Effect of Health Education on Knowledge Attitude Practice towards Malaria among Basic Schools Pupils in Taiz

Background: Malaria is one of the main health problems in Yemen. Health education is essential for the control of diseases such as malaria. School-age children represent 25% of Yemen’s population. Schools children can convey the knowledge and skills that they acquire at school to the community, thus increasing general community awareness about malaria. Aim to determine the impact of school-based malaria education intervention on knowledge, attitude and practice of school children towards malaria prevention and control. Methods: We conducted a community-based trial, intervention and non-intervention comparison (exposed & non-exposed), in four randomly selected districts (rural and urban) of Taiz governorate. This study was conducted in four districts of Taiz governorate. The study population was 2130 pupils of grade 6, 7, and 8 in primary schools, chosen from four randomly selected districts; two of them were rural and the others urban. Four schools and their pupils (1065 pupils) were chosen for the study and assigned as an intervention group and four schools with their pupils (1065) were chosen for the study and assigned as a non-intervention (1065 pupils). Data were col-lected using questioner in intervention and non-intervention schools (three months apart). The major intervention activities included lectures about malaria, distribution of educational materials. χ 2 was used to analyze differences. Results: with (35%) in the non-intervention group the difference was statistically significant. The knowledge of mode of malaria transmission was higher in the intervention schools to be (86.2%) compared with (59.1%) in the non-intervention group, with statistically significant deference (P value <0.001). The knowledge of fever as a main malaria symptom was higher to be (90.4%) in the intervention schools compared with (63.6%) in the non-intervention schools with statistically significant deference. Conclusions and Recommendation: This study concludes that the health education intervention in primary schools in Taize governorate had a positive impact on the knowledge, attitude and practice of pupils. We recommend to conduct similar methods of the health education activities in schools with suitable modifications to reach all schools level.


Introduction
Malaria is a preventable and treatable infectious disease that kills more than one million people each year [1]. It is a protozoal infection transmitted to human beings by mosquitoes biting mainly between sunset and sunrise. Human malaria is caused by four species of Plasmodium protozoa: Plasmodium falciparum, P. vivax, P. ovale and P. malariae [2]. According to the latest estimates, released in December 2014, there were about 198 million cases of malaria in 2013 (with an uncertainty range of 124 million to 283 million) and an estimated 584,000 deaths (with an uncertainty range of 367,000 to 755,000). Malaria mortality rates have fallen by (47%) globally since 2000, and by (54%) in the WHO (World Health Organization) African Region [3]. It is a deadly mosquito-born disease, which took 781,000 lives in 2009 and affects as many as half a billion people in 106 countries in Africa, Asia and Latin America [4]. Malaria is endemic in 109 countries and territories in tropical and sub-tropical zones, spanning all continents of the world except Antarctica and Australia, with intensities of transmission that vary from very low to extremely high [5]. Every 45 seconds, a child dies from malaria. Each of these deaths is avoidable [6]. By the mid-19th century, malaria had been eliminated from several countries in temperate zones in which it had been endemic [7]. The recent impact of malaria control interventions, showing in multiple countries over a few years over (75%) reduction in malaria cases with high coverage of effective treatment and vector control, has renewed global interest in malaria elimination and eradication [8]. P. falciparum is the dominant species of parasite in Djibouti, Saudi Arabia, Sudan and Yemen, but the majority of cases in Afghanistan and Pakistan and almost all cases in the Islamic Republic of Iran and Iraq are due to P. vivax [9]. The renewed effort to control malaria world-B. A. Farea et al. wide and move towards elimination in some countries is founded on the latest generation of effective tools and methods for prevention and treatment, increasing use of long-lasting Insecticide Nets (LLINs), Artemisinin-based Combination Therapies (ACTs) and Indoor Residual Spraying (IRS) for scaling up global malaria control and elimination [10] [11]. Malaria is considered as one of the main health problems in the Republic of Yemen [12]. Regarding malaria epidemiologically Yemen is classified as an Afro-tropical, because of the predominance of the life-threatening species, and the plasmodium falciparum [13] [14]. Malaria in Yemen is unstable, seasonal malaria that is closely related to altitude, rainfall and topography [15]. In some settings, it is estimated that up to (40%) of severe pediatric admissions can be due to malaria during the malaria transmission season, with appreciable mortality [16]. Pregnant women and children under five are the mostly affected groups whereas all age groups are at great risk in the epidemic prone areas where epidemics may occur following unusual climatic changes, e.g. heavy rainfalls, unexpected warming and rise in humidity [17]. WHO estimated the malaria cases in Yemen in 2009 to be around 265,074 cases and 779 estimated related deaths [18]. In an effort to reduce the burden of malaria in the country, the national malaria control program has implemented an integrated malaria control strategy of case management, integrated vector control including long-lasting insecticide nets, indoor residual spraying (IRS) and larviciding and community health education (increasing the capability of the community to recognize, prevent and control malaria is the nine strategic directions of the NMCP (National Malaria Control Program) [19]. Skills-based health promotion and disease prevention should be integrated into school health curricula as schools have the responsibility to equip children of all ages with the skills and information that they need in order to make informed choices to lead healthy and productive lives [20].
In general, skills-based health education aims to help children develop attitudes, knowledge and skills necessary to allow them to maintain and enhance their own health [20].
In Yemen malaria cause a major cause of absenteeism [21]. School-age children represent 25% of Yemen's population, and an increased proportion of these children are going to school, who could benefit from a systematic approach to school-based malaria control [22]. The aim of the study is to determine the impact of school-based malaria education intervention on knowledge, attitude and practice of school children towards malaria prevention and control.  Based on that, calculated sample size was found to be 1065 pupils. A larger sample size is needed, because of the design effect. Since we will use (Urban-Rural), the design effect (DEFF) was estimated as 2. So, the actual sample size was double of the calculated sample size. N = 1065 × 2 = 2130 pupils.
Half of the sample size (1065) is as an intervention group, and another half (1065) as a non-intervention group.
Multistage sampling technique was used.
1) A list of all districts in Taiz governorate (sampling frame) was made, including rural and urban district. Random selection of two districts one rural and other urban for the study as a study group, according to that Alqahera (urban district) & Ataizea (rural district) have been selected, and two districts one rural and other urban as a control group, according to that Salah (urban district) & Alshameaten (rural district) have been selected.
2) A list of schools (sampling frame) was made, including schools in Alqahera district (urban) and Ataizea (rural district), random selection of one school for boys & one school for girls from each district (4 schools: 2 from urban (one for boys & one school for girls) and 2 from rural (one for boys &one school for girls) as a study schools).
3) A list of schools (sampling frame) was made, including schools in Salah Specially designed Arabic questionnaire was prepared by principle investigator; the questions related to bed net was taken from previous study, that conducted in Yemen about BN [24]. The questionnaire was validated by experts from Sana'a University and from the Ministry of Public Health & Population.
The questionnaire includes demographic, knowledge, attitudes and practice da-ta, the same questionnaire was used in pre and post intervention by the same training team at the same time for the intervention group (schools receive health education) and the control group (schools not receive health education activities), the control group (1065). One pilot study was conducted before implementation of the study in two schools urban and rural. At the end of each day of data collection, the completed questionnaires were checked for errors, and completeness, the data were coded to be suited for computer feeding.
Data collectors were trained to ensure the standardization of data collection. Researcher supervised all steps of data collection that was performed by trained data collectors. Educated teachers from the same localities were trained by researcher to perform the data collection.
The same questionnaire was used in intervention and non-intervention by the same trained data collection team.
The completed questionnaires were checked for completeness; the data were coded and entered to the computer. Quality control of the data was ensured by rechecking 25% of entered data.
The statistical analysis was performed using the Statistical Package for Social sciences (SPSS) program V.17.0, Chi-square test was used to analyze categorical variables, P value < 0.05 was considered as statistically significant difference.
T-test was used to analyze variables that have mean scores. The 23 items of knowledge (one item was a mode of malaria transmission, 8 items were symptoms of malaria, 7 were complication and 7 items were for malaria prevention) were aggregated to create a single knowledge outcome for each student with possible scores ranging from 0 (23 were incorrect) to 23 (23 were correct). The  Very good knowledge = correct answer for (75%) and above of questions.
Poor knowledge = correct answer for questions (less than 50%).
We used 5 points Likert scales to measure the attitude (strongly agree, agree, uncertain, disagree and strongly disagree).
Informed consent was obtained from parents/caretakers along with the children's assets. Institutional approval was obtained from the Ministry of Education Office in Taiz Governorate, district education offices, and headmasters of targeted schools before the start of data collection.

Result
In our study 2130, pupils were included; half of them 1065 from urban (Al-Qaherah, Salah districts) and another half from rural (Attaezzeiah, Alshameatan districts).
Females constituted 50%, and males constituted 50%. The mean age of participants was 12.9 ± 1.4 years, with slight difference according to sex; it was 12.8 ± 1.4 for females and 13.1 ± 1.4 for males. After the intervention (health education activities), the number of pupils who recognized mosquito bite as a route of malaria transmission in the intervention schools was higher 918 (86.2%), compared with 600 (56.3%) in the non-intervention group with statistically significant difference.
From Table 1, it is clear that the statistically significant difference was observed in all symptoms, the highest difference occurred in mentioning the cardinal symptoms of malaria like fever, headache, loss of appetite, and vomiting.
After introducing the malaria education campaign, the mean knowledge of malaria complications increased to 3.9 ± 1.8, compared with 1.6 ± 1.1 in the non-intervention group. A statistically significant difference was detected between the intervention and the non-intervention groups. Details are presented in Table 2. After introducing the malaria education campaign, the mean score of knowledge of malaria protection increased to be 4.8 ± 1.8, compared with 2.8 ± 1.8 in the non-intervention group. The difference was statistically significant between the intervention and the non-intervention. Details are illustrated in Table 3.
After  Percent cannot be accumulated due to the possibility of mentioning more than one signs from the same pupil. in the intervention group, but without statistically difference. The schools as a source of information was the main source 533 (50%) in the intervention group, while it was only 9 out of 1065 pupils (0.8%) in the non-intervention group.
To assess the attitude of pupils, researchers put some statements and compared numbers and percent of those pupils who mentioned strongly agree of some attitudes towards malaria intervention and non-intervention groups. Almost all items were higher in intervention group compared with non-intervention group. Details are illustrated in Table 4.
Level of knowledge: The mean of the knowledge was higher in the intervention group 60.2% ± 17%, compared with 31.7% ± 13.5% in the non-intervention group (P value < 0.001).
The number of pupils who scored seventy-five percent 75% or more (high level that answered correctly more than 16 of 23 questions) after the activities of health education was 213 (20%) in the intervention group, compared with 2 (0.2%) in the non-intervention group. The difference between score of pupil's knowledges in the intervention and the non-intervention schools is illustrated in Figure 2.
Level of attitude: The mean level of attitude increased in the intervention group to 48% ± 14% compared with 34.7% ± 13.5% in the non-intervention group.
The number of pupils who scored seventy five percent 75% or more (high level who answered correctly 6 or more question of 8 question) after the activities of health education was increased to 301 (28.3%) in the intervention group, compared with 124 (11.5%) in the non-intervention schools. The difference between score of pupil's attitudes in the intervention and the non-intervention schools is illustrated in Figure 3.
The relationship between an age group of the intervention group and the grade of knowledge and the grade of attitude illustrates that, the grade of knowledge is associated with increase in the age, the difference is statistically significant, P value was <0.001, while the grade of attitude was increased with increase   in the age, but the difference is not statistically significant, P value was 0.401.
The relationship between sex of study population and the grade of knowledge and the grade of attitude illustrates that, the grade of knowledge is higher in the male, the difference is statistically significant, P value is <0.001. While the grade of attitude has some difference but the difference is not statistically significant (P value was 0.430). The relationship between the schooling level of the study population and the grade of knowledge and the grade of attitude illustrates that, the very good grade of knowledge was increased by increasing the schooling level of pupils, the difference was statistically significant (P value is <0.001).
The very good grade of attitude was higher among pupils of seventh class the difference was statistically significant (P value is <0.001).
Analysis of the relationship between a residence of the pupils and grade of knowledge and grade of attitude illustrates that, although the grade of knowledge was difference, but there was no statistically significant, P value is 0.581.

Discussion
The knowledge of mode of malaria transmission is very important because a lack

B. A. Farea et al. Health
of understanding of the linkage between malaria and mosquito bites is associated with poor adherence to vector control interventions [26].
The knowledge of correct mode of transmission was higher in among intervention group, 918 (86.2%), compared with 600 (56.3%) in the non-intervention group with statistically significant difference P value < 0.001. This agrees with studies conducted in Thailand the improvement was 62.1% [26], and in Ghana where the positive change was 99% in the intervention group [27], while in rural areas in China the improvement was 76.31% [28]. Good knowledge of the symptoms of malaria is crucial to recognizing the disease and to seeking appropriate health care [26]. The initial knowledge of symptoms in our study population was less than the findings of a study conducted in southeastern Iran which obtained that (80% -90%) of the respondents were aware about signs and symptoms of malaria [29]. In our study the knowledge of fever as a cardinal symptom of malaria increased to 952 (89.4%) in intervention group compared with 585 (54.9) in non-intervention group, the difference was statistically significant (P value < 0.001), this is in agreement with the study conducted in Thailand, where the knowledge of fever as a symptom of malaria was increased with statistical significant difference to be (56%) [26] in intervention group, it was denoted a substantial increase of the awareness of malaria, (67.6%) of pupils knew how to prevent malaria, while a study conducted in south-western Tanzania obtained that more than (85%) of the respondents were knowledgeable of malaria preventive measures [30]. Our findings revealed that mosquito-proof nets on windows increased to (84.5%) in the intervention group, compared with (41.5%) in the non-intervention group, and water storages were increased to (86.8%) compared with (55.2%) in the non-intervention group. This is in agreement with the study conducted in Ghana where the mosquito-proof nets on windows was higher (69.4%) in the intervention group compared with (41.3%) in the non-intervention group. And water storages were more likely to be covered in the intervention group (77.9%) than in the non-intervention group (67.3%), P value = 0.028 [31]. Personal protection is widely used and accepted to prevent mosquito bites [31].
In our study the positive change toward malaria prevention in the intervention group obtained that the percent of pupils who own BN increased with statistical significant difference to (16.8%), compared with (12.1%) in the non-intervention group, Regarding to using of BN the sleeping under BN was higher (72.1%) in the intervention group compared to (62%) in the non-intervention group with statistical significant difference. This is in agreement with the study conducted in Uganda [32]. Where the sleeping under bed net increased to (99%) in the intervention group compared to (93.6%) in the non-intervention group, and it showed a significantly higher proportion of respondents in the intervention group reported use of effective malaria prevention methods (BN, insecticide spray and coils) in their house hold (73.8%) compared with respondents in the non-intervention group (59.6%) (P = 0.037) [32]. Health Factors such as vulnerability, economic constraints, inadequacy or unavailability of appropriate health services, and other related factors play an important role in explaining health seeking behavior of the people [28]. In our study seeking medical care increased to be (87%) in the intervention group. Our findings in this regard are in agreement with the study conducted in rural areas of China where the positive change toward correct treatment was significantly increased from (10.7%) to be (39.4%), and the seeking medical care increased from (78 %) to be (91%) [28]. After the intervention, the mean knowledge increased to (4.4) ± 1.9 (P < 0.001), it was less than studies conducted in Kenya where the study pupils had a significant increase in the mean knowledge, the difference was (6 ± 4.0) (P value < 0.0001) [33], and in India where an average overall pretest score was 11.8 ± 5.03 which increased significantly to 19.3 ± 4.69 after intervention [34].
The schoolchildren who answered correctly more than 12 of the 23 knowledge questions increased significantly in the intervention schools to be (64.5%). This also agreed with the study conducted in Oudomxay province, Lao PDR that revealed the scores of the mean knowledge, attitudes and practices were significantly increased one month after the intervention [35]. As a result of HE, the schoolchildren changed their behavior positively towards malaria prevention.
The schoolchildren in our study who answered correctly four or more out of eight attitude questions were increased to (58.2%). This was similar to the studies conducted in Thailand where schoolchildren changed their behavior positively towards malaria prevention with significant difference in 6 of 7 questions [26], and in India when schoolchildren exposed to different health education activities a high average of correct score (47.9%) in the intervention group comparing to that of the control group (26.6%) was noticed [34]. The relationship between the schooling level of the study population and the grade of knowledge and the grade of attitude illustrates that, the very good grade of knowledge was increased by increasing the schooling level of pupils, the difference was statistically significant (P value is <0.001), This was similar to the studies conducted in rural areas of China where was the grade of knowledge was increased by increasing the schooling level of pupils (P value is <0.001) [28]. Our study revealed that our school-based malaria HE activities had a positive impact in changing schoolchildren's behavior with regards to malaria prevention in Yemen. Such positive results were seen as we encouraged the teachers to use various participatory learning methods, including child-to-child approaches. The importance of these types of approaches, has been recognized. Some studies reported effectiveness of health education using child-to-child approaches for malaria control, it was reported that participatory learning methods were effective in changing behaviors [26]. Nonetheless, our positive findings with respect to knowledge and attitudes are encouraging in light of the tenets of health behavior theory which indicate that the knowledge and attitude change are important precursors to behavioral change [36].

Limitation of Study
1) There was a short time between exposure and follow-up that required us to focus on proximal learning outcomes and not on behavioral change.
2) Activities that are not based on the curricula increased teachers work load.
3) There is a need to study the effect of HE activities on the prevalence of malaria cases among schools' children by comparing cases of malaria among schools' children before and after HE activities.

Conclusions and Recommendation
In The knowledge of mode of malaria transmission was increased in the intervention schools to be (86.2%) compared with (59.1%) in the control group with statistical significant deference (P value < 0.001).
The knowledge of fever as a malaria symptom was increased with statistical significant deference to be (90.4%) in the intervention schools compared to (63.6%), in the control schools with statistical significant deference (P value < 0.001).
The positive change in knowledge of mode of transmission was not significantly associated with age, sex, schooling level, and education status of father or mother of pupils.

Recommendation
Since our study shows a positive impact, conduct health education campaigns to improve health education regarding malaria infection in schools is highly recommended. Since activities that are not based on the curricula could increase teachers work load. It may be recommended to integrate the activities of health education in to the curricula.
Our school-based malaria control study was effective in schools' children behavior, and could be effective and recommended in the wider community groups to comprehend malaria control activities.
Although short-term malaria health education campaigns may have a positive impact on the control of malaria, a multi-channel approach is required.
Further studies on the impact of health education activities on knowledge, attitude and practice toward malaria control among schools' pupils, and the impact of these activities on the prevention and control of malaria in the community.

A Questionnaire on the Impact of Health Education of Malaria Disease on the Knowledge and Behavior of School Students in Taiz Governorate
Dear student, put a mark (√) or (×) on the phrase you deem appropriate.  Third: Attitudes, beliefs and behavior: 1. Who are the most vulnerable groups to severe malaria (more than one answer can be chosen)?