27286 Practical Epidemiology and Statistics
Hello, dear friend, you can consult us at any time if you have any questions, add WeChat: daixieit
Practical Epidemiology and Statistics
(Module code: 27286)
Assessment brief 2023/24
Referencing
You are not required to cite additional evidence for this assignment, but you may find it
helpful to do so, e.g. to justify your choices or explain/support certain aspects. However, if you do use other sources, you MUST reference these appropriately by providing a citation
within the text and a list of references at the end (separately for each part of the assignment).
Acknowledging and Citing the Use of Generative Artificial Intelligence (AI)
The University of Birmingham permits the use of Generative AI for the following purposes: 1. Scope out a topic or develop your initial ideas
2. Develop personalised study resources
3. Test your critical thinking skills
4. A revision aid to test knowledge
5. Improve your time management skills
6. Understand the fundamentals of academic writing
7. Editing and proofreading work
8. Summarise and understand research
9. Develop effective search techniques using prompts
For more information, please see:
https://intranet.birmingham.ac.uk/as/libraryservices/asc/student-guidance-gai.aspx
Use of generative AI in assessments
Where you have used generative AI to support an assessment you must acknowledge its
use. Failure to appropriately reference externally sourced, non-original work can result in
academic misconduct as per the University’s Code of Practice on Academic Integrity. You
should name the tool (e.g. ChatGPT) and how it was used by including the appropriate
statement (see below) in an appendix at the end of your assignment titled Acknowledgement of Generative AI Use:
. No content generated by AI tools has been presented as my own work.
. I acknowledge the use of [insert AI system(s) and link] to generate materials for background research and self-study in the drafting of this assessment.
. I acknowledge the use of [insert AI system(s) and link] to generate materials that were included within my final assessment in modified form. The following were initially generated by AI:
Where you have used a generative AI tool to generate material included in your assignment, you must describe how the information or material was generated (including the prompts you used with the generative AI tool), what the resulting output was and how the output was then changed by you. As noted above, failure to appropriately reference externally sourced, non- original work can result in academic misconduct as per the University’s Code of Practice on Academic Integrity. You should use the following style of wording, depending on the nature of use:
. The following prompts were input into [AI system]:
. The output obtained was:
. The output was changed by me in the following ways:
Part A1 – Critical appraisal
Task: Using the checklist on the following page, write a structured critical appraisal of the paper below by Smulders et al (available on Canvas).
Paper:
Smulders E, et al, Efficacy of a Short Multidisciplinary Falls Prevention Program for Elderly Persons With Osteoporosis and a Fall History: A Randomized Controlled Trial. Archives of Physical Medicine and Rehabilitation 2010;91:1705-11. DOI: 10.1016/j.apmr.2010.08.004
Reporting structure and allocation of marks
Your answer to Part A1 should be structured using the sections provided in the checklist on the following page (i.e. study question, study design, results, implications). Under each of
these sections, you need to address the items listed in the checklist. The appraisal should be written in the style of an essay, using the sections from the checklist as subheadings.
Marks for part A1 will be awarded in the following proportions:
. Study question: 10%
. Study design: 40%
. Results: 30%
. Implications: 10%
. Presentation, including logical flow, correct use of grammar/spelling 10%
Maximum marks will be given based on the following criteria:
(1) Clarity and accuracy of the descriptions of the study aspects. For example, what was the focus of the question, how were participants recruited, what was the exposure and how was it measured, what were the study results etc.
(2) Critical appraisal of the methods used (e.g. was the study design appropriate, to
what extent have the methods chosen minimised bias, confounding and chance)
(3) Correct description and interpretation of the study results
(4) Appropriateness of considerations regarding implications of the results
Critical appraisal checklist
This checklist contains the key elements that need to be included in your critical appraisal of the paper. The checklist is drawn from the Critical Appraisal Skills Programme (CASP)
checklists created by University of Oxford.
1. Study question (10% of marks)
a. What is the study question? Did the study address a clearly focused issue? b. What are the PICO/PECO elements?
2. Study design (40% of marks)
c. What type of study design did the authors employ and was this appropriate for answering the study question?
d. How were participants recruited, what were the inclusion/exclusion criteria and could this method have introduced bias?
e. What methods were used for randomisation and allocation concealment? Could these have introduced bias?
f. Were patients, researchers and other study staff blind to the intervention assignment? Why is blinding important?
g. Aside from the experimental intervention, were the groups treated equally?
h. How were the outcomes measured? Could this method have introduced bias? Was there a power calculation?
i. The authors say that analyses were performed according to the intention-to-treat principle. What does this mean, and what is the strength of this type of analysis?
3. Results (30% of marks)
j. Were all participants accounted for at the end of the trial (in terms of those included in the analysis for the primary outcome only)? Was there any missing or incomplete data? Could bias have been introduced?
k. What were the baseline characteristics, and did these differ by group? Could any differences have impacted the findings?
l. What were the results in terms of the primary outcome? Discuss the precision and statistical significance of the results. Very briefly summarise the findings for
secondary outcomes.
4. Implications (10% of marks)
m. How strong is the evidence, and what are the implications for public health in the UK
and globally? Consider generalisability/external validity in your answer.
(The remaining 10% of marks is for presentation and clarity.)
Part A2 – Research study design
Childhood obesity has been linked to an increase in the risk of a number of different health
issues and diseases later in life, including type II diabetes. It has been suggested that
childhood obesity increases insulin demand, which could place greater stress on the β-cells and make them more vulnerable to autoimmune attack – leading to the destruction of these cells and development of type I diabetes. However, more evidence is needed to confirm a
link between childhood obesity and risk of type I diabetes. Type I diabetes is most commonly diagnosed at about age 14.
Research question
“Does obesity in childhood increase the risk of developing type I diabetes?”
Task: As a first step in designing a study to answer the research question above, answer the following questions and justify your choices:
(1) What study design will you use, and why?
(2) Who will the study population be, including inclusion/exclusion criteria? How will participants be recruited?
(3) What would you measure in order to investigate this study question and how would you collect data about these study measures? Think about your
exposure/intervention, comparator/control, outcome(s), and any relevant
confounders/covariates. Also think about study duration and follow-up period. Describe any important definitions you will use, for instance, in defining diagnoses.
Note: You do not need to describe how you would analyse the data collected.
Marks for part A2 will be awarded in the following proportions:
. Study design 20%
. Population 20%
. Study measures 60%
Part B – Data analysis
Assessment submission
Section B (data analysis) is worth 40% of the overall assessment mark for the Practical
Epidemiology and Statistics module. This assignment requires you to produce a poster
which answers the questions posed on the following page. The poster should contain a brief description of the study, details of the methods used to analyse the data, a summary of the results of the analysis, and a discussion of the results. You should include tables and figures (as appropriate) to illustrate your analysis.
The poster should be created as an A0 size (1189mm x 841mm) and it can be produced
using commonly available software e.g. Microsoft PowerPoint. You can choose whether to
present the poster in portrait style or landscape style. A template can be found on Canvas, in the Practical Epidemiology and Statistics module page (in the assignment section). Like
other assessments, you should not put your name on your poster, but only your student ID number. Please put your student ID in the top right-hand corner. The minimum font size is 12pt. References and abstract are optional – however, please note that if you do make
use of other sources, these must be referenced. There is no need for pictures or diagrams, and only include figures and tables that support your results. You should attempt all stages of the analyses that are described below.
Guidance on general rules for production of this document is given in the MPH assignment submission guidelines (see the Handbook, or Canvas), which you must follow. In particular, you are reminded of the Department’s guidelines on Plagiarism, which can be found in the Handbook. Check to ensure that your submission meets the regulations.
A PDF version of the poster should be submitted via Canvas by 12th December 2023.
Task: What patient characteristics impact the time to cardiovascular disease (CVD) occurrence among adults
Study details
The dataset you are provided with (peas_dataset_2023.dta) is from a cohort study to
examine the relationship between patient characteristics and the time to cardiovascular
disease (CVD) occurrence among adults. The study included a large population-based
cohort of adults aged 40-65 years without a history of CVD at baseline. Participants were
followed for an extended period, during which incident cases of CVD were identified. You are provided with information on various patient characteristics, including age, gender, smoking status, body mass index (BMI), air pollution exposure, and socio-economic status.
The task
Using the dataset provided, your task is to investigate the relationship between participant characteristics and cardiovascular disease. You should also investigate whether the relationship between participant characteristics and cardiovascular disease differs by smoking status.
The dataset you have been given includes, for each participant, the outcome, cardiovascular disease (cvd_occurrence), and participants’ demographic information. The variables are summarised in the table below.
The stages of your analysis should be:
Summary statistics table
1. Create a table of descriptive summary statistics to describe all the characteristics of the study population, stratifying as appropriate. You should briefly discuss how the characteristics compare across the stratification variable.
Analysis
1. Perform univariable analysis. Examine and describe the impact of the participant
characteristics on the outcome. Place particular emphasis on the point estimate and the confidence interval.
2. Perform multivariable analysis. Examine and describe the impact of the participant characteristics on the outcome. Place particular emphasis on the point estimate and the confidence interval.
3. Perform multivariable analyses separately for smokers and non-smokers. Discuss whether you think there is any evidence of a differential effect of the participant
characteristics in smokers and non-smokers.
4. Summarise the limitations of the study (or the information you have been told about it) and what, if any, the clinical value might be from your findings.
Table of variables in the dataset
Variable |
Variable name |
Description |
ID |
ID |
Individual patient identifier |
Age (year) |
age |
Participant age in years |
Exposure to air pollution |
air_pollution_exposure |
0 Low; 1 High |
Body Mass Index |
bmi |
Body Mass Index (kg/m2) |
Follow up time |
follow_up_time |
Follow up time (years) |
Gender |
gender |
0 Male; 1 Female |
Smoking Status |
smoking_status |
0 Non-smoker; 1 Smoker |
Socio-economic status |
socioeconomic_status |
1 Low; 2 Medium; 3 High |
Outcomes |
|
|
Diagnosis of CVD |
cvd_occurrence |
0 No CVD; 1 CVD occurred |
Accessing the data file
Data for this assessment is available on Canvas, in the assignments section of the Practical Epidemiology and Statistics module page. The file is called peas_dataset_2023.dta. In the event of any problems with accessing the data, you can email James Martin (j[email protected]) for a copy.
Note: you should ensure that you have a working copy of this dataset as soon as possible. Requests for extensions due to last minute problems with file access are unlikely to be viewed favourably.
Reporting structure and allocation of marks
The task for your assessment is to analyse the dataset appropriately to answer the questions posed as clearly and fully as you can.
Your work should be submitted as an academic poster, with sections (such as: Introduction, Methods, Results, and Discussion), tables and figures (as appropriate), to illustrate your investigations. Statistical output should be presented carefully, and interpreted and discussed where necessary. Statistical output should not come directly from Stata. You may wish to look at the level of presentation in published papers and other academic posters as guidance. The focus of marking for this assignment is on answering the question posed, including the reporting of your methods and statistical analysis, the interpretation of your results, and the structure of the poster. The introduction is not a key focus for our marking, so we only expect students to briefly summarise the study based on the information provided in the question.
You are not expected to find outside sources for your introduction.
You may wish to use the following sections:
. An Introduction section. Here you should very briefly summarise the study and describe the objectives of your investigations. There is no need to include a review of literature. This brief summary should come from the information provided in the question.
. A Methods section detailing all data and analysis-related methods that you have used, justifying and explaining their selection and any assumptions made.
. A Results section. Here you should present your findings. Tables and Figures should be used where appropriate, with an explanation of their context. Be sure that the results presented answer the assignment question.
. A Discussion section. Here you should explain what the results mean in the context of the assignment question. You should also detail any limitations of the results.
Credit will be given for:
. Clarity of structure and organisation of your poster.
. Quality of presentation of your work:
- Use sections & headings (see above for examples).
- Application of principles of good presentation in tables and graphs.
- You should NOT present output directly cut and pasted from Stata (or any other analysis package). You should edit and arrange it to show the results as clearly as possible.
- Avoid unnecessary repetition.
. Selection and execution of appropriate statistical methods for the questions you are investigating, i.e.:
- Description and summary of data set.
- Appropriate data analysis.
. Demonstration of knowledge of when the methods you use may be applied, any assumptions you make in using them, and consideration of their shortcomings or limitations.
. Clear, plain English discussion of what the analyses mean, and the conclusions you draw.
Marks will be awarded in the following proportions:
. Clarity of structure, organisation, quality of presentation: 30%
. Content: selection and execution of appropriate methods: 40%
. Content: discussion and interpretation of analyses: 20%
. Critical insight into conclusions, summarisation: 10%
2023-12-04