
AJTCCM VOL. 29 NO. 1 2023 7
RESEARCH
healthrisk for both active and passive smokers.[1] There is an
erroneous perception that shisha is less hazardous than tobacco
cigarettes,[4-6] although mounting evidence indicates that it is
even more harmful.[4-6] Like tobacco cigarettes, shisha has been
shown to be associatedwith a wide range of detrimental eects on
health such as cancer, heart disease, lung disease, and many other
illnesses.[4-6] Moreover, sharing of shisha pipes has been linked
to the spread of infectious diseases such as hepatitis B, herpes,
tuberculosisandinuenza.[4-6]
Shisha has recently been gaining increased popularity in many
developed and developing countries.[4,7] It has been documented
that shisha smoking is common among young people, mainly high-
school children and college and university students.[7] For example,
according to a survey involving >100 000 students from 152 colleges
and universities in the USA, current shisha smoking was reported
by 8.4% of students, second only to cigarettes.[7] Two recent studies
in 15 secondary schools in London, UK, and on the campus of the
University of Florida in the USA have documented for the rst time
a higher prevalence of use of shisha than of cigarettes.[7] In Africa,
shisha smoking has rapidly become increasingly popular in major
cities. e vogue is pervading society, and smoking is commonly
practised by university undergraduates, adolescents and the older
population in restaurants and hotels and at social gatherings.[8-10] Lack
and/or weakness of regulations have also contributed to its increased
consumption. A high prevalence (36.4%) of shisha smoking and poor
knowledge regarding its eects on health was reported in youths
attending bars in Kampala, Uganda.[9] In Nigeria, the prevalence of
shisha smoking in nightclubs was 7.1%.[10]
Burkina Faso has not escaped this trend, and despite the growing
popularity of shisha, to date there has been no study on the prevalence
of shisha consumption in a student environment. e present study
aimed to determine the prevalence of shisha use among university
students in Ouagadougou, Burkina Faso, and associated knowledge,
practices and beliefs about health effects, to generate helpful
information for interventions against shisha smoking.
Methods
Design and study population
Burkina Faso is a landlocked country located in the heart of West
Africa. Ouagadougou, the capital city, has a population of 2637 303.[11]
We conducted a cross-sectional study from October to December 2019
at UniversitéSaint omas d’Aquin (USTA) in Ouagadougou. USTA
is the biggest private university in the country, with 2 442 students
during the 2019/2020 academic year. It currently has ve faculties
(Legal and Political Sciences, Economics and Management, Health
Sciences, Science and Technology, Human Sciences and Society), one
institute (Higher Institute of Tertiary Trade) and one doctoral school
(Graduate School of Science, Health and Technology).
e study population was students from dierent faculties at USTA
enrolled during the 2019/2020 academic year, who consented to
participate in the study.
Sample size and sampling technique
A stratied sampling method was used, as students were stratied
into faculties. e number of students per stratum was proportionally
allocated. Students were selected by simple random sampling
using a student list made available by each department on request.
Studentswere approached in lecture rooms shortly before or aer
a class.
We determined the minimum sample size (no=385) using the
Cochrane formula for estimating a single proportion:
where p is the anticipated shisha prevalence in the population,
is the percentage of the standard distribution corresponding
to the two-sided signicance level (for the signicance level of 5%,
=1.96), and =0.05 is the level of precision.
A 10% non-response rate gives a total sample size of N=424 for
data collection.
Data collection
The questionnaire was adapted from previous studies on the
prevalence of shisha use conducted in Kampala, Uganda, and Kigali,
Rwanda.[8,9] We used CSpro soware, version 7.1 (US Census Bureau
and ICF International, USA), through CAPI (Computer Assisted
Personal Interviewing) using a mobile phone for data collection.
is soware allows administration of a face-to-face questionnaire in
which the interviewer uses a tablet on an Android system to conduct
the interview. e questions were closed, open or dichotomous.
Dependent variable
e dependent variable was shisha smoking. e information sought
on prevalence was based on the participants’ response to the question
‘Have you ever smoked shisha?’ We dened three types of smokers:
regular (an individual who smoked daily or at least once a week),
occasional (an individual who smoked less than once a week), and
experimenter (an individual who had smoked only once or twice
during their lifetime).
Independent variables
Our independent variables included sociodemographic characteristics
(including age, sex, marital status, level of education and place of
residence), smoking status of the parents and youths, educational
level of the parents, and knowledge about and attitudes of the youths
towards shisha smoking.
We dened knowledge about the health eects of shisha smoking as
the respondent’s ability to identify diseases associated with the practice
based on a list of eight diseases, knowledge about the harmfulness
of shisha to health, and knowledge about the harmfulness of shisha
compared with cigarettes. Correct responses to questions were
allocated 1 point, giving a possible total of 10 points for the questions.
Participants were then categorised as having satisfactory knowledge,
with a score ≥5, or poor knowledge, with a score of 0 - 4.
Data analysis
e data generated were entered into the system and analysed using
SPSS Statistics soware for Windows, version 25.0 (IBM Corp., USA).
At the univariate level, we calculated proportions for categorical
variables and summarised age using means and standard deviations
(SDs). A forward stepwise-ordered logistical regression model
established the factors independently associated with shisha smoking