See discussions, stats, and author profiles for this publication at: https://www.researchgate.net/publication/43181049
School-based participatory health education for malaria control in Ghana:
Engaging children as health messengers
Article  in  Malaria Journal · April 2010
DOI: 10.1186/1475-2875-9-98 · Source: PubMed
CITATIONS READS
58 111
10 authors, including:
Irene Ayi Daisuke Nonaka
Noguchi Memorial Institute for Medical Research University of the Ryukyus
69 PUBLICATIONS   512 CITATIONS    36 PUBLICATIONS   252 CITATIONS   
SEE PROFILE SEE PROFILE
Masamine Jimba Kwabena M Bosompem
The University of Tokyo Leiden University Medical Centre
288 PUBLICATIONS   2,608 CITATIONS    73 PUBLICATIONS   1,276 CITATIONS   
SEE PROFILE SEE PROFILE
Some of the authors of this publication are also working on these related projects:
SATREPS View project
TB/HIV in Cambodia View project
All content following this page was uploaded by Daniel Boakye on 15 May 2014.
The user has requested enhancement of the downloaded file.
Ayi et al. Malaria Journal 2010, 9:98
http://www.malariajournal.com/content/9/1/98
R E S E A R C H Open AccessRSecseharcohol-based participatory health education for 
malaria control in Ghana: engaging children as 
health messengers
Irene Ayi1, Daisuke Nonaka2,3, Josiah K Adjovu4, Shigeki Hanafusa5, Masamine Jimba2, Kwabena M Bosompem1, 
Tetsuya Mizoue3, Tsutomu Takeuchi6, Daniel A Boakye1 and Jun Kobayashi*5
Abstract
Background: School children have been increasingly recognized as health messengers for malaria control. However, 
little evidence is available. The objective of this study was to determine the impact of school-based malaria education 
intervention on school children and community adults.
Methods: This study was conducted in the Dangme-East district of the Greater Accra Region, Ghana, between 2007 
and 2008. Trained schoolteachers designed participatory health education activities and led school children to 
disseminate messages related to malaria control to their communities. Three schools and their respective communities 
were chosen for the study and assigned to an intervention group (one school) and a control group (two schools). 
Questionnaire-based interviews and parasitological surveys were conducted before and after the intervention, with 
the intervention group (105 children, 250 community adults) and the control group (81 children, 133 community 
adults). Chi-square and Fisher's Exact tests were used to analyse differences in knowledge, practices, and parasite 
prevalence between pre- and post-intervention.
Results: After the intervention, the misperception that malaria has multiple causes was significantly improved, both 
among children and community adults. Moreover, the community adults who treated a bed net with insecticide in the 
past six months, increased from 21.5% to 50.0% (p < 0.001). Parasite prevalence in school children decreased from 
30.9% to 10.3% (p = 0.003). These positive changes were observed only in the intervention group.
Conclusions: This study suggests that the participatory health education intervention contributed to the decreased 
malaria prevalence among children. It had a positive impact not only on school children, but also on community 
adults, through the improvement of knowledge and practices. This strategy can be applied as a complementary 
approach to existing malaria control strategies in West African countries where school health management systems 
have been strengthened.
Background infections are treated at home and are, therefore, not
Ghana is located in sub-Saharan Africa, where an esti- often reported [1,2]. In Ghana, the national malaria con-
mated 90% of the world's malaria-attributable deaths trol programme focuses chiefly on pregnant women and
occur. In Ghana, malaria accounts for more than 44% of children under five years of age, as malaria leads to more
reported outpatient visits and an estimated 22% of deaths serious consequences in this group. The main activities of
in children under the age of five. Reported malaria cases the malaria control programme in Ghana are facility-
represent only a small proportion of the actual number of based and implemented at the health centre level, i.e.,
episodes, as the majority of people with symptomatic providing intermittent preventive treatment (IPT) and
distributing insecticide-treated bed nets (ITNs) to preg-
* Correspondence: j-kobayashi@it.ncgm.go.jp nant women during antenatal care, and distributing ITNs
5 Bureau of International Medical Cooperation, National Center for Global to women with young children during immunization [3].
Health and Medicine, 1-21-1 Toyama, Shinjuku, Tokyo, Japan
Full list of author information is available at the end of the article© 2010 Ayi et al; licensee BioMed Central Ltd. This is an Open Access article distributed under the terms of the Creative Commons At-
BioMed Central tribution License (http://creativecommons.org/licenses/by/2.0), which permits unrestricted use, distribution, and reproduction in any
medium, provided the original work is properly cited.
Ayi et al. Malaria Journal 2010, 9:98 Page 2 of 11
http://www.malariajournal.com/content/9/1/98Children of school-going age have also been targeted Methods
for malaria control in some endemic countries including Study site
Ghana [4-6]. As more and more children attend school, The study was conducted in the Dangme-East District of
governments are increasingly recognizing the importance the Greater Accra Region, Ghana, between July 2007 and
of child health for educational achievement [7]. A num- June 2008. Dangme-East District is located approximately
ber of studies that focus primarily on evaluating the effec- 100 km east of the capital, Accra. The district is partially
tiveness of providing treatment for children have been rural and partially urban in nature and covers an area of
conducted in the school setting, [5,8,9]. Recently, the 721 km2. The indigenous Ga-Dangme people form the
impact of IPT among school children has drawn majority of the residents in the area. According to the
increased attention [10-12]. However, while IPT is rele- nation's 2000 census, the population of the district is
vant only in high-transmission areas, skills-based health 93,193. Annual rainfall ranges between 740 and 900 mm.
education for malaria control is recommended to be The highest and lowest mean monthly temperatures are
effective in all transmission settings [13]. approximately 30°C and 26°C, respectively. The relative
Recently, school-based health education interventions humidity throughout the year ranges between 65 and
have been conducted for malaria control. They use an 80%. The region's vegetation is basically coastal savannah,
innovative approach that engages school children to consisting of grass with isolated patches of trees and
reach community adults with health messages and shrubs. There are wet and dry seasons; the wet season
hygienic practices through action-oriented and participa- begins in May and ends in October. The district has four
tory learning action (PLA). For example, Okabayashi et al government-run health centres and a district hospital.
[14] reported that school children disseminated informa- According to the data available at the District Health
tion on malaria to the community through a variety of Office, in 2006, the malaria prevalence of clinically diag-
approaches including issuing newsletters, placing bill- nosed and laboratory-confirmed cases among outpa-
boards, and holding village events in Thailand. As a tients in the district was 45.0% (19,273/42,974).
result, both children and community adults showed In collaboration with local education and health
improved knowledge, attitudes, and practices pertaining authorities, three primary schools (Afiadenyigba, Dor-
to malaria. Onyango-Ouma et al [15] evaluated the gobom and Tojeh) were purposively selected from the 15
potential of school children as health change agents in a located within the Kasseh-East Circuit of the Dangme-
rural community in Kenya and observed improved East District Education Directorate of the Ghana Educa-
knowledge pertaining to malaria among children and tion Service. The selection was based on specified criteria
guardians. In Lao PDR, Nonaka et al [16] demonstrated including: similar ecological setting, school master's will-
that school children reached out to and influenced ingness to participate in the study, and school master's
women who were not caregivers of target school children recognition of malaria as one of the major problems
to improve their behaviour in relation to malaria control. affecting school children. The selected schools are
These studies consistently reported that school children located within several kilometres of each other, in rural
were not merely recipients of health education, but also settings (Figure 1). The Afiadenyigba village and school
contributed to malaria control by playing the role of
health change agents in the community. In the present
study, this approach was employed in a different area in
order to confirm the effectiveness of the strategy, with the
consideration that socio-cultural factors in the different
area might influence its effectiveness.
The West African Centre for International Parasite
Control (WACIPAC) was established at the Noguchi
Memorial Institute for Medical Research of the Univer-
sity of Ghana in 2004, with support from the Japan Inter-
national Cooperation Agency. As part of the strategy to
control parasitic diseases in the West African sub-region
of sub-Saharan Africa, the WACIPAC set up a parasite
control project model site in the Ada-Foah sub-district of
the Dangme-East District to facilitate capacity building
and evaluate intervention strategies. In this study, the Figure 1 Map of the Dangme-East District showing the study vil-
objective was to determine the impact of school-based lages. Small map inserted at the upper right shows Ghana; Dangme-
malaria education intervention, not only on children but East District is highlighted in green. Green circles indicate the locations 
also on adult community members. of the study villages and schools (Afiadenyigba, Dorgobom and Tojeh).
Ayi et al. Malaria Journal 2010, 9:98 Page 3 of 11
http://www.malariajournal.com/content/9/1/98served as the intervention area whilst the Dorgobom and Post-teacher training activities
Tojeh villages and schools were used as controls. Accord- The research team (IA, DN, JKA, and SH) provided pic-
ing to the sample size calculation, approximately one ture charts and posters on malaria transmission and pre-
hundred children were required in each of the interven- vention, which were requested by teachers and used in
tion and control groups. As neither the Dorgobom nor teaching children in the participating grades. The posters
Tojeh schools had enough children to reach the required were displayed on the walls of their classroom after
sample size, the two control schools and their respective teaching. Teachers guided the children through dramatiz-
communities were grouped as one. ing the transmission of malaria and prevention methods.
The teachers then led the children to observe their school
Study populations compound and cleared possible mosquito breeding sites
The study population consisted of 186 children in grades such as open cans and dumped containers. The children
3-5 from the selected schools and 383 adults who resided were also encouraged to draw pictures on malaria accord-
in villages where the schools are located. School children ing to their understanding of the malaria education they
in the first and second grades were excluded due to per- received and voluntarily use the pictures to educate their
ceived difficulties in involving them in the intervention. peers and adults in the village. The teachers composed a
For community adults, an attempt was made to invite one song in the local language (Ga-Dangme) on malaria enti-
adult participant from each household of the villages tled 'pumi (mosquito) song', to educate the children and
where the schools are located. In recruitment of adult community on malaria transmission and prevention. The
participants, priority was given to those that were care- song, which became popular, was aimed at correcting
givers of children at the target school; in case there were common misperceptions that eating green mangoes or
no household members who were caregivers of target standing in the sun gives malaria and it promoted the use
school children, priority was given to caregivers of non- of bed nets for preventing mosquito bites and malaria
target children or any woman of reproductive age. transmission. The teachers made slogans such as "mos-
Intervention quito: malaria provider" which the children happily
Teacher training chanted at school and in the village. These activities were
The health education intervention started from October mainly conducted in the morning before the day's les-
2007. At the beginning of the intervention, a two-day sons.
training was conducted for all teachers at the interven- Campaign
tion school. On the first day, malaria-related information In February 2008, the teachers and 3
rd to 5th grade chil-
was shared, such as mosquito biology, malaria signs and dren of the intervention school also conducted a one-day
symptoms, and treatment and prevention. Additionally, anti-malaria campaign in which they educated the village
teaching methods using PLA were introduced. On the residents on malaria through a number of recreational
second day, strategies for effective implementation of activities. First, they marched through the village singing
malaria education activities in the schools and villages the 'pumi song' accompanied by the school band, at
were discussed. Finally, teachers developed action plans which time they invited community members to their
for the health education intervention (See Additional file school compound for a durbar. At the gathering, the chil-
1). dren educated the people about malaria through dramaand poetry recitals. A community health nurse demon-
strated the correct procedure for treatment of conven-
Table 1: Characteristics of school children.
Characteristic Intervention school Control schools p-valuea
(n = 105) (n = 81)
Age (yrs), median (range) 13 (8--20) 12 (8-21) 0.132b
Sex, male/female 59/46 44/37 0.799
Grade: n (%) 0.598
3rd 35 (33.3) 31 (38.3)
4th 37 (35.2) 23 (28.4)
5th 33 (31.4) 27 (33.3)
a, Chi-square test
b, Mann-Whitney U test
Ayi et al. Malaria Journal 2010, 9:98 Page 4 of 11
http://www.malariajournal.com/content/9/1/98Table 2: Characteristics of the community adults.
Characteristic Community Adults
Intervention village Control villages p-valuea
(n = 250) (n = 133)
Age(yrs), median (range) 35.0 (16-93) 35.5 (18-85) 0.504b
Sex, male/female 9/241 8/125 0.275
Occupation, n (%) 0.146
Farmer 170 (68.0) 103 (77.4)
Trader 43 (17.2) 17 (12.8)
Others 37 (14.8) 13 (9.8)
Education, n (%) 0.162
No formal education 86 (34.4) 56 (42.1)
Elementary (primary or 120 (48.0) 63 (47.4)
middle)
Senior (secondary or 16 (6.4) 3 (2.3)
higher)
Other 28 (11.2) 11 (8.2)
Ethnicity: n (%) 0.271
Ga-Dangme 235 (94.0) 121 (91.0)
Other 15 (6.0) 12 (9.0)
a, Chi-square test
b, Mann-Whitney U test
tional bed nets using insecticide tablets and explained the undertaken by the children. Native speakers of the Ga-
benefits of sleeping under treated bed nets and other Dangme language administered the questionnaire-based
malaria information. Volunteer residents had their con- interviews. Before the administration of questionnaire,
ventional bed nets freshly treated or re-treated at no cost. the interviewers were trained for a day by one of the
authors (JKA) who is a native speaker of Ga-Dangme.
Surveys Each of the participating children was interviewed at
Questionnaire-based interview school. The adult participants were interviewed at home.
Pre- and post-intervention questionnaire-based inter- Observational survey
views on malaria-related knowledge and practices were The research team made an observational survey on con-
conducted with all study participants (186 children and ditions in and around the households before the interven-
383 adults) from July to September, 2007, and April to tion in August, 2007. The observation with a simple
June, 2008, respectively. Questionnaires were originally check-list focused on possible mosquito breeding sites
developed on the basis of previous studies investigating around the house, presence or absence of mosquito-proof
perceptions, knowledge and behaviours pertaining to netting on windows/trap doors and bed nets, and the
malaria in Ghana and other countries [17-20]. The ques- condition of those mosquito prevention tools. As enter-
tionnaires were modified appropriately after pre-test was ing the house was perceived as an intrusion of privacy,
conducted with 22 mothers and 38 school children. the researchers asked an adult member of the households
There were two questionnaires: one for school children to bring one net, even though some of the households
and one for community adults. Each of the questionnaires had multiple nets.
shared 14 question items pertaining to knowledge and Parasitological survey
five pertaining to practices. The questionnaire for adults Finger-prick blood for preparation of thick and thin films
contained three additional items pertaining to practices. on microscope glass slides was taken from the school
For the post-intervention questionnaire interviews, three children that volunteered in the intervention and control
more question items were included to verify whether the schools. The blood films were Giemsa-stained after fixing
community adults were exposed to the malaria education the thin film in methanol; they were then examined
Ayi et al. Malaria Journal 2010, 9:98 Page 5 of 11
http://www.malariajournal.com/content/9/1/98Table 3: Comparison of changes in knowledge and practices among school children.
Knowledge/Practices Knowledge: % of respondents who knew
Practices: % of respondents who practiced very often or sometimes
Intervention (n = 105) Control (n = 81)
Before (%) After (%) p-valuea Before (%) After (%) p-valuea
Knowledge
Meaning of ITN 80.4 90.5 0.039 88.9 81.5 0.185
Effectiveness of ITN against mosquitoes 96.3 95.8 1.000b 98.6 91.0 0.054b
Necessity of ITN re-treatment 30.5 55.2 0.001 36.1 37.9 0.830
Vulnerability of ITN to sunshine exposure 46.8 58.7 0.121 38.9 42.4 0.673
Place to get re-treatment service for ITN 69.5 92.6 <0.001 66.7 72.7 0.440
Paracetamol alone cannot cure malaria 34.0 69.5 <0.001 46.9 55.0 0.305
Breeding site of mosquitoes 42.2 74.3 <0.001 32.1 53.1 0.007
Habitat of mosquito larvae 94.3 95.2 1.000b 95.1 92.6 0.746b
Resting place for mosquitoes 91.4 93.3 0.603 93.8 95.1 1.000b
Mango cannot cause malaria 10.5 79.8 <0.001 18.5 4.9 0.013b
Heat from the sun cannot cause malaria 11.4 75.0 <0.001 6.3 6.3 1.000b
Mosquito bites can cause malaria 76.2 99.0 <0.001 90.1 90.1 1.000
Drinking dirty water cannot cause malaria 16.3 43.8 <0.001 8.6 8.6 1.000
Sex difference in biting behaviour of 15.2 24.8 0.084 5.0 5.0 1.000b
mosquitoes
Practice
Weeding around house in the past 6 months 90.5 90.5 1.000 85.2 80.2 0.406
Talking with family members about malaria in 55.2 43.3 0.084 17.5 7.6 0.060
the past 6 months
Talking with neighbours about malaria in the 51.4 37.5 0.043 22.2 4.9 0.002b
past 6 months
Burning something to prevent mosquitoes in 92.4 80.4 0.012 88.9 78.8 0.080
the past 6 months
Covering arms and legs when going outside at 79.0 70.2 0.141 82.3 55.6 <0.001
night in the past 6 months
a, Chi-square test
b, Fisher's Exact test
microscopically to detect malaria parasites. This exercise Baseline differences in socio-demographic variables,
was performed pre-intervention in October 2007 and conditions of malaria prevention tools, and possible mos-
post-intervention in June 2008. quito breeding site were analysed by Chi-square test or
Mann-Whitney U test. Differences between pre- and
Data analysis post-intervention in knowledge, practices, and parasite
The sample size of 91 children in each group was calcu- prevalence were analysed by Chi-square test or Fisher's
lated to detect statistical significance in differences in Exact test. Statistical analysis was performed with SPSS
prevalence of malaria between pre- and post-interven- 17.0 (SPSS Inc., Chicago, IL). A P value of < 0.05 was
tion, with 80% power at 95% significance level. It was accepted as statistically significant.
hypothesized that the initial prevalence would be 40%,
and that a 20% decrease would be achieved in the inter- Ethical clearance
vention school. Approval and ethical clearance for the study were
obtained from the Scientific and Technical Committee
Ayi et al. Malaria Journal 2010, 9:98 Page 6 of 11
http://www.malariajournal.com/content/9/1/98Table 4: Comparison of changes in knowledge and practices among community adults.
Knowledge/Practices Knowledge: % of respondents who knew
Practices: % of respondents who practiced very often or sometimes
Intervention (n = 250) Control (n = 133)
Before (%) After (%) p-valuea Before (%) After (%) p-valuea
Knowledge
Meaning of ITN 86.3 74.6 0.006 93.1 89.1 0.266
Effectiveness of ITN against mosquitoes 84.7 93.2 0.006 86.9 84.7 0.635
Necessity of ITN re-treatment 51.4 52.1 0.891 59.5 33.6 <0.001
Vulnerability of ITN to sunshine exposure 54.9 75.1 <0.001 65.0 64.1 0.885
Place to get re-treatment service for ITN 62.5 72.4 0.032 63.6 65.0 0.832
Paracetamol alone cannot cure malaria 66.0 74.3 0.044 73.8 63.2 0.062
Breeding site of mosquitoes 57.4 66.7 0.034 65.4 51.1 0.018
Habitat of mosquito larvae 96.8 94.4 0.184 95.5 91.7 0.210
Resting place for mosquitoes 96.4 95.2 0.492 99.2 97.7 0.622b
Mango cannot cause malaria 15.6 61.1 <0.001 11.6 10.5 0.776
Heat from the sun cannot cause malaria 12.4 47.1 <0.001 4.6 2.3 0.334b
Mosquito bite can cause malaria 95.2 92.2 0.171 97.0 97.7 1.000b
Drinking dirty water cannot cause malaria 16.9 40.9 <0.001 3.0 2.3 1.000b
Sex difference in biting behaviour of 8.4 30.5 <0.001 0.8 3.0 0.370b
mosquitoes
Practices
Sleeping under bed net in the past 6 99.0 93.6 0.006b 92.9 94.9 0.525
months
Washing bed net in the past 6 months 95.7 93.6 0.391 91.3 95.3 0.280b
Treating bed net with insecticide in the 21.5 50.0 <0.001 25.3 30.5 0.406
past 6 months
Weeding around house in the past 6 94.8 85.0 <0.001 94.7 83.1 0.003
months
Talking with children about malaria in the 73.0 49.4 <0.001 40.6 34.8 0.334
past 6 months
Talking with neighbours about malaria in 43.4 33.1 <0.001 27.8 24.2 0.507
the past 6 months
Burning something to prevent mosquitoes 83.9 79.5 0.212 71.8 78.2 0.227
in the past 6 months
Covering arms and legs when going 79.0 80.3 0.721 65.9 50.8 0.013
outside at night in the past 6 months
a, Chi-square test
b, Fisher's Exact test
and the Institutional Review Board, respectively, of the was obtained from participating children before conduct-
Noguchi Memorial Institute for Medical Research with ing the survey.
the certified protocol number: CPN 038/06-07. Partici-
pants were informed that their participation was purely Results
voluntary and assured of the confidentiality of all data Study participants
collected. Informed written consent was obtained from From the total of 217 school children in the target grades,
adult participants, and parent/guardian of school chil- 186 participated both in the pre-and post-intervention
dren, with the children giving their assent. Verbal consent questionnaire-based interviews and were used for analy-
Ayi et al. Malaria Journal 2010, 9:98 Page 7 of 11
http://www.malariajournal.com/content/9/1/98Table 5: Results of household observational survey before the intervention.
Item/condition observed Community households
Intervention Control p-valuea
(n = 172) (n = 104)
Bed nets
Possession of at least one 149 (86.6%) 86 (82.7%) 0.558
net
Type of nets
Long-lasting 80 (53.7%) 49 (57.0%) 0.626
insecticide-treated 
nets
Conventional nets 69 (46.3%) 37 (43.0%)
State of the nets
Intact 114 (76.5%) 61 (70.9%) 0.618
Torn/hole 29 (19.5%) 21 (24.4%)
Not in use 6 (4.0%) 4 (4.7%)
Mosquito-proof nets on 
windows
Furnished 36 (20.9%) 46 (44.2%) <0.001
State of the nets
Intact 25 (69.4%) 19 (41.3%) 0.011
Torn/hole 11 (30.6%) 27 (58.7%)
Household compound
Presence of trash 83 (48.3%) 49 (47.1%) 0.722
container that can be 
breeding site of 
mosquitoes
Presence of covered 134 (77.9%) 70 (67.3%) 0.028
water storage facility
a, Chi-square test
sis. Participants consisted of 105 children (59 boys) out of Of a total of 447 households, 383 were involved in the
a total of 128 in the intervention school, and 81 children study and one participant from each household partici-
(44 boys) out of a total of 89 in the control schools (Table pated in the questionnaire-based interviews. Of the adult
1). The median age was 13.0 (range: 8-20) years in the participants, 115 were caregivers of target children. Adult
intervention school, and 12.0 (range: 8-21) in the control participants consisted of 250 adults (241 women) from
schools. No baseline difference was found between the the intervention village and 133 (125 women) from the
intervention and control school children in age, sex, or control villages (Table 2). Median age was 35.0 (range: 16
distribution ratio by grade. to 93) years in the intervention village, and 35.5 (range: 18
Table 6: Prevalence of P. falciparum infection among target school children before and after intervention.
School children P. falciparum prevalence (%) p-valuea
Pre-intervention Post-intervention
Intervention (n = 68) 30.9 10.3 0.003
Control (n = 63) 9.5 15.9 0.285
a, Chi-square test
Ayi et al. Malaria Journal 2010, 9:98 Page 8 of 11
http://www.malariajournal.com/content/9/1/98to 85) in the control villages. Most of the adults were < 0.001), although it was almost the same between pre-
engaged in farming (intervention village 68.0%, control (25.3%) and post-intervention surveys (30.5%) in the con-
villages 77.4%); over 90% were native Ga-Dangme speak- trol area. In the intervention area, the percentages signifi-
ers. Most of them had elementary or no formal educa- cantly decreased in "Sleeping under bed net" (p = 0.006),
tion. "Weeding around house" (p < 0.001), "Talking with chil-
dren about malaria (p < 0.001), and "Talking with neigh-
Questionnaire-based interview bours about malaria" (p < 0.001). In the control area, the
According to the post-intervention interview, 23.1% (24/ percentages also significantly decreased in "Weeding
104) of the children in the intervention school responded around house" (p = 0.003) and "Covering arms and legs
that they had, at least once, presented a picture to family when going outside at night" (p = 0.013).
members and/or other community members. For adults
in the intervention community, 37.1% (92/248) confirmed Observational survey
that a child showed a picture to them at least once; 80.7% One hundred and seventy-two out of 263 (65.4%) house-
(201/249) responded they had heard the "pumi song" holds in the intervention area and 104 out of 184 (56.4%)
sung by the school children; and 59.0% (147/249) households in the control areas were observed. Most of
responded that they attended the durbar at the interven- the households (intervention area: 86.6%, control area:
tion school. 82.7%) possessed at least one bed net (Table 5). Among
School children in the intervention school significantly the nets observed, nearly half (46.3%, 43.0%) were con-
improved their knowledge on ITN, cause of malaria, ventional nets. Most of the nets (76.5%, 70.9%) had no
mosquitoes, and paracetamol. Before the intervention, obvious holes or tears through which mosquitoes could
poor knowledge on cause of malaria according to invade. In nearly half of the households (48.3%, 47.1%), a
response to question item was striking. After the inter- dumped container, which could be a potential breeding
vention, those who correctly responded to the question site for mosquitoes, was found in the compound. Baseline
items "Mango cannot cause malaria" increased from differences were observed in the condition of mosquito
10.5% to 79.8% (p < 0.001), from 11.4% to 75.0% on "Heat proof nets and covers for water storage; fewer mosquito
from the sun cannot cause malaria" (p < 0.001), and from proof nets were furnished on the windows in intervention
16.3% to 43.8% on "Drinking dirty water cannot cause households (20.9%) than in control households (44.2%) (p
malaria" (p < 0.001) (Table 3). Among school children in < 0.001). In contrast, more mosquito proof nets were
the control schools, the difference between pre- and post- intact in the intervention households (69.4%) than in con-
intervention was only statistically significant in one item. trol households (41.3%) (p = 0.011). Water storages were
In the practice questions, no positive change was found in more likely to be covered in the intervention households
the intervention group of children. Rather, "Talking with (77.9%) than in control households (67.3%) (p = 0.028).
neighbours about malaria" and "Burning something to
prevent malaria" decreased with statistical significance (p Parasitological survey
= 0.043, p = 0.012). In control school children, "Talking Of the 105 children at the intervention school who partic-
with neighbours about malaria" and "Covering arms and ipated in the questionnaire survey, 64.8% (68/105) partic-
legs when going outside at night" also showed a statisti- ipated in both pre- and post-intervention surveys; at the
cally significant decrease (p = 0.002, p < 0.001). control schools, 77.8% (63/81) participated. In the inter-
As shown in children, adult participants' knowledge on vention school, the median age (range) and male:female
ITN, cause of malaria, mosquitoes, and paracetamol also ratio were 14 (9-20) and 40:28, respectively, in the partici-
significantly changed in the intervention area. In most of pating children; and 14 (8-18) and 19:18, respectively, in
the question items, the percentage of respondents who the non-participating children. In the control schools, the
had correct knowledge was slightly higher in adult partic- median age (range) and male:female ratio was 14 (10-19)
ipants than in children before the intervention. However, and 29:34, respectively, in the participating children; and
adult participants who believed that mango, heat from 15 (11-21) and 15:3, respectively, in the non-participating
the sun, or drinking dirty water can cause malaria children. The difference in the male:female ratio between
accounted for more than 80%. This knowledge was signif- participating and non-participating children was statisti-
icantly improved after the intervention (Table 4). In con- cally significant in the control schools (p = 0.007). Plas-
trast, adult participants in the control area did not show modium falciparum was the only malaria parasite found
an increase in any knowledge question items. In the prac- among the school children. Parasite prevalence signifi-
tice section, the percentage of those who treated a bed cantly reduced from 30.9% to 10.3% (p = 0.003) at post-
net with insecticide in the past six months significantly intervention in the intervention school, while it increased
increased from 21.5% to 50.0% in the intervention area (p from 9.5% to 15.9% in the control schools (Table 6).
Ayi et al. Malaria Journal 2010, 9:98 Page 9 of 11
http://www.malariajournal.com/content/9/1/98Discussion showed us were conventional nets. In response to the
After the intervention, the malaria prevalence was signifi- intervention, community adults who treated nets with
cantly decreased in school children in the intervention insecticide increased from 21.5% to 50.0%, compared
school, although no decrease was observed in children in with the 25.3% to 30.5% in the control area. This fact sug-
the control schools. Previous studies conducted in Ghana gests that the intervention was effective in increasing the
reported that falciparum prevalence was higher in the coverage of ITNs.
rainy season than in the dry season, and a little higher in Although the extra opportunity to treat their nets was
the middle of the rainy season than at the beginning or provided to community people during the one-day cam-
end [21,22]. In the present study, pre- and post-interven- paign, providing an opportunity alone is unlikely to
tion surveys were conducted at the end and middle of the increase the net treatment rate. According to the local
rainy season, respectively. As is consistent with the trend health authority, community people rarely participate in
reported from the previous studies, there was a slight free net treatment services which health workers offer
increase in prevalence in the control schools at the post- regularly in the study villages. Previous studies reported
intervention survey. Therefore, in the intervention school that barriers of insecticide-treatment were not only cost
the effect of seasonal variation is unlikely to be the princi- and access to treatment place, but also fear about insecti-
pal reason for the decrease. Although a marked difference cide, and poor linkage between malaria and ITNs [27-29].
in prevalence between intervention and control schools Thus, community awareness raised by the children about
was observed at the baseline, intervention impacts can be the malaria likely had a substantial impact on increasing
considered as one of the main factors influencing preva- net treatment practices.
lence reduction. The results also showed that the frequencies of talking
Previous studies conducted in sub-Saharan African with children and guardians/neighbours about malaria
countries including Ghana showed that people believed unexpectedly decreased at the post-intervention survey.
that not only mosquito bites, but also eating mango, This trend was seen both in intervention and control
drinking dirty water, and being exposed to hot sun were groups. There is a possibility that community adults were
causes of malaria [17,19,20,23,24]. The same finding was busier at post-intervention period than pre-intervention
observed in this study at pre-intervention. As Table 4 period because of seasonally related changes to farming
shows, 95.2% of the community adults correctly answered labour intensity. Unexpectedly, at the baseline, the fre-
that mosquito bites can cause malaria. However, only a quency of talking between children and guardians/neigh-
small proportion of the respondents disagreed with the bours in intervention groups was much higher than those
incorrect statements that "heat from the sun", "eating in control groups. This might be due to a higher burden
mango", and "drinking dirty water" cause malaria. After of malaria, suggested by the higher prevalence at pre-
the intervention, this knowledge was significantly intervention among school children in the intervention
improved in the intervention area. The improvement is school.
important because a lack of understanding of the linkage This study has five major limitations. First, although
between malaria and mosquito bites is associated with conventional ITNs and long-lasting insecticide-treated
poor adherence to vector control interventions [20,25]. nets (LLITNs) co-existed in the study site, no attempt was
In this study, both community adults and school chil- made to teach the respondents to recognize the two
dren showed significantly increased knowledge in the major differences to avoid confusion which might arise
item "Paracetamol alone cannot cure malaria". In Ghana, among study participants; conventional ITNs should be
most malaria cases have been managed at the household regularly treated and not be washed frequently. In con-
level [2]. However, during the fever episode, nearly 40% of trast, LLITNs have no need of treatment until nets are
children under 5 years old were not treated with any anti- washed many times and should be washed to activate
malaria medicine [3], and in the absence of anti-malari- insecticide agents. This limitation might be reflected in
als, paracetamol alone was commonly administered as the result showing that knowledge "Necessity of ITN re-
treatment [18]. Knowledge improvement regarding treat- treatment" was not improved in the intervention area.
ment could be beneficial even to children. In Kenya, Gei- Additionally, almost all of the community respondents
ssler et al [26] reported that a considerable number of reported that they washed their nets "very often" or
children self-treated their febrile illness without help "sometimes" in the past six months, although some of the
from their caregivers. nets must have been conventional ITNs. Second,
In Ghana, most children under five years of age have although we used the local word "Asra" to define malaria,
yet to be protected by ITNs [3]. It has been emphasised "Asra" does not necessarily correspond to "malaria" as
that untreated conventional nets should be treated with defined by modern medicine. As shown in the previous
insecticide to increase coverage of ITNs [3]. The observa- study in Ghana, "Asra" was used interchangeably to define
tion survey found that 45% of the nets that respondents both malaria and fever [19]. Third, in data analysis, clus-
Ayi et al. Malaria Journal 2010, 9:98 Page 10 of 11
http://www.malariajournal.com/content/9/1/98tering of individuals within the same school was not Competing interests
taken into account, resulting in failure in addressing the The authors declare that they have no competing interests.
cluster effects including class differences within the Authors' contributions
school. Fourth, results of the parasitological survey might IA and DN were the principal investigators and responsible for the whole pro-
be biased because of selection bias. Although no age and cess. JKA coordinated field implementation and contributed to the data collec-
tion. SH and JK contributed to field implementation, data analysis and
sex difference was observed between children who par- manuscript drafting. MJ, TM, and TT contributed to reviewing the manuscript.
ticipated in the survey and those who did not, sex differ- KMB and DAB were involved in protocol development and contributed to
ence was observed in the children in the control schools. study design development. All authors read and approved the final manu-
script.
Finally, a randomised controlled design was not
employed. However, no baseline difference was observed Acknowledgements
in demographic characteristics of the study participants The authors would like to sincerely acknowledge Prof. Alexander K. Nyarko, 
Director of the Noguchi Memorial Institute for Medical Research and the WACI-
and bed net related characteristics. PAC Project, the Ghanaian staff, and the JICA experts on the WACIPAC project 
For the intervention, schoolteachers successfully for their interest and contributions to the study. This project was funded by the 
adopted education activities using the PLA approach, Ministry of Health, Labour and Welfare of Japan (Kosei Kagaku Research Grant, 
such as role-playing, poetry recitals, slogan chanting, International Cooperation Research Grant 21S3).
song composition and dramatization. These activities Author Details
could be socially and culturally acceptable, because 1West African Centre for International Parasite Control, Parasitology 
teachers themselves designed these activities. The results Department, Noguchi Memorial Institute for Medical Research, University of 
Ghana, Legon, Accra, Ghana, 2Department of Community and Global Health, 
showed that most of the community adults were exposed Graduate School of Medicine, the University of Tokyo, 7-3-1 Hongo, Bunkyo, 
to these activities. Moreover, participants in the interven- Tokyo, Japan, 3Department of Epidemiology and International Health, 
tion area were more likely to talk about malaria than International Clinical Research Center, National Center for Global Health and 
Medicine, 1-21-1 Toyama, Shinjuku, Tokyo, Japan, 4Zoology Department, 
those in the control area. Thus, the application of this University of Ghana, Legon, Accra, Ghana, 5Bureau of International Medical 
strategy in other malaria endemic areas is recommended. Cooperation, National Center for Global Health and Medicine, 1-21-1 Toyama, 
Scaling up school-based health education interventions Shinjuku, Tokyo, Japan and 6Department of Tropical Medicine and Parasitology, 
School of Medicine, Keio University, 35 Shinano-machi, Shinjuku, Tokyo, Japan
should be easy if a well-established school health system
is available. The study in Thailand utilized a school health Received: 8 January 2010 Accepted: 18 April 2010 
system based on the Health Promoting School concept Published: 18 April 2010TM©h a2ils0a ia1rsir0 ta i AncJol yeOui ipersnte a navll a;A2 li0cilcac1ebe0nls,e s9 e fa:re9ro 8tBmiciol:e Mh dtetispdt: r/Ci/bewunwterwadl. muLtnaddl.a erria tjhoeu trenraml.cso omf t/hcoen Ctreenatt/i9v/e1 C/9o8mmons Attribution License (http://creativecommons.org/licenses/by/2.0), which permits unrestricted use, distribution, and reproduction in any medium, provided the original work is properly cited.
[14]. In recent years, WACIPAC has introduced the strat- References
egy for the setting up and establishment of school health 1. WHO, UNICEF: The World Malaria Report 2005 Geneva: World Health 
management systems as a national programme in 10 Organization; 2005. 
West African countries. Moreover, WACIPAC has rec- 2. Owusu-Agyei S, Awini E, Anto F, Mensah-Afful T, Adjuik M, Hodgson A, Afari E, Binka F: Assessing malaria control in the Kassena-Nankana 
ommended the necessity of coordination among donors district of northern Ghana through repeated surveys using the RBM 
interested in school health. Partnership has been pro- tools.  Malar J 2007, 6:103.
moted between governments in target countries and 3. WHO, UNICEF: The World Malaria Report 2008 Geneva: World Health Organization; 2008. 
donors for the establishment of school health manage- 4. Brooker S, Guyatt H, Omumbo J, Shretta R, Drake L, Ouma J: Situation 
ment systems. Thus, school-based intervention has the analysis of malaria in school-aged children in Kenya - what can be 
potential to be scaled up on the basis of the systems in done?  Parasitol Today 2000, 16:183-186.5. Afenyadu GY, Agyepong IA, Barnish G, Adjei S: Improving access to early 
each country in West Africa. treatment of malaria: a trial with primary school teachers as care 
providers.  Trop Med Int Health 2005, 10:1065-1072.
6. Kobayashi J, Jimba M, Okabayashi H, Singhasivanon P, Waikagul J: Beyond 
Conclusions deworming: the promotion of school-health-based interventions by 
School-based malaria education intervention engaging Japan.  Trends Parasitol 2007, 23:25-29.
school children as health messengers had a substantial 7. Bundy DA, Lwin S, Osika JS, McLaughlin J, Pannenborg CO: What should 
impact not only on school children, but also on commu- schools do about malaria?  Parasitol Today 2000, 16:181-182.8. Magnussen P, Ndawi B, Sheshe AK, Byskov J, Mbwana K: Malaria 
nity adults in improving knowledge on cause and preven- diagnosis and treatment administered by teachers in primary schools 
tion and bed net impregnation practices. The improved in Tanzania.  Trop Med Int Health 2001, 6:273-279.
knowledge and practices could be associated with the 9. Fernando D, de Silva D, Carter R, Mendis KN, Wickremasinghe R: A randomized, double-blind, placebo-controlled, clinical trial of the 
decrease in the malaria prevalence observed in the school impact of malaria prevention on the educational attainment of school 
children. children.  Am J Trop Med Hyg 2006, 74:386-393.
10. Pasha O, Del Rosso J, Mukaka M, Marsh D: The effect of providing 
fansidar (sulfadoxine-pyrimethamine) in schools on mortality in 
Additional material school-age children in Malawi.  Lancet 2003, 361:577-578.
11. Clarke SE, Jukes MC, Njagi JK, Khasakhala L, Cundill B, Otido J, Crudder C, 
Estambale BB, Brooker S: Effect of intermittent preventive treatment of 
Additional file 1 Appendix: Teacher training programme. The appen- malaria on health and education in schoolchildren: a cluster-
dix shows detailed contents of the teacher training and teaching aids used randomised, double-blind, placebo-controlled trial.  Lancet 2008, 
in the training. 372:127-138.
Ayi et al. Malaria Journal 2010, 9:98 Page 11 of 11
http://www.malariajournal.com/content/9/1/9812. Temperley M, Mueller DH, Njagi JK, Akhwale W, Clarke SE, Jukes MC, 
Estambale BB, Brooker S: Costs and cost-effectiveness of delivering 
intermittent preventive treatment through schools in western Kenya.  
Malar J 2008, 7:196.
13. Brooker S, Clarke S, Snow RW, Bundy DA: Malaria in African 
schoolchildren: options for control.  Trans R Soc Trop Med Hyg 2008, 
102:304-305.
14. Okabayashi H, Thongthien P, Singhasvanon P, Waikagul J, Looareesuwan 
S, Jimba M, Kano S, Kojima S, Takeuchi T, Kobayashi J, Tateno S: Keys to 
success for a school-based malaria control program in primary schools 
in Thailand.  Parasitol Int 2006, 55:121-126.
15. Onyango-Ouma W, Aagaard-Hansen J, Jensen BB: The potential of 
schoolchildren as health change agents in rural western Kenya.  Soc Sci 
Med 2005, 61:1711-1722.
16. Nonaka D, Kobayashi J, Jimba M, Vilaysouk B, Tsukamoto K, Kano S, 
Phommasack B, Singhasivanon P, Waikagul J, Tateno S, Takeuchi T: 
Malaria education from school to community in Oudomxay province, 
Lao PDR.  Parasitol Int 2008, 57:76-82.
17. Agyepong IA: Malaria: ethnomedical perceptions and practice in an 
Adangbe farming community and implications for control.  Soc Sci Med 
1992, 35:131-137.
18. Agyepong IA, Manderson L: The diagnosis and management of fever at 
household level in the Greater Accra Region, Ghana.  Acta Trop 1994, 
58:317-330.
19. Ahorlu CK, Dunyo SK, Afari EA, Koram KA, Nkrumah FK: Malaria-related 
beliefs and behaviour in southern Ghana: implications for treatment, 
prevention and control.  Trop Med Int Health 1997, 2:488-499.
20. Adongo PB, Kirkwood B, Kendall C: How local community knowledge 
about malaria affects insecticide-treated net use in northern Ghana.  
Trop Med Int Health 2005, 10:366-378.
21. Browne EN, Maude GH, Binka FN: The impact of insecticide-treated 
bednets on malaria and anaemia in pregnancy in Kassena-Nankana 
district, Ghana: a randomized controlled trial.  Trop Med Int Health 2001, 
6:667-676.
22. Kobbe R, Neuhoff R, Marks F, Adjei S, Langefeld I, von Reden C, Adjei O, 
Meyer CG, May J: Seasonal variation and high multiplicity of first 
Plasmodium falciparum infections in children from a holoendemic 
area in Ghana, West Africa.  Trop Med Int Health 2006, 11:613-619.
23. Okrah J, Traoré C, Palé A, Sommerfeld J, Müller O: Community factors 
associated with malaria prevention by mosquito nets: an exploratory 
study in rural Burkina Faso.  Trop Med Int Health 2002, 7:240-248.
24. Alaii JA, Borne HW van den, Kachur SP, Mwenesi H, Vulule JM, Hawley WA, 
Meltzer MI, Nahlen BL, Phillips-Howard PA: Perceptions of bed nets and 
malaria prevention before and after a randomized controlled trial of 
permethrin-treated bed nets in western Kenya.  Am J Trop Med Hyg 
2003, 68(Suppl 4):142-148.
25. Alaii JA, Borne HW van den, Kachur SP, Shelley K, Mwenesi H, Vulule JM, 
Hawley WA, Nahlen BL, Phillips-Howard PA: Community reactions to the 
introduction of permethrin-treated bed nets for malaria control during 
a randomized controlled trial in western Kenya.  Am J Trop Med Hyg 
2003, 68(Suppl 4):128-136.
26. Geissler PW, Nokes K, Prince RJ, Odhiambo RA, Aagaard-Hansen J, Ouma 
JH: Children and medicines: self-treatment of common illnesses 
among Luo schoolchildren in western Kenya.  Soc Sci Med 2000, 
50:1771-1783.
27. Winch PJ, Makemba AM, Makame VR, Mfaume MS, Lynch MC, Premji Z, 
Minjas JN, Shiff CJ: Social and cultural factors affecting rates of regular 
retreatment of mosquito nets with insecticide in Bagamoyo District, 
Tanzania.  Trop Med Int Health 1997, 2:760-770.
28. Schellenberg JA, Minja H, Mponda H, Kikumbih N, Mushi A, Nathan R, 
Abdulla S, Mukasa O, Marchant TJ, Tanner M, Lengeler C: Re-treatment of 
mosquito nets with insecticide.  Trans R Soc Trop Med Hyg 2002, 
96:368-369.
29. Rhee M, Sissoko M, Perry S, McFarland W, Parsonnet J, Doumbo O: Use of 
insecticide-treated nets (ITNs) following a malaria education 
intervention in Piron, Mali: a control trial with systematic allocation of 
households.  Malar J 2005, 4:35.
doi: 10.1186/1475-2875-9-98
Cite this article as: Ayi et al., School-based participatory health education 
for malaria control in Ghana: engaging children as health messengers 
Malaria Journal 2010, 9:98View publication stats