Article Review: Hazardous Alcohol Consumption in Irish Universities

This report presents a detailed review of the article "Hazardous alcohol consumption among university students in Ireland: a cross-sectional study" by Davoren, Shiely, Byrne, and Perry (2015). The review follows STROBE guidelines, focusing on study size, statistical methods, and other analyses. The study involved 2,275 undergraduate students from University College Cork (UCC), selected using probability proportional to size (PPS) sampling. Various statistical methods were employed, including BMI measurements and regression/correlation analyses, to determine relationships between variables such as gender, courses of study, and alcohol consumption prevalence. The results indicated that a significant percentage of students met the criteria for hazardous alcohol consumption (HAC), with notable differences based on gender, field of study, and year of study. Further analyses revealed associations between HAC and factors like illicit drug use and house ownership. The study also utilized TeleForm TM scanning to estimate data entry error rates and employed descriptive statistics via IBM SPSS Statistics V.20 and WEMWBS score to examine the mental wellbeing of the sample population.

Running head: ARTICLE REVIEW 1
Article Review
Name
Institution

ARTICLE REVIEW 2
Article Review
This paper presents a review of the article, “Hazardous alcohol consumption among
university students in Ireland: a cross-sectional study” by Davoren, Shiely, Byme, and Perry
(2015). The review is guided by STROBE’s statement checklist numbers 10, 12 and 17
respectively. As such, the paper reports the study size, statistical methods, and other analyses
in the article.
Methods
Study Size
Checklist item number 10 of the STROBE statement requires the article review to
discuss how the researchers arrived at the study size. According to Davoren et al. (2015) a
total of 2,275 participants were used in the study. The sample was made up of undergraduate
students from the University College Cork (UCC) in Ireland. The sample was selected from a
population of 12,475 undergraduate students “using probability proportional to size (PPS)
sampling” (p. 2).
Moreover, the researchers ensured that ethical considerations and governance were
included in the research. Students were allowed to participate in the study voluntarily. Their
privacy and confidentiality was also guaranteed.
Statistical Methods
Various statistical methods were employed in the study by Davoren et al. (2015). For
instance, the article employed the probability proportional to size (PPS) sampling to select the
sample size of the study. Further, body mass index (BMI) was used to determine the weight
of each participant. The article utilized these statistical methods to determine the measures of

ARTICLE REVIEW 3
central tendency including the mean and mode of the sample population. For example the
article uses the measure of central tendency to determine the mean/average number of
participants in both genders (p. 2).
Moreover, Davoren et al. (2015) employ regression and correlation to determine the
relationship between the subgroups in the study. For example, the article uses regression and
correlation analysis to determine the relationship the statistical relationship between the
participants, the courses they are studying and the prevalence of alcohol consumption.
Results
Participants
First the study estimated a sample size of 2,686 participants according to a previous
unpublished masters dissertation and approached lecturers/module coordinators to allow them
distribute questionnaires during 15 minutes of lecture time between March 12 and March 23,
2012. Only 94.3% of those approached agreed to cooperate. Thus, a total of 2332
questionnaires were completed; but 57 students were found to be postgraduates and were
subsequently excluded from the study. Therefore, the data available was for study size of
2,275 undergraduate students.
Descriptive Data
The participating sample comprised of 63.2% females and 56% males. 39.7% were in
the college of Arts, Celtic Studies and Social Sciences, 20.1 % were in the School of
Business and Law, 24.6% in Science, Engineering and Food Science, 14.2% in Medicine and
Health. According to the college year, 46.8 % were first years, 7.7 % were fourth year
students and the rest were either in second or third year in college.

ARTICLE REVIEW 4
Outcome Data
Approximately, 57.4% of the participating females met the HAC threshold. Only 8.4
% of males and 5.8% of females were non-drinkers. “Approximately 17% of males and 5%
of females had AUDIT-C score of 10 or higher,” Davoren et al (2015). As such, the article
reports that 65.2% of males and 67.3% of females had HAC.
Main Results
The controlling variable in the study was that of age only. The results of the
multivariate analysis revealed that “males (OR=2.26, 95% CI 1.46 to 3.49; p<0.001) and
females (OR=2.12, 95% CI 1.44 to 3.14; p<0.001) studying Law and Business were over
twice as likely to report HAC as their peers studying Science and Engineering” (p. 3). Among
men, 56 % of third year students were “likely to report HAC (OR=1.56, 95% CI 1.02 to 2.41;
p<0.001), while among females those in fourth year were 80% more likely to report HAC
than their counterparts in first year (OR=1.80, 95% CI 1.14 to 2.86,” (p. 3).
While controlling for age, accommodation, course of study, type and year of study,
further analyses revealed that males and females using an illicit drug were more likely to
report HAC with males 95% CI 1.52 to 3.26; p=0.001 and females 95% CI 1.51 to 2.96;
p<0.001.
The analyses found that house ownership is negatively associated with HAC for both
genders as well as being a male second year. On the contrary, being a La or Business student
was found to have a positive association with HAC. Having multiple partners and illicit drug
use showed positive correlation with HAC including smoking among women.
Other Analyses
In the analysis of data, the study the TeleForm TM scanning process to estimate the
error rate for data entry. The study used descriptive statistics in the analysis of data on IBM

ARTICLE REVIEW 5
SPPSS Statistics V.20. The descriptive statistics is used to describe the relationship between
the sample and the data collected. For example, what the data tells us about the gender
distribution, the course, and the year of study in relation to HAC. More, the univariate and
multivariate logistic regression analyses were undertaken to investigate the factors associated
with high alcohol consumption separately in men and women” (p. 3). The article also uses the
WEMWBS score to examine the mental wellbeing of the sample population. Similarly, an
AUDIT-C score was employed to identify the patterns in alcohol consumption Davoren et al.
(2015).

ARTICLE REVIEW 6
Reference
Davoren, M. P., Shiely, F., Byrne, M., & Perry, I. J. (2015). Hazardous alcohol consumption
among university students in Ireland: a cross-sectional study. BMJ open, 5(1): 1-8.