interviewed in the communities Community Population (X) Sample of population interviewed N = X/T*300 Aboso 9737 132 Nsuaem 5250 70 Dompim 4749 64 Pieso 2202 30 Fanti Mines 267 4 Total (T) 22205 300 7.9 Data analysis The qualitative data was analyzed manually by transcription, coding, and sorting. Epi Info version 5 was used to analyze the quantitative data. 14 University of Ghana http://ugspace.ug.edu.gh CHAPTER FOUR 8.0 RESULTS As already stated both FGDs and cross-sectional household survey m ethods were used to collect the data. Four focus group discussions were conducted in Dompim, which is one of the randomly selected five communities where the survey was conducted. The findings are presented below. 8.1 Characteristics of study communities and population 8.1.1 The study community characteristics Out of the total of 300 respondents interviewed, 133(44.3%) were from Aboso, 71(23.7%) from Nsuaem, 64(21.3%) from Dompim, 29(9.7%) from Pieso and 3(1%) from Fanti Mines. This is presented in table 1 below. Table 1: The communities and the respondents interviewed. Community No. of respondents Percentage % Aboso 133 44.3 Nsuaem 71 23.7 Dompim 64 21.3 Pieso 29 9.7 Fanti Mines 3 1 Total 300 100 15 University of Ghana http://ugspace.ug.edu.gh 8.1.2 Access to health facility A total of 268(89.3%) respondents had access to clinic, whilst 32(10.7%) had no access to a health facility. This is presented in fig 1 below. Respondents accessibility to health facility 11% □ Present ■ Absent 89% Fig 1: Respondents access to health facility The figure shows that 89% of respondents had access to health facility whilst 11 % did not. N.B Access to a health facility in this study was defined as whether a clinic was present in the community or not. 16 University of Ghana http://ugspace.ug.edu.gh 8.1.3 Characteristics of participants in FGD The focus group discussions were held with mothers, fathers, grandfathers and grandmothers of children under-5. The groups were averagely made up of 8 participants. The issues discussed included knowledge, perceptions and treatment practices of malaria in small children. The characteristics of the mothers group are presented in table 2 below. The characteristics of the other groups are presented in tables 17, 18 and 19 (see annex). Table 2:Characteristics of the mother’s group in the focus group discussion No. Age Marital Status Occupation Educational Level Tribe No. of Children 1. 34 Married Trader Primary Wassa 2 2. 28 Married Trader Primary Wassa 4 3. 24 Married Trader Primary Wassa 2 4. 30 Divorced Seamstress Primary Wassa 4 5. 30 Divorced Trader None Wassa 4 6. 30 Divorced Farmer None Dwura 5 7. 29 Married Seamstress J.S.S Nzema 3 8. 25 Married Trader J.S.S. Wassa 1 8.1.4 Age sex distribution of respondents In the cross-sectional survey the minimum age of respondents was 15 years and the maximum 75 years. The mean age of the respondents was 31.74 with a standard deviation of 10.74 and the modal age 30 years. Out of the 300 respondents, 261(87.0%) were females whilst 39(13.0%) were males. 17 University of Ghana http://ugspace.ug.edu.gh 8.1.5 Ethnic distribution of respondents The Wassaws were the highest represented ethnic group among the respondents, constituting 141(45%) of the total respondents of 300. This was also the same with the participants of the FGDs, where the majority were Wassaws. Table 3 below illustrates the ethnic distribution. Table 3: Ethnic Distribution of respondents Ethnic group No. of respondents Percentage % Wassaw 141 47 Asante 47 15.7 Nzema 40 13.3 Fante 33 11.0 Ewe 12 4.0 Northerner 11 3.7 Pepesa 4 1.3 Others 12 4.0 Total 289 100 8.1.6 Marital status of respondents Out of the 299 respondents interviewed, 178 (59.5%) were married, 56 (9.7%) cohabiting, 36 (12%) divorced or separated or widowed and 29 (9.7%) single. 8.1.7 Educational status The highest educational level achieved by the respondents was university level forming 0.3% of the total 300 respondents. However, most of the respondents completed middle school and Junior Secondary School both representing 75 (25%) of the total respondents. 67 (22.3%) had no formal education at all, whilst 61(20.3%) completed only primary level. The rest completed senior secondary school, Technical/vocational training, teacher training and polytechnic representing 10 (3.3%), 6 (2.0%), 4 (1.3%) and 1 (0.3%) respectively. In the FGDs majority of the participants had no formal education. 18 University of Ghana http://ugspace.ug.edu.gh 8.1.8 Occupational status Of the total of 298 respondents interviewed, majority were petty traders forming 36.6%, peasant farmers 24.5% and a significant proportion of 21.1% were unemployed. The rest were hairdressers 14 (4.7%), teachers 12 (4.0%), students 3 (1.0%) and others 13 (4.4%). However, majority of the participants in the FGDs were farmers and only a few were traders. 8.1.9 Relationship of respondents with child The distribution of respondent’s relationship with the child in the cross-sectional survey is summarized in table 4. Table 4: Relationship of respondents with child Relationship No. % Mother 237 79.3 Father 30 10.0 Grandmother 22 7.4 Grandfather 3 1.0 Auntie 3 1.0 Sister 1 0.3 Brother 1 0.3 Other 2 0.7 Total 299 100% As noted, survey participants were caregivers of a child or children under-5 years. Majority of the respondents were mothers constituting 79.3%. 8.1.10 Knowledge of malaria All the respondents (100%) knew what malaria was. 19 University of Ghana http://ugspace.ug.edu.gh 8.1.11 Perception of malaria In the cross-sectional survey, people were asked whether malaria worries them or not and out of a total of 298 respondents, as many as 277(93%) said malaria worries them, whilst only 20(6.7%) said malaria does not worry them and 1(0.3%) did not know whether malaria worries them or not. 8.2 Terminology related to malaria All the participants in the focus group discussions conducted agreed that the local term they use to refer to malaria is “ebun ", The term "fever" is used interchangeably with "ebun”to refer to a set of symptoms including hot body and chills, vomiting, yellowish urine and stool, yellow eyes, the child not eating, the child weak, sleeping and not playing. Most of them use the three terms “malaria”, “fever” and “ebun”interchangeably to mean the same. As a participant put it, “fever is the commonest name all of us use, but malaria is what the hospital people use. “Ebun ” is the local term we use here”. These were confirmed in the cross-sectional survey, where out of a total of 298 people interviewed as many as 202 (67.7%) respondents said the local name for malaria was "ebun", whilst 120 (40.2%) said "fever” and only 12 (4.0%) mentioned "atridii”. 24 (8.1%) of the respondents did not know the local name for malaria. (See table 5 below) Table 5: Local names for malaria (n = 298*) Local name Frequency % Ebun 202 67.7 “fever” 120 40.2 “atridii” 12 4.0 Don’t know 24 8.1 *Total number o f names mentioned is more than the sample size due to multiple responses Among other terms mentioned during the FGDs was "niwura ”, this is used to refer to a set of signs and symptoms that approximate the biomedical disease defined clinically as complicated malaria, although the respondents claim is not the same as malaria. The signs mentioned included very hot body, excessive cry and shouting, static attacks and convulsion. As one woman described it, "niwura starts with hot body (as hot as burning fire, hotter done that of malaria), the child shouts and cries, becomes spastic and holds you firmly”. Another woman also said, “ if my child has "niwura" he has 20 University of Ghana http://ugspace.ug.edu.gh static attacks if he hears someone shouting, and if I carry him on my back and a car passes-by, the child becomes spastic and grips me firmly”. Other local terms used synonymous to complicated malaria were “atowosu", “asinsin" and “abobrim". 8.3 Mothers or caregivers diagnostic criteria for malaria in children under-five years Mothers or caregivers diagnose malaria in small children based on their own knowledge of certain signs and symptoms observed in the children. During the focus group discussions, all the participants agreed that they diagnose malaria in their children if the child’s body is hot. “Ahodo”is the local term used to refer to a rise in body temperature, which is synonymous to English clinical term fever. As one of the participants in the focus group discussion said and I quote “ if malaria attacks my child the child becomes hot and if I give some medicine he becomes cold and after sometime becomes hot again”. Another participant also said, “if I am breast feeding the child and I feel the mouth of the child is hot then I know my child has malaria”. A father also said “I use my hand to touch the head to feel whether the child is hot, if the child is hot then I know the child has malaria”. The other symptoms mentioned were vomiting, yellowish urine and stool, yellow eyes, mouth and palm, child weak and not playing, child doesn’t eat. As a participant said, “if my child has malaria the toilet becomes yellow and even stains my cloth”. Another also said, “if my child has malaria and wakes up in the morning the face becomes “bosaa” (meaning ill looking face), if I give him food he doesn’t eat and vomits every food and even water given”. “In the morning the child plays but in the evening the child becomes weak and sleeps, the body becomes hot and child refuses breast milk, urinates frequently then I know child has malaria”; a participant said. Most mothers were also able to recognize complicated malaria. The signs recognized were convulsion and anaemia. As one woman puts it, “if the child has malaria the child becomes very hot and if you don’t see it early and sponge the child with cold water, convulsion attacks the child. The child’s face becomes white and the more the child becomes hot the more the child’s blood falls”. A grandmother also said, if “ebun" attacks my grandson, the eyes become white and the child’s blood reduces”. The same findings were obtained during the cross-sectional survey where it was found that, mothers or other caregivers diagnosed "ebun" or “fever" in small children based on one or more symptoms. 17.6% of the respondents diagnosed malaria based on only one symptom. Out of this, 10.7% mentioned change in child’s body temperature-fever as the only sign of diagnosing malaria, 21 University of Ghana http://ugspace.ug.edu.gh whilst the remaining 6.9% used other single symptoms like vomiting, loss o f appetite, convulsion, yellow urine/eyes and white face, hands and feet. Majority of the mothers and other caregivers 138 (46.3%) diagnosed malaria on the basis of two symptoms only and the rest 109 (36.3%) diagnosed malaria on the basis of 3 or more symptoms. The three most common signs by which mothers and other caregivers recognized malaria was found to be fever, vomiting and loss o f appetite representing 32 (10.7%) of the total respondents of 297. The percentage o f mothers and other caregivers who perceived symptoms o f malaria in small children to be fever was found to be 91%. Furthermore, the percentage of mothers or caregivers who perceived symptoms of childhood malaria to be gastrointestinal manifestations (lost o f appetite, vomiting and diarrhoea) was found to be 73.9%. The other symptoms such as convulsion, chills, child not playing/child sleeping a lot, yellow urine/eyes, white face, hands and feet and headache were mentioned by 32.2% of the respondents. (Tables 6 & 7 below) Table 6: Perceived symptoms of malaria in small children (respondents n = 297)* Symptoms No % Fever 273 91.3 Vomiting 147 49 Lost of appetite 116 38.8 Child not playing 86 14.3 Diarrhoea 53 18.5 Yellow urine/ eyes 25 8.4 Convulsion 17 5.6 White face, hands and feet 2 0.6 *Total number o f symptoms is more than the sample size due to multiple responses 22 University of Ghana http://ugspace.ug.edu.gh Table 7: Number of signs mentioned as indicating malaria in small children Number o f respondents giving: No % 1 sign of malaria 51 17.6 2 signs of malaria 138 46.3 3 signs 74 24.7 4 signs 34 11.3 Total 297 100 8.4 Treatment seeking practices for uncomplicated and uncomplicated malaria 8.4.1 Treatment seeking practices for uncomplicated malaria In the focus group discussions, most mothers said that they first give paracetamol and chloroquine syrup to their small children at the onset of "fever". If the child does not recover, they send the child to clinic or hospital. As one mother puts it, “ if my child is sick I first give para, malarex, B’co and sponge him for 3 days, if he does not improve I send him to hospital”. A father also said, “as for malaria we normally give first aid by giving chloroquine, paracetamol, and multivite and if the child does not recover before we send to clinic”. However, some participants claimed that they first give the child enema and herbal drink if there is no improvement, and then they send the child to clinic. A grandmother said and I quote, “ we were brought up with herbs so I will first use herbs, if the child does not recover before I would send to hospital”. Another also said, “I will go to the farm and collect herbs for the child”. As a father puts it, “if my child has “ebun ” I first use small herbs to give the child enema and if the child does not recover then I send to clinic”. A father also claims “because of financial problems we resort to herbal medicine and give the child enema and if the child does not recover before we send to hospital, but if there is money we send the child to hospital straight”. Some also said they go to the chemical seller to buy “fever " medicine for the child. The chemical seller usually sells chloroquine and paracetamol syrup over the counter if they describe the symptoms of the child: “if my child is hot I will go and buy medicine at the drug store”, a participant said. A participant also said, “we send the children to the drug store because the chemical seller have been trained to know what is wrong with the child and give treatment”. Another also maintains that, “if my child has fever I go to the drug store to buy a previously prescribed drug at the clinic for the same condition”. 23 University of Ghana http://ugspace.ug.edu.gh The majority of the participants in the focus group discussion agreed that they do not send their children to the herbalist. As one of the participants put it, “as for herbalist I don’t have fowl so I will not send the child there”. Almost all the participants disagreed going to the “injectionist”. A grandmother said, “ if I send my child to clinic for injection nothing bad will do the child but if the child takes injection outside the clinic the child may paralyse A grandfather also claim he saw a child die after receiving injection from an injectionist in the house. A few send their children to the priest because there is some illnesses that are not hospital sickness so need prayers from the priest to heal the child. The findings from the FGDs were confirmed in the cross-sectional survey where it was also found that, the majority of the respondents reported giving some form of self-medication at home at the onset of "fever” in their small children. 60.4% of mothers or caregivers treat children under-5 years with malaria at home with either modem or traditional medicine. At the onset of “fever" in their small children, 48.4% of the respondents gave paracetamol to their children at home, 35.2% gave chloroquine and 11% gave herbal treatment, which included herbal drinks and enema. The drugs were either used alone or in combination. As for example, 10.7% used paracetamol alone, 3.7% chloroquine alone and 27.9% combined paracetamol with chloroquine. There was also combination of European medicine and that of herbs. For example, some combined chloroquine with enema 0.3%, paracetamol with enema 3.4% and chloroquine with herbal drink 0.3%. Only 26.2% of mothers or caregivers sent their children under-5 years to the clinic or hospital as first line of action in the event of their child having malaria. Also 13.4% o f the respondents took their sick children to the chemical seller and 1% to the herbalist for treatment as the first line of action. (See table 8 below). 24 University of Ghana http://ugspace.ug.edu.gh Table 8: First treatment/actions for small children with malaria (n = 298)* T reatment/action No. % Go to hospital/clinic 78 26.2 Go to chemical seller 40 13.4 Paracetamol 144 48.4 Chloroquine 105 35.2 Sponging/bathing 16 5.3 Traditional treatment (herbal drink, enema) 33 11 Go to herbalist 3 1 T o ta l actions exceed number o f respondents as respondents undertook two or more actions concurrently The first line of action taken by a mother or caregiver if child has malaria was strongly associated with the educational level o f the mother or caregiver (Chi square =160 P value = 0.00). Respondents with educational level above primary are more likely to send their sick children to clinic/hospital at the onset of the illness than those without any formal education. (See table 9 below). Table 9: Comparison of first treatment/actions for malaria in a child and educational level of Respondents Treatment/action Educational level of respondents No formal Primary Above primary Total education education Education Go to clinic/hospital 17 (28%) 12 (23.5%) 49(31.6%) 78 (29.3%) Go to chemical seller 7(11.7%) 9 (17.6%) 13 (8.4%) 29(10.9%) Home treatment with paracetamol, 36 (60%) 30 (58.8%) 93 (60%) 159 (59.8%) chloroquine or herbs at home Total 60 (100%) 51 (100%) 206 (100%) 266 (100%) Chi square = 112.79 p value = 0.00 Table 9 shows that 28.3% of respondents with no formal education sent their children to hospital or clinic at the onset of the illness whilst 31.6% of those with educational level above primary sent their ill children to hospital or clinic as first line of action. 25 University of Ghana http://ugspace.ug.edu.gh The first line o f treatment/action by mothers or caregivers was also significantly associated with the duration/time the treatment or action was taken. This is presented in table 10 below. Table 10: Comparison of first treatment or action for child with malaria and duration/time the treatment/action was taken Treatment/action Duration/time o f treatment/action taken Same day of 2 days 3 and more Total illness days Go to clinic/hospital 43 (53.1%) 19(18.6%) 15(13.2%) 77 (25.9%) Go to chemical seller 19(23.5%) 7 (6.9%) 14(12.3%) 40(13.5%) Home treatment with paracetamol, 19 (23.5%) 76 (74.5%) 85 (74.6%) 180 (60.6%) chloroquine or herbs Total 81 (100%) 102(100%) 114(100%) 297*(100%) Chi square = 2 3 0 .5 0 * less than the sam ple size due to non- responses Degrees o f freedom =100 P value =0.00 Table 10 indicates that 53% o f respondents sent their ill children with malaria to hospital or clinic on the same day o f child’s illness, 18.6% after 2 days, 13.2% after 3 or more days and overall 26% sent their ill children to hospital or clinic. 23.5% of respondents also went to the chemical seller the same day of child’s illness. Furthermore, 74.6% of mothers or caregivers treated children under-5 at home with paracetamol, chloroquine or herbs for 3 days. 26 University of Ghana http://ugspace.ug.edu.gh There was also a significant association between duration/time mothers or caregivers took the first line of treatment or actions when their small children had malaria and educational level achieved. (See table 11 below). Table 11: Comparison of duration/time first line of treatment or action was taken and education of mothers/caregivers. Duration/time first treatment/action Educational level of respondents was taken No formal Primary Above primary Total Education Education Education Same day of illness 20 (29.9%) 18(30%) 44 (25.9%) 82 (27.6%) 2 days 27 (40.3%) 15(25%) 61 (35.9%) 103 (34.7%) 3 and more days 20 (29.9%) 27 (45%) 65 (38.2%) 112(37.7%) Total 67(100%) 60(100%) 170(100%) 297* (100%) An expected value is < 5. chi square not valid *less than sample size due to non-responses. Chi square = 31.38 Degrees o f freedom = 40 P value = 0.833 Table 11 shows that of those of the respondents who took the first line of treatment or action on the same day of the illness in the child, 29.9% had no formal education, 30% had completed primary level, 25.9% had above primary education. Overall 27.6% took the first treatment/action the same day of the illness. Furthermore, o f those respondents who took the first treatment/action for 3 or more days 29.7% had no formal education whilst 38.2% had above primary level education. However, there was no association between the duration/time the first line o f treatment or action was carried out and the knowledge of malaria. (Chi square = 0.00: p value = 1.00) 27 University of Ghana http://ugspace.ug.edu.gh However, in the event child did not recover with self-medication, 85% of mothers or caregivers said they would send their children with malaria to clinic or hospital, 8.6% said they would seek herbal treatment and 2.7% would go to the chemical seller. (See table 12 below) Table 12:Subsequent treatments/actions taken when child did not recover with first line of action T reatment/actions No. % Go to clinic/hospital 254 84.9 Seek herbal treatment 26 8.6 Go to chemical seller 8 2.7 Home treatment with chloroquine and paracetamol 5 1.7 Others (wait and see) 6 2.1 Total 299 100 There was an association between the subsequent treatments or actions taken by mothers or caregivers if the child did not recover from the first treatment or action with accessibility to health facility. (Table 13 below). N.B Accessibility to health facility was defined as the presence or absence of a clinic in the community. Table 13 shows that, of a total of 262 respondents who had access to clinic, 85.9% took their sick child to clinic/hospital as the second line o f treatment/action, 2.7% went to the chemical seller, whilst 11.4% self medicated at home with chloroquine, paracetamol or herbs. Table 13: Comparison of second treatment/actions if child not recovered and access to clinic Second treatment/actions Clinic Present Absent Total Go to clinic/hospital 225 (85.9%) 28 (90.3%) 253 (86.3%) Go to chemical seller 7 (2.7%) 1 (3.2%) 8 (2.7%) Home treatment with chloroquine, paracetamol 30(11.4%) 2 (6.5%) 32(11.0%) or herbs Total 262(100%) 31 (100%) 293*(100%) An expected value is < 5. Chi square not valid Chi square= 10.97 Degrees of freedom = 11 P value= 0.446 ♦Less than sample size due to non-responses. 28 University of Ghana http://ugspace.ug.edu.gh 8.4.2 Treatment seeking practices for complicated malaria (childhood convulsions) Most of the participants in the focus group discussions claim they would self-medicate with herbs first and if the child doesn’t recover before sending to clinic or hospital. They usually send the child to “a person in the community who knows the herb for treating convulsion”. A grandmother said “I give enema first for the child to defeacate all the ‘phlegm in the body’ and squeeze the liquid from the herb into the child’s nostrils for the child to sneeze and wake up”. A grandfather also said, “As for here we have people who know herbal medicine to treat childhood convulsion so we don’t normally send the children to hospital. The medicine man plugs the herbs and squeezes it’s fluid into the nostrils of the child causing the child to vomit all the phlegm in the stomach and head and after, the child gets up and start playing”. A few sponge or bath the child and rush to the clinic. Others turn the child’s head upside down before running to the hospital. The man usually carries the child because “the man is brave and can run faster”. Similar findings were obtained during the cross-sectional survey. Mothers’ or caretakers perception and knowledge on the management of childhood convulsions are shown in table 14. Table 14: Perceptions and knowledge on management of childhood convulsions (n = 300*) Remedies/actions for childhood convulsions No. % Go to hospital/clinic 186 62.2 Herbal treatment at home 95 31.8 Go to herbalist 12 4 Bathing & paracetamol & chloroquine 8 2.6 Others 6 2 *Total number o f remedies/actions exceeds sample size due to multiple responses Table 14 shows that, 62.2% of respondents sent children with convulsion to clinic/hospital whilst 2.6% self-medicated with sponging/bathing, chloroquine and paracetamol, 31.8% with herbs. 4.0% of respondents send the child with convulsion to the herbalist. 29 University of Ghana http://ugspace.ug.edu.gh 8.4.3 Perceived causes of childhood convulsions Almost all the participants in the focus group discussions mentioned phlegm in the child’s body as the main cause o f childhood convulsion. As a participant puts it, “as for this illness everybody knows is caused by phlegm in the stomach of the child and every child is supposed to get it”. The phlegm they claim is got from eating only starchy foods like cassava, rice etc. Another also said “I always give my children palm oil and fish so that they do not get convulsion. A few mentioned hot body and malaria as the cause of convulsion in children. A participant also said “worm can cause convulsion”. Another cause mentioned was that “if during pregnancy the mother runs unnecessarily, the water around the baby in the womb will enter into the baby so after delivery the child will be having convulsion”. In the cross-sectional survey, the perceived causes of childhood convulsions mentioned by the respondents are presented in table 15 below. Table 15: Mothers or caregivers perceived causes of childhood convulsions (n = 300*) Perceived causes of childhood convulsion No. % Fever (hot body) 114 38.1 Phlegm in the child’s body 178 59.5 Malaria 25 8.2 Worms 3 0.9 Don’t know 43 14.4 Others 4 1.3 *Total causes o f convulsions mentioned exceeds sample size due to multiple responses Table 15 indicates that majority of the respondents, 59.5% mentioned phlegm in the child’s body as the cause of childhood convulsion, 38.1% said fever (hot body) and only 8.2% said malaria. 14.4% did not know the cause o f convulsion. 30 University of Ghana http://ugspace.ug.edu.gh The main reasons respondents sent their children to the hospital/clinic are presented in table 16. Table 16: Reasons mothers/caregivers send sick children to hospital/clinic Reasons No. % For better treatment 120 65.2 I do not know the cause/treatment for disease 31 16.8 Child not recovered from self-medication 24 13.1 For injection 3 1.6 Others 6 3.3 Total 184* 100 * Less than sample size o f 300 because only respondents who sent their children to hospital were interviewed Table 16 shows that, the majority of mothers or caregivers, 65.2% said they send their children to hospital/clinic for better treatment, 16.8% said because they do not know the cause or treatment of the disease, 13% because the child did not recover with self-medication and only 1.6% went to hospital/clinic for injection. 31 University of Ghana http://ugspace.ug.edu.gh Mothers or caregivers were also asked the main reasons they treat their sick children at home. Their responses are presented in fig 3 below. Reasons child treated at home bO -55- 50 U) o> > U i 40 - 0)