Authors: Tahir Azhar1, Omar Hasan Kasule1, Syed Hassan Al-Mashor2, Aziz Baba3, Gerard Lim4, Fuad Ismail5, Hooi Lai Ngoh6, Zarihah Md Zain7. Presented at the First International Conference on Lung Cancer – Prevention and Treatment 21-23 November 2000 jointly organized by the Kuwait Society for Preventing Smoking and Cancer, the Kuwait Foundation for the Advancement of Sciences in cooperation with the International Union for Cancer Control (UICC) By Prof Dr. Omar Hasan Kasule, Sr. MB ChB (MUK), MPH, DrPH (Harvard); Deputy Dean for Research & Post-graduate Affairs; Professor of Community Health (Epidemiology), Kulliyah of Medicine, International Islamic University, Kuantan, MALAYSIA.

Authors: Tahir Azhar1, Omar Hasan Kasule1, Syed Hassan Al-Mashor2, Aziz Baba3, Gerard Lim4, Fuad Ismail5, Hooi Lai Ngoh6, Zarihah Md Zain7. Presented at the First International Conference on Lung Cancer – Prevention and Treatment 21-23 November 2000 jointly organized by the Kuwait Society for Preventing Smoking and Cancer, the Kuwait Foundation for the Advancement of Sciences in cooperation with the International Union for Cancer Control (UICC) By Prof Dr. Omar Hasan Kasule, Sr. MB ChB (MUK), MPH, DrPH (Harvard); Deputy Dean for Research & Post-graduate Affairs; Professor of Community Health (Epidemiology), Kulliyah of Medicine, International Islamic University, Kuantan, MALAYSIA.




This paper describes a intervention study carried out on a nation-wide sample of 1875 secondary school teachers and students who were given a baseline questionnaire on 102 KAP items covering knowledge of cancer information sources, cancer sites, cancer symptoms, cancer detection methods, cancer treatment methods, cancer risk factors, cancer prevention methods, and health seeking behaviors. The paper is based on the analysis of only 8 KAP indicators deemed most relevant to lung cancer. An intervention program was then carried out consisting of seminars, posters, brochures, and video shows. A follow-up questionnaire was administered 6 months following the intervention. Matched pair analysis using the MacNemar technique was used to identify KAP items that changed significantly after intervention. Conditional logistic regression suitable for matched data was used to identify interventions that independently predicted change of KAP.



A high proportion of the respondents, 86.7%, knew the lung as a cancer site; 96.4% knew that cigarette smoking was a risk factor for cancer; 72.8% knew that bloody cough was a symptom of cancer; 96.8% knew that stopping smoking was a cancer prevention measure; 79.6% knew that eating green vegetables was protective against cancer; 88.2%, responded that they would not like to sit next to someone smoking. There were significant changes in many KAP indicators following cancer education intervention. The paper presents the independent effects of the intervention modalities (seminar, poster, brochure, video) on change in KAP.



The study found that KAP on lung cancer was satisfactory compared to other less common cancers. There were improvements in KAP following intervention. It is therefore recommended that cancer education modules should be introduced in secondary schools before the teenagers learn risky behavior that is more difficult to change after.



The study had the primary objective of assessing the impact of currently-used health education methods on cancer KAP. A secondary objective of the study was to use the extensive data collected on cancer KAP, its determinants, and its change to design school-based cancer education modules that will involve teachers and students. Cancer education in the high school will have immediate impact as the students become responsible adults in a few years and will leave a life-long impact because at that age youths are very receptive to new ideas;  their opinions and attitudes are still amenable to change and modification. The study pioneered intervention at the school level as a mainstay of cancer intervention. School intervention had the dual impact of raising cancer awareness in general in addition to imparting specific information on cancer KAP.




The study was carried out in three states. Two districts, one rural and the other urban, were selected in each state. Two secondary schools were selected at random from each district for inclusion in the study. All selected schools agreed to participate in the study. Schools were allocated to the intervention (experimental) and non-intervention (control) groups by a quasi-random process.



The study consisted of a base-line questionnaire study of KAP form 3 students and their teachers. This was followed six months later by an intervention package that consisted of lectures, video shows, posters, and distribution of brochures on cancer. A post-intervention questionnaire survey of KAP was carried out 1-2 months after the end  of the intervention. Questionnaire administration was organized as class sessions where the students and their teachers took time to complete the questionnaires. Study research assistants were at hand to make clarifications where needed. State study coordinators sent all answered questionnaires to the study statistical center in Kuantan where the data was keyed and edited for consistency and completion.



Cancer education materials, brochures and posters, currently used were used. A special video on cancer education was prepared by the study. Seminars were held in the schools at which students and teachers were taught about various aspects relating to cancer KAP. They were given an opportunity to ask questions. Posters on cancer were displayed in the school. Brochures on cancer KAP were distributed. A video was shown.



One-way tabulations were carried out to determine the base-line prevalence of KAP indicator variables. Two-way tabulations of baseline KAP were carried out to study prevalence by gender and family income. Multivariate logistic regression models were used to study independent prediction of baseline KAP by gender and income level. Matched analysis using the Mac-Nemar statistic was carried out to identify KAP indicators that changed significantly at the repeat survey. Conditional logistic multivariate analysis appropriate for paired data was used to study how specific intervention modalities affected change of KAP while controlling for putative confounding factors.




The total study sample was 1875, 1695 (90.4%) being students and 180 (9.6%) being teachers. Nine hundred and eighty one, 981, (52.3%) were from urban districts and 894 (47.7%) were from rural districts. The majority of the study subjects, 1689 (90.1%) were aged 20 years and below. There was no marked gender difference; 863 (46.0%) were males and 1012 (54.0%) were females. Muslims constituted the largest religious group, 1030 (56.0%), followed by Buddhists 564 (30.6%), Christians 197 (10.7%), and Hindu 50 (2.7%).. The majority 1439 (81.4%) had lived in the study area for the past 5 years and only 329 (18.6%) had in-migrated.



The study sample had been divided into intervention and non intervention groups. The intervention group was 1221 (65.1%) and the non-intervention group was 654 (34.9%). No cancer educational activities were carried out for the non-intervention group. All respondents were asked to state what study interventions they had been exposed to. The following are proportions of study participants who reported receiving specific interventions: attendance at only 1 seminar 461/930 (49.5%), attendance at 2 or more seminars 175/926 (18.9%), seeing a poster 691/935 (73.8%), receiving a brochure 447/927 (48.2%), and watching a video 324/929 (34.9%).



Previous exposure to risk factors can affect KAP; an investigation was therefore made of prior exposure to tobacco products. Of 1812 respondents, 1547 (85.4%) never smoked, 174 (9.6%) were current smokers, and 91 (5.0%) had stopped smoking more than 1 year ago. Among 213 current and former smokers, 172 reported smoking 1-10 cigarettes a day, 30 reported smoking 11-20 cigarettes a day, 6 reported smoking 6 cigarettes a day, and only 5 reported smoking above 40 cigarettes a day. Eighty percent of the smokers started smoking below the age of 20 years. Reported tobacco chewing was much lower than cigarette smoking: 1704/1715 (99.4%) of the respondents had never chewed tobacco, 5/1715 (0.3%) were current chewers, and 6/1715 (0.3%) had stopped chewing more than a year ago. The respective statistics for batel nut chewing were 1664/1699 (97.9%), 13/1699 (0.8%), and 22/1699 (1.3%).



Table 1 shows KAP indicators at the baseline survey. The following proportions knew the various indicators: the lung is a cancer site, 86.7%; cigarette smoking is a risk factor for cancer; stopping smoking prevents cancer, 96.8%; and bloody cough is a cancer symptom 72.8%. The proportion who knew that low vegetable intake was a risk factor for cancer was low, 45.6%, but a higher proportion knew that including green vegetables in diet was protective against cancer, 79.6%. Only 7.7% thought that smoking was the only cause of cancer. A high proportion, 88.2%, did not like sitting next to someone smoking.



Table 2 shows KAP indicators by gender. Significant gender differences in KAP were found only for knowledge of the lung as a cancer site, knowledge that a bloody cough was a symptom of cancer, knowledge that inclusion of green vegetables in the diet was a preventive measure, and the attitude of not wanting to sit next to a smoker. With the exception of knowledge of the lung site, females had a higher proportion of correct knowledge and attitudes.



Table 3 shows KAP indicators by income level. The proportion of correct knowledge, opinions, and attitudes increased significantly with income level for all except three KAP indicators related to tobacco: knowledge of the lung cancer site, knowledge of smoking as a risk factor, and knowledge that stopping smoking was a preventive measure. The lack of significant increase with income in these three indicators is that they are widely known by the whole population irrespective of income level due to the intense anti-smoking advertising campaigns.



Table 4 shows that there was an increase at the repeat survey in the proportion knowing selected KAP indicators but this reached statistical significance only for 4 indicators: lung cancer site, bloody cough as a cancer symptom, low intake of green vegetables as a risk factor, and inclusion of green vegetables in diet as a preventive factor. Knowledge of cigarette smoking as a risk factor did not show significant change at the repeat survey because it was already too high at the base-line survey, 96.1%. It is worth noting that there was no significant change in either opinion or attitude.



Tables 5-9 show the results of multivariate analysis showing the independent effects of each intervention on change in the given KAP indicators. Interventions did not have consistent independent effects on the KAP indicators. Each intervention impacted KAP indicators in a different way. Being in an intervention school had an effect on KAP No. 1, 3, and 8 that was independent of any specific intervention. Attending a seminar had an effect on KAP No. 1 and 2. Seeing a poster had an independent effect on KAP indicators No. 2 and 3. Reading a brochure had an independent effect on KAP indicator No. 7. Watching a video had an independent effect on KAP indicator No. 6.



The study sample: The study sample is a national sample of the school population including teachers and students. The findings can therefore be used as the basis for a nation-wide school-based cancer education program. Intervention at the secondary school level when students are in a formative stage susceptible to influence is more likely to produce life-long changes in healthy life-style and behavior than intervention at an adult stage.


Prior exposure to tobacco, a cancer risk factor: Prior exposure is likely to increase awareness of cancer and thus affect KAP. The low rates of exposure to cigarette smoking and alcohol are expected in a school population; although the data is thought to be an under=reporting of the actual situation.


Study intervention modalities: There were wide variations in the proportions of respondents exposed to each intervention modality. There was also significant variations among districts. This wide variability provided an opportunity to study the separate effects of each intervention modality on specific KAP indicator variables. Seminar was the most common intervention reported. This is because of the seminars that were held in the schools. A high proportion of respondents reported receiving cancer information from sources outside the study. These are not likely to have been new or special sources. They are the usual sources that were described in the baseline survey. These sources were included in the multivariate model that determined the interventions that independently predict change of KAP.


KAP indicators: A high proportion of respondents knew the lung cancer site. The results indicate that cancer education messages should be individualized because knowledge of the different sites was predicted by gender; the proportion of knowledge among males was higher. There was however no association between knowledge of the lung cancer site and income level, a reflection of wide-spread knowledge that cuts across socio-economic classes. Knowledge of bloody cough as a cancer symptom was only 72.8%. This data highlights the importance of more education on cancer symptoms and signs to enable earlier diagnosis and treatment. Knowledge of smoking as risk factor was almost universal but knowledge of low vegetable intake as a risk factor was lower. Knowledge of cancer prevention methods paralleled knowledge of cancer risk factors; the most known preventive method being stopping smoking.


Intervention: The data shows significant changes in KAP indicators in all districts and all socio-demographic groups. The changes occurred in both the schools that had intervention programs and those that did not; being more marked in the former. It seems that the mere action of completing the questionnaire at baseline was educative by raising interest and curiosity about cancer. The study demonstrated more changes in indicators of knowledge than indicators of opinions and attitudes. A significant finding of this study is that specific intervention modalities (seminar, poster, brochure, video) had different impacts on different KAP indicators. Thus packaging of cancer education methods requires consideration of both the type of information and the medium to be used.



The study has demonstrated improved KAP due to increased awareness without any specific information input. Additional change was shown to be due to specific interventions. It has demonstrated the feasibility and positive impact of a KAP intervention program in schools. The detailed data-base collected can be used to prepare cancer education modules for use in schools. The current cancer education material, methods, and media need extensive change in light of the findings of this study. Some content is already known by the public and need not be the area of emphasis. The study has for example shown that knowledge of cigarette smoking as a risk factor for cancer was high; the target audience will be bored if this fact is given emphasis in an education program. The close connection between KAP indicators at baseline and at repeat survey with specific socio-demographic variables is an important finding in a multi-cultural country. Programs should be tailored to specific niche populations based on gender, ethnicity, income, and district of residence; a common centrally planned program is not likely to be effective. Extensive tabulations of the data were made to provide a data-base that can be used in planning cancer education programs as well as preparing cancer education modules in schools as a strategy to achieve long-term change in KAP of future adults.

ęCopyright Professor Omar Hasan Kasule, Sr November 2000