This was a cross-sectional study. For data collection, a validated anonymous structured self-administered questionnaire was used, which was written in English language and consisted of 49 items divided in to four major sections. Section I included 12 items about participants’ socio-demographic data such as age, gender, smoking behaviour, etc. Smoking behaviour was categorised into non-smoker, smoker or ex-smoker. A smoker is defined as a person who is currently smoking while a person who already stopped smoking for at least one year is considered as an ex-smoker. Section II included 12 multiple choice questions about participants’ knowledge about tobacco and smoking. The participants were asked to select one out of three options (“yes”, “no” or “I don’t know”). Section III included 14 items that focused mainly on participants’ attitude towards tobacco and smoking. In this section, the participants were requested to show their level of agreement with the statements using three-point Likert scale (“agree”, “disagree” or “unsure”). Section IV included eleven items with the same format as section II questions (except the 11^th item) and focused on participants’ smoking practice. The 11^th question was open-ended where the participants were asked to recommend strategies which could help to reduce the prevalence of cigarette smoking among pharmacy students. To validate and evaluate the questionnaire, a pilot study was conducted with the participation of 40 third year pharmacy students. The reliability of the questions was tested using the reliability test.

The study population was pharmacy students in the International Islamic University Malaysia (IIUM) from November 2012 to December 2012. The sample size was estimated based on total number of IIUM pharmacy students, which were 439 students. All registered students of year one, two, three and four in the Kulliyyah of Pharmacy were eligible for study participation. Only students who were absent during data collection or rejected to participate were excluded from the study.

The sample size calculated by Raosoft sample size calculator, to achieve a confidence level more than 95% and margin of error less than 5% when the response distribution is 50%, was 206. The participants were chosen by using simple random sampling technique.

An approval to conduct the study was obtained from the Department of Pharmacy Practice, Kulliyyah of Pharmacy, IIUM, since the study was exempted from approval by the Ethics Committee of IIUM. A total of 330 questionnaire forms were distributed to pharmacy students and collected by the end of the day. The additional questionnaires were included to make up for the drop out of participation. Prior to questionnaire distribution, the nature and the purposes of the study were explained to participants and informed consent was obtained. Of the 330 questionnaires distributed, 280 questionnaires were received; however, 29 of the received questionnaires were excluded because of incompleteness.

Statistical Package for Social Science (SPSS) 18.0 software package (SPSS Inc., Chicago, IL) was used to analyse the data. Descriptive statistics were used to summarise the data. Categorical data are presented as percentages and frequencies. Median split method was applied to categorise respondents as having either poor or good knowledge about smoking and either positive or negative attitude towards it. Inferential statistics were used to examine the relationships between variables. Mann-Whitney test and Kruskal-Wallis test were used to test the difference in knowledge, practice, attitude or total Knowledge, Attitude and Practice (KAP) scores based on respondent demographics when the number of groups are two and more than two, respectively. All statistical tests were two-tailed and maintained a significance level (α ≤0.05) and a confidence interval ≥95%. Answers to open ended questions were compiled, sorted out and post-coded.

The majority of respondents were in the age range of 21-23 years (183, 72.9%), female sex (191, 76.1%), Malay race (248, 98.8%), and single (246, 98%). This reflects the demography of student population in this faculty. Only one student (0.4%) of the respondents was a smoker. Forty nine (19.5%) of the respondents claimed that their brothers were smokers and 33 (13.1%) students claimed that their fathers were smokers. About half of the respondents did not have any family members who are smoking. More details are presented in [Table/Fig-1].

As shown in [Table/Fig-2], the median knowledge score was 16. The respondents with knowledge score of ≤16 were classified as having poor knowledge, while those with knowledge score of >16 were classified as having good knowledge [Table/Fig-3]. There was a significant difference in smoking knowledge level based on study year (p=0.001). Second year students showed the highest proportion of students with poor smoking knowledge (47, 73.4%), while fourth year students showed the highest proportion with good smoking knowledge (35, 56.2%) compared to students from other study years. More female respondents have good knowledge (88, 46.1%) compared to male respondents (20, 33.3%). However, the difference is not statistically significant (p=0.250).

Students have been categorised as having either positive or negative attitude based on the median of the attitude score (≤17 or >17, respectively) [Table/Fig-4]. There is no statistically significant difference between the study year and attitude towards smoking (p=0.644). First year students showed the most positive attitude (37, 61.7%) compared to students from other years. In contrast, more than half of second year students showed negative attitude (33, 51.6%) towards smoking. The evaluation of attitude towards smoking showed a statistically significant difference between genders (p=0.047), as females showed more positive attitude (113, 59.2%) compared to males (27, 45.0%).

Based on the median of smoking practice score, students were categorised as either having poor or good practice (≤11 or >11, respectively) [Table/Fig-2]. There was no significant difference in smoking practice based on level of study or gender (p=0.544, p=0.071, respectively) [Table/Fig-5]. Fourth year students showed the highest percentage with poor smoking practice (44, 68.8%), while third year students showed the highest percentage with good smoking practice (27, 42.9%), compared to other years of study. A higher percentage of students with good smoking practice was observed among males (28, 46.7%) compared to females (68, 35.6%).

Based on the cumulative score of KAP, students were categorised as either having poor or good KAP (≤45 or >45, respectively) [Table/Fig-2]. There was no difference in KAP score based on the level of study or gender (p=0.327 and p=0.120, respectively) [Table/Fig-6]. Third year students showed the highest percentage of students with good KAP (30, 47.6%). A higher percentage of students with good KAP was observed among male (29, 48.4%) compared to female students (75, 39.3%).

This study revealed that only 0.4% of IIUM pharmacy students are smokers. Other studies on smoking among medical students showed higher smoking prevalence (5.07% in Lao People’s Democratic Republic [20], 25.8% in Lebanon [21] and 17.6% in Saudi Arabia [22]). Similarly, smoking prevalence among young adults in Pakistan was reported to be 23% [2]. According to Azhar A et al., smoking prevalence among medical students is expected to be low because they have a greater chance of exposure to awareness programs and greater knowledge regarding the risk of smoking [19]. The reported low smoking prevalence among IIUM students is probably due to the policy strictly banning smoking in IIUM campus. With regards to gender, the smoking prevalence among males was higher than females, although overall prevalence among study sample were low compared to previous scholars studies, as majority of the sample were females. Similar trend was reported in previous studies [23-25]. This may be attributed to the social norms [24]. More than one third of participants reported that at least one of their siblings were smokers, while less than quarter of the participants has a close friend who is smoking. This is an important factor that should be considered because smoking status among family members and peers, more likely, will encourage them to become smokers in the future, which was reported by few previous studies [2,26,27].

Less than half of respondents appeared to have a good knowledge about smoking. They knew that nicotine is the main component of cigarettes, which reacts with body receptors and leads to addiction and development of diseases such as cancers and fetal developmental abnormalities [11]. Majority of the fourth year students had good knowledge compared to other students. This probably is due to the knowledge acquired through previous study years. Zhu t et al., reported that increasing the level of knowledge about smoking may provide awareness to those students about smoking hazards, where students that have medical education have a great impact towards reducing cigarette smoking compared to those who do not have any medical education [24].

On the other hand, more than half of the participants had positive attitude towards anti-smoking behaviour. Female participant’s had a higher percentage of students with positive attitude towards anti-smoking compared to male student. Several studies had documented no female smokers at all, while others revealed a significantly lower smoking prevalence among female students [28].

Study participants were noticed to have poor practice regarding establishing a new initiative towards smoking prevention, smoking free policy and smoking cessation programs. Therefore, to promote smoking cessation among pharmacy students in IIUM, it is suggested to conduct smoking cessation training programs and provide counselling sessions to the smokers. Awareness would be changed by health education methods using individual interview, group discussion and from anti-smoking campaign at the university. Furthermore, the presence of smoking ban policy is not enough to erradicate smoking. Hence, the level of students’ compliance with the smoking policy in the campus should be monitored to achieve complete tobacco control. Moreover, the price of cigarettes pack should be increased as the current price is still reasonable for students.

The findings of this study should be interpreted with caution as they are only related to the IIUM pharmacy students and cannot be generalised to other colleges in Malaysia. The data were collected by using self-administered survey, which may be limited by respondent bias. The smoking is considered as socially unacceptable action, as a result under-reporting could not be ruled out.

Smoking prevalence among IIUM pharmacy students was very low as majority was female participants, which might be considered a limitation of the current study. Also, the majority of students showed positive attitude towards smoking and good smoking practice. However, most of them had a poor smoking knowledge. This can be improved by targeted training and educational programs. Starting counselling sessions with smokers could also change their attitude towards smoking and help them to quit.

Funding: This study was funded in part by Research Initiative Grant Scheme [RIGS 17-005-0580] from International Islamic University Malaysia (IIUM).