Analysis of Hazardous Alcohol Consumption Among University Students in Ireland

This article analyses the research paper on hazardous alcohol consumption among university students in Ireland. The sample size calculation, confounding variables, descriptive statistics, logistic regression, and sensitivity analysis are discussed. The article also includes a regression analysis of self-reported average alcohol consumption and well-being of students of different genders.

401077 Introduction to Biostatistics, Autumn 2018
Assignment 3
Due Sunday June 10, 2018
“When submitting your assignment to Turnitin you are implicitly ticking these statements:
 I retain a backup file of this assignment in case the original file is lost or damaged.
 I hereby certify that no part of this assignment or product has been copied from any
other student’s work or from any other source except where due
acknowledgement is made in the assignment.
 I hereby certify that no part of this assignment or product has been submitted by me
in another (previous or current) assessment.
 I hereby certify that no part of the assignment has been written or produced by any
person.
I hereby certify that no part of this assignment has been made available to any other
student.
 I am aware that this work will be reproduced and submitted to plagiarism detection
software for the purpose of detecting possible plagiarism. This software may retain
a copy of this assignment on its database for future plagiarism detection.
 I understand that failure to uphold this declaration may result in academic
proceedings in line with the UWS Student Academic Misconduct Policy.”
Your name:
Your student number:
Question 1:
The present analysis is based on the research article “Hazardous alcohol consumption among
university students in Ireland: a cross-sectional study” by Davoren et al., (2015) published in BMJ
Open. Strobe Checklist for reporting of clinical observational studies has been taken as the basis for
analysing the paper.
Item 10
The required sample size was based on the information that the number of undergraduate students
at the University College Cork is 12,475. From previous studies (unpublished master’s thesis by
McCarthy, 2010) the authors have found that the incidence of HAC is 73%. Thus with a precision of
1.5% the required number of students was 2686.

Since the authors have based their sample size calculation from an unpublished thesis thus the
prevalence of 73% of students having HAC cannot be verified. However, if it is taken into
consideration that 73% of students do have HAC then the sample size calculation has been correctly
done.
Item 12
The research article did not use any confounding variable.
Thus the Strobe item 12a was not essential.
No subgroups were selected in the study.
Thus the Strobe item 12b was not essential.
The study did not provide for how missing data was handled.
Hence, the Strobe item 12c was not addressed.
The present research is a cross-sectional study. In the research paper under the section Results there
are clear mention of frequencies (with percentage) of the socio-demographic profile of the students.
The socio-demographic profile is segregated into different gender. In addition, the significance of
prevalence of HAC based on the socio-demographic is also provided. Moreover, HAC of age adjusted
odds ratio (with 95% CI) and multivariate logistics are also given
Thus the Strobe item 12d was well addressed.
Item 13
Davoren et al. in the methods and participants have stated that the number of students were
sampled using the method of probability proportional to size. Since the study was based on
university students all the students at the college were eligible for the study. The students were
asked to fill up a questionnaire. Thus there was no need for further follow-up.
Thus the Strobe item 13 was well addressed.
Item 14
In table 1, information is provided for descriptive statistics (counts and percentages) of the socio-
demographic profile of the students. Further, the self-reported body mass index, WEMWBS

(Warwick Edinburgh Mental Well-being scale) were also provided. Further statistical differences in
sub-groups was also provided.
Thus it can be concluded that item 14a has been addressed.
Item 15
In table 1 descriptive summaries of the socio-demographic profile of the students were provided.
The descriptive statistics for sub-groups were also provided.
Hence, Item15 for the Strobe’s Statement has been addressed.
Item 16
Table 2 presents the results of the Logistic Regression done to assess risk associated with HAC for
males and females. Age adjusted results for odds-ratio are given in the table. Further, the 95% CI for
age adjusted odds-ratio are given in the table. The results of multivariate analysis with 95% CI is also
given in the table.
In Table 3 the adverse consequences of alcohol consumption for male and females for both HAC and
non-HAC are provided.
Thus, Item 16 for the Strobe’s Statement has been addressed.
Item 17
No sensitivity analysis was done in the study. The subgroups in the study were not analysed in the
study.
Thus, Item 17 for the Strobe’s Statement is not necessary.

Question 2: Note: Students will get different answers as the data sets differ.
To assess the well-being of the student’s dataset 18861466 was taken. The variables of
concern is self-reported average alcohol consumption and sex of the students.
The analysis shows that the Gender distribution is as follows:
Table: Frequency count of Gender
Gender Frequency Percentage
Male 175 64.1
Females 98 35.9
Further the summary statistics for self-reported average alcohol consumption and well-
being score of 273 students is:
Table: Summary Statistics
Statistics Average Alcohol Consumption WEMWBS
Mean 8.699634 44.403663
sd 10.46868 3.82459
Median 5.0 44.3
IQR 11.0 5.1
1st Quartile 1.0 41.8
3rd Quartile 12.0 46.9
Min 1.0 33.9
Max 52.0 53.1

The average alcohol consumption of the students is skewed to the right. 50% of the
students reported that the average intake is below 5.0. On the other hand, the average
intake of alcohol is 8.699634 with a standard deviation of 10.46868. In addition, the
minimum and maximum average intake is 0.5 and 52.0 respectively. Moreover, from the
boxplot we find that there are outliers in the higher side. Thus it can be inferred that some
of the students consume higher level of alcohol.
The WEMWBS of the students is normally distributed. 50% of the students reported that the
WEMWBS is below 44.3. On the other hand, the average Well-being score is 44.403663 with

a standard deviation of 3.82459. In addition, the minimum and maximum Well-being score
of the students is 33.9 and 53.1 respectively. Moreover, from the boxplot we find that there
is one outlier in the lower side. Thus it can be inferred that the well-being score of one of
the students is poor.
Table: 95% Confidence Interval
Lower Limit Upper Limit
Alcohol 7.452263 9.947004
WEMWBS 43.94795 44.85937
Present the findings of relevant regression models and inferential analyses (about 150-200
words, 10 marks)
To report the self-reported average alcohol consumption from well-being of the students of
different gender a regression analysis was used. Gender was taken as a categorical variable
with male = 0 and female = 1.
The findings from the regression model is
Estimate Std. Error t-value Pr
Intercept 60.6072 6.7285 9.008 < 2e-16
Sex(T.female) -6.8538 1.1865 -5.776 2.10e-08
WEMWBS -1.1136 0.1491 -7.469 1.12e-12
Multiple R-Squared : 0.2215
Thus the average alcohol consumption can be predicted as:
Alcohol Consumption = 60.6072 – 6.8538*sex – 1.1136*WEMWBS
Since, sex is a categorical variable hence it had been factored with male = 0 and female = 1.
Thus the average alcohol consumption for males is
 Alcohol Consumption = 60.6072 – 1.1136*WEMWBS
And for females is
 Alcohol Consumption = 53.7534 – 1.1136*WEMWBS
Further, from the regression it is found that the coefficients of sex and wellbeing
significantly impact alcohol consumption (p-value < 0.05).
Further, from the regression analysis it is found that 22.15% of the average alcohol
consumption can be predicted from sex and wellbeing of the students.

Thus it is found that the alcohol consumption in male is higher than females.
Provide your answer to the research question (40-80 words, 4 marks)
Thus from an analysis of the data on wellbeing and alcohol consumption it can be inferred
that the average alcohol consumption of students of both sexes can be predicted from
wellbeing of the students of University of Western Sydney.

References
Davoren, M.P., Shiely, F., Byrne, M. and Perry, I.J., 2015. Hazardous alcohol consumption
among university students in Ireland: a cross-sectional study. BMJ open, 5(1)
Vandenbroucke, J. P., Von Elm, E., Altman, D. G., Gøtzsche, P. C., Mulrow, C. D., Pocock, S. J.,
... & Egger, M. (2007). Strengthening the Reporting of Observational Studies in Epidemiology
(STROBE): explanation and elaboration. Annals of internal medicine, 147(8), W-163.
Von Elm, E., Altman, D. G., Egger, M., Pocock, S. J., Gøtzsche, P. C., Vandenbroucke, J. P., &
Strobe Initiative. (2007). The Strengthening the Reporting of Observational Studies in
Epidemiology (STROBE) statement: guidelines for reporting observational studies.
Preventive medicine, 45(4), 247-251.