ST332 & ST409 Medical Statistics: 2024-2025
Assignment [20%]
Deadline: 13:00 Thursday 13th March 2025
Background
Colleagues at University of Warwick and University Hospitals Coventry & Warwickshire (UHCW) NHS Trust have undertaken an unmatched case-control study to assess whether vaping (i.e. use of e-cigarettes) is a potential risk factor for Chronic obstructive pulmonary disease (COPD). They as well as collecting information on whether individuals had ever used or were current users of vape/e-cigarette products and whether they had been diagnosed with COPD, they also collected data on various covariates, including; their age, smoking history, socio-economic deprivation and sex.
Dataset
The R dataset ccvape.RData contains synthetic data on 300 cases (people who had been diagnosed with COPD) from clinics at UHCW NHS Trust together with 300 controls (people who had been diagnosed with COPD, but had been treated at UHCW NHS Trust for other conditions). The dataset contains the following variables and associated codes;
case (0=control,1=case)
vape (0=never user,1=former or current user of vapes/e-cigarettes)
age (age in years)
smoke (0=never smoked,1=current or former smoker)
dep (0=none,1=deprived)
male (0=female,1=male)
Assignment Tasks
PART A [Maximum 5 pages – 15 marks]
Answer the following questions based on the ccvape.RData dataset described above. Note it is not necessary to write a report – just answer the questions, but do not include R code or paste R output (construct appropriate tables – you can include high quality plots from R if you wish) – marks will be lost for this.
1. Undertake an Exploratory Data Analysis of the ccvape.RData dataset, in
particular identify any covariates which may be potential confounders. [5 marks]
2. Calculate and interpret the unadjusted Odds Ratio (OR) and it’s 95% CI and P-value to assess the association between vaping and being a case/control. [2 marks]
3. Using a binomial GLM calculate and interpret the Odds Ratio (OR) and it’s 95% CI and P-value to assess the association between vaping or not and being a case/control adjusting for the potential confounding variables you identified in Q.1. [3 marks]
4. Are any other variables also required to be included in the binomial GLM you fitted in Q.3? Justify your ‘final model’ and contrast the results obtained with those you obtained in Q.2 and Q.3. [3 marks]
5. Your clinical colleagues suggest that a 20% relative increase in the risk of being
diagnosed with COPD associated with vape use would be of public health importance (given the number of people in the population who now vape/use e-cigarettes). They also suggest that a 10% probability of making a Type II error and a 5% probability of making a Type I error would be appropriate in any future study. Describe how you would design (including a sample size calculation) a future study based on this information (and the results of your analyses in questions 1 to 4). [2 marks]
PART B [500 words maximum - 5 marks]
Based on your answers to PART A write a Press Release about the study for the University to make available to the media.
Marks will be awarded for;
• Appropriate interpretation of the results in PART A
• Discussion of strengths and limitations of the study
• Appropriate use of non-technical language for an informed lay audience
• Inclusion of an “anonymous” quote from you as “Researcher Jo Bloggs said … “” ”
• Discussion of potential next steps/further research
You can find examples of Press Releases on the University website at;
https://warwick.ac.uk/newsandevents/pressreleases/and two good recent health-related examples are;
https://warwick.ac.uk/newsandevents/pressreleases/?newsItem=8a17841b8defef98018df5c098261254and
https://warwick.ac.uk/newsandevents/pressreleases/?newsItem=8a17841a8d79730b018d9e2bbb0e054b
General Instructions
It is expected that PART A and B will be professionally presented (e.g. using LaTeX, WORD or R Markdown) and submitted on Moodle as a single pdf document. Do NOT put your name on the document or in the filename, please just use your student number. For PART B include a word count at the end of the Press Release. No appendices are allowed.
General questions about the assignment can be posted (anonymously if you wish) to the Discussion Forum on Moodle page for ST332, and these will be answered, questions emailed directly will not be answered.