Analysis of Statistical Methods in Public Health Research Articles

This report analyzes two research articles. The first article, published in the Journal of Public Health, examines alcohol and drug use at school leavers' celebrations, utilizing Wilcoxon signed-rank tests and logistic regression to analyze data on adolescent substance use, negative consequences, and safety strategies. The second article, published in the International Journal of Public Health, investigates the rising trend of diabetes mellitus in a Chinese population through a population-based household survey, employing binary logistic regression to assess factors associated with diabetes prevalence. Both studies highlight the importance of statistical methods in public health research, demonstrating how these techniques can be applied to understand complex health issues, identify risk factors, and inform public health interventions. The report provides an overview of the methodologies, findings, and implications of each study, emphasizing the role of statistics in drawing meaningful conclusions from the data.

Running Head: IDENTIFICATION AND IMPLEMENTATION OF STATISTICS IN
RESEARCH ARTICLES
IDENTIFICATION AND IMPLEMENTATION OF STATISTICS IN RESEARCH ARTICLES
Name of the Student:
Name of the University:
Author Note:

1IDENTIFICATION AND IMPLEMENTATION OF STATISTICS IN RESEARCH ARTICLES
Paper 1: Lam, T., Liang, W., Chikritzhs, T., & Allsop, S. (2014). Alcohol and other drug use
at school leavers' celebrations. Journal of Public Health, 36(3), 408-416. Retrieved from:
http://jpubhealth.oxfordjournals.org/content/early/2013/08/26/pubmed.fdt087.full.pdf+html
It has been observed that a major proportion of youngsters in countries like USA, UK and
Australia are often engaged in substantial alcohol and other drugs consumptions and thus
suffering from blackouts, hangovers and violence. The main objective of this paper is to study
alcohol and drug uses at school graduates’ celebrations by adolescents.
1. The analysis is composed of two parts. In the first part, the aim is to collate the levels of
AOD use at a school leavers’ programme and other peer-based social events. For this, the
null and alternative hypotheses can be considered as,
Null hypothesis: There is no significance difference in the levels of AOD use at an end of
school function and other social celebrations. In other words, adolescents are found to be
addicted by same amount in a school leaving celebration or any social function.
Alternative hypothesis: The amount of AOD usage at a school-end programme differ
significantly from that at other social gatherings.
Therefore, the experimental variable is the levels of AOD used and the independent
variables are school-leaving celebration and other peer-based social programmes.
The second part of this study shows the negative consequences of the AOD use and
implementation of protective strategies. The hypotheses for this part are taken as,
Null hypothesis: There is no negative consequences of AOD use at the celebrations and
safety strategies has been used frequently to avoid common drug-related harms.

2IDENTIFICATION AND IMPLEMENTATION OF STATISTICS IN RESEARCH ARTICLES
Alternative hypothesis: There are notable negative issues related with AOD use and other
risk factors at the celebration.
Here the independent variables are the levels of AOD use, safety strategies used, gender,
accommodation location, survey modality and the dependent variable is the different
types of negative consequences related to AOD use.
2. The data for this venture were collected based on a two-stage survey plan with the help of
a self-report methodology. The location of the study was Rottnest Island which is a very
popular place for events in Western Australia. The first survey was conducted among
young persons who were interested to be at the event and the post-celebration survey was
conducted on the youngsters attended the party. The surveys were available both online
and person-to-person however the second survey was mostly conducted face-to-face.
A self-rated study is a type of survey to ask participants about their feelings, attitudes and
so on. The motive of this study is to observe the outcomes of substantial alcohol and
other drugs uses on the youngsters by assessing their intentions and what happened at the
party. Hence, the self-report study is appropriate to collect data in this project. The main
advantage of this study is that it is easy to conduct and privacy of the participant is
maintained. However people are often biased when they report on their own experience
which reduces the accuracy of the sampling design.
3. The teenagers of 17-18 years old represent the demographic characteristics of people in
the study. It has been observed that both male and female participants have almost equal
consumption of AOD use. The legitimate purchase age for alcohol in Australia is 18
which leads to drug use at young age. The respondents of this survey were the young
adults who seem to be more interested in the activities like substantial alcohol and other

3IDENTIFICATION AND IMPLEMENTATION OF STATISTICS IN RESEARCH ARTICLES
drugs consumption and other adulthood enjoyments. The descriptive study shows that
most of the significant AOD users are the youngsters who have just completed schooling.
4. At first, a sequence of Wilcoxon signed rank tests were executed to compare if the level
of AOD used at the last event was different from an average day at the school leavers’
programme. Then a logistic regression is performed to observe the repercussions of six
factors on the negative consequences.
The Wilcoxon-signed rank test shows that a remarkable larger proportion of
adolescents consumed alcohol during the celebration period (93%), compared with the
last peer-based social event (78%). Male and female participants consumed 18.44 and
13.24 Australian SDs on average per day during the school leavers’ celebration. There
were 112 students at the last event and 30 at the programme who did not drink. Further,
males were more prone to use of amphetamine, cannabis, ecstasy and illicit drugs in a
greater proportion during the celebration than the last event, compared with that of
females.
The logistic regression shows that majority of the participants (87%) faced at least
one negative consequences at the celebration after AOD use.
5. In the logistic regression analysis, 17 negative consequences of AOD use have been
showed. Among them, four were associated with the safety strategies and these are
hangover, blackouts, vomiting and unprotected sex. The Odds Ratio shows that
respondents who did not take protective strategies too frequently were 10.92 times more
likely to report unprotected sex, compared to those engaged in safety schemes with
highest frequency. Further, the prevalence percentage of unprotected sex is 13.62 among
301 male and female participants. This indicates that the youngsters are incapable to

4IDENTIFICATION AND IMPLEMENTATION OF STATISTICS IN RESEARCH ARTICLES
control the situation in the field. They have insufficient knowledge about the drug usage
effects which lead them to such negative experiences. Consequently, the rate of
unprotected sex after AOD use is significantly high among the teenagers. Hence it can be
concluded that the condition of adolescents in these Western countries are highly critical
since they are too much attracted to substantial alcohol and drug use at social events and
specially at their school leavers’ function.
6. The original sample was collected from school students of Western Australia which
shows a significant tendency of the teenagers to consume alcohol and other drugs at
different events. In Australia, teens get legal permission for purchasing AOD at age 18.
This is the reason that around 90-95% of the young adults have tendency to use alcohol
and other drugs at social events and school leaving parties. They consider it as an
essential part for any social function. The sample shows that the attitudes of the teenagers
is almost same within the western countries. This makes the sample a good representative
of the conditions of youngsters across Western Australia.

5IDENTIFICATION AND IMPLEMENTATION OF STATISTICS IN RESEARCH ARTICLES
Paper 2: Wong, M. C., S., Leung, M. C., M., Tsang, C. S., H., . . . Griffiths, S. M. (2013). The
rising tide of diabetes mellitus in a Chinese population: A population-based household survey on
121,895 persons. International Journal of Public Health, 58(2), 269-276. Retrieved from:
http://dx.doi.org.ezproxy.laureate.net.au/10.1007/s00038-012-0364-y
Diabetes mellitus has been recognised as a potential health crisis worldwide. According to some
reports, a major part of the world population will be affected by this disease in next few years.
The paper shows research on the upward trend of diabetes mellitus among Chinese people.
1. The principal motto of this study is to find out the increasing trend of diabetes among
Chinese people. It is a population based household survey and the survey was performed
by interviewing person from territory wide household during the period 2001-2008.
Moreover, the research involves the examination of the factors that are independently
associated with diabetes. Here the null and alternative hypotheses can be considered as,
Null hypothesis: There is no noteworthy difference in effect of different factors
associated with diabetes.
Alternative hypothesis: The effect of the factors associated with diabetes differ
significantly.
2. The demographic features of the people selected for consideration are the Chinese
household population and most of the respondents are more than 15 years old. There
were 121,895 male and female respondents in total and among them 103,367 aged 15
years or more. According to the paper, the level of the participants was notably good and
it is capable of showing the demographic characteristics of Chinese community. The
report is based on the characteristics and lifestyle habit of the people. Moreover the

6IDENTIFICATION AND IMPLEMENTATION OF STATISTICS IN RESEARCH ARTICLES
household survey covered almost 96% of the total Hong Kong residents. The estimated
response rate was 75% across the survey rounds.
3. In this paper, binary logistic regression has been used to analyse the data taking self-
reported diabetes as dependent variable and age, sex, monthly household income as set of
predictors. Binary logistic regression is an extension of simple linear regression which
models the probability of an event of interest based on some factors. Here the adequate
sample size for observing the Diabetes mellitus in Chinese population is 121,895. The
analysis is performed based on the data collected from 2001, 2002, 2005 and 2008. The
regression shows that advanced age, low monthly income of the respondent and the
household are directly related to the self-report of diabetes mellitus. In other words, if age
increases or the financial crisis of the household increases, then number of diabetes cases
will also increase. The result helps the researcher to understand the medical condition of
the households. Overall, the main objective of the study that is the rising trend of diabetes
can be visualised with the help of the logistic model.
4. The survey was conducted among 121,895 people and among them, 130,367 were above
age 15. The prevalence escalated by approximately 50% in the period 2001-2008, and the
increment was extreme among females (69.3%) than males (47.9%). However, there was
no significant incompatibility in prevalence based on gender when the prevalence was
balanced with age and sex of an individual. Among male participants, the age and sex
modified prevalence of diabetes in 2001,2002,2005,2008 was 2.80%, 2.87%, 3.32% and
4.66% respectively. For female respondents, the age and sex adjusted prevalence in those
years were 3.25%, 3.37%, 3.77% and 4.31% respectively. It can be observed that the
adjusted prevalence increased sharply among the poor Chinese people in these years.

7IDENTIFICATION AND IMPLEMENTATION OF STATISTICS IN RESEARCH ARTICLES
Further, a progressive growing tendency of self-reported diabetes was observed among
people who aged 75 years or more across the years 2001-2008. The study shows the
efficiency and important information regarding the development of the situation.
5. The Odds Ratio obtained from the binary logistic regression shows that people with low
income are at higher risk of diabetes than rich people due to their lifestyle and food habit.
These people are not able to afford healthy foods which may prevent diabetes and other
health hazards. Moreover, they have lack concern about health problems until it becomes
vital. The research also shows that women are more exposed to the risk of diabetes
mellitus due to their working function and lifestyle activities. The trend of self-reported
diabetes shows a sharp increase with increase in age. Further, if a family has diabetes
history, then the members of the family are exposed to the risk of diabetes. The report
shows that people are facing high health issues with rising trend of self-reported diabetes.
This is a very common disease and anyone can be affected from it despite of their age or
sex. However unhealthy lifestyle and food habit and genetic history play important roles
in rising trend of diabetes.
6. The study included population based random sampling method to generalize the findings
across different time periods. However there are some limitations of this study. The
sample is based on self-reported information which is not always a reliable source since
there is no knowledge about the validity of these data. Further, variables like lifestyle,
family history, BMI which may be have significant influence on the prevalence of
diabetes are not considered in the study. Children (below age 15) are kept out of the study
hence any information regarding diabetes among children are not taken into account. The
regression model can explain only 20% of the divergence in diabetes prevalence

8IDENTIFICATION AND IMPLEMENTATION OF STATISTICS IN RESEARCH ARTICLES
approximately and if other risk factors are considered, then that might improve the
model.

9IDENTIFICATION AND IMPLEMENTATION OF STATISTICS IN RESEARCH ARTICLES
Bibliography
Corder, G. W., & Foreman, D. I. (2014). Nonparametric statistics: A step-by-step approach.
John Wiley & Sons.
Jongenelis, M. I., Pettigrew, S., Biagioni, N., & Hagger, M. S. (2017). Western Australian
students’ alcohol consumption and expenditure intentions for Schoolies. Australian journal of
primary health, 23(3), 268-271.
Lam, T., Liang, W., Chikritzhs, T., & Allsop, S. (2014). Alcohol and other drug use at school
leavers' celebrations. Journal of Public Health, 36(3), 408-
416. https://doi.org/10.1093/pubmed/fdt087 Retrieved from:
http://jpubhealth.oxfordjournals.org/content/early/2013/08/26/pubmed.fdt087.full.pdf+html
Lubman, D. I., Droste, N., Pennay, A., Hyder, S., & Miller, P. (2014). High rates of alcohol
consumption and related harm at schoolies week: a portal study. Australian and New Zealand
journal of public health, 38(6), 536-541.
On Biostatistics and Clinical Trials (2012). Retrieved from:
http://onbiostatistics.blogspot.com.au/2012/02/how-to-interpret-odds-ratios-that-are.html
Ott, R. L., & Longnecker, M. T. (2015). An introduction to statistical methods and data analysis.
Nelson Education.
Shih, W. J., & Aisner, J. (2015). Statistical design and analysis of clinical trials: principles and
methods. CRC press.
Taheri, S. M., & Hesamian, G. (2013). A generalization of the Wilcoxon signed-rank test and its
applications. Statistical Papers, 54(2), 457-470.

10IDENTIFICATION AND IMPLEMENTATION OF STATISTICS IN RESEARCH ARTICLES
Wong, M. C., S., Leung, M. C., M., Tsang, C. S., H., . . . Griffiths, S. M. (2013). The rising tide
of diabetes mellitus in a chinese population: A population-based household survey on 121,895
persons. International Journal of Public Health, 58(2), 269-276.
http://dx.doi.org.ezproxy.laureate.net.au/10.1007/s00038-012-0364-y
Yang, L., Shao, J., Bian, Y., Wu, H., Shi, L., Zeng, L., ... & Dong, J. (2016). Prevalence of type
2 diabetes mellitus among inland residents in China (2000–2014): A meta‐analysis. Journal of
diabetes investigation, 7(6), 845-852.