MRI imaging for investigating medium-term effects of SARS-CoV-2
Category: Imaging Studies
This meeting will bring together statisticians, clinicians and other methodologists for discussion of common statistical issues in imaging studies. This meeting will focus on small group discussion of an article using MRI is the imaging method to monitor patient follow up. The issues discussed will be widely applicable to other imaging modalities.
The meeting will focus on small group discussions about the statistical challenges in this article: Statistical issues in designing and analysing a study to look at medium-term effects of SARS-CoV-2 infection after hospital discharge on multiple vital organs , exercise capacity, cognition, quality of life and mental health (PMC7808914)
The meeting will include small group discussion of statistical issues linking to this article including:
- Patient population enrolled
- Interpretation of study results
- Exploring multiple outcomes
- Communication of results
- Science in the time of COVID
Places are limited and will be allocated on a first-come, first-served basis.
To join the NIHR Statistics Group for information about future events, email: firstname.lastname@example.org.
13:30 Meeting introduction (Sue Mallett, Tom Nichols)
13:35 Statistical issues in designing and analysing a study to look at medium-term effects of SARS-CoV-2 infection after hospital discharge on multiple vital organs , exercise capacity, cognition, quality of life and mental health (PMC7808914)
- Dr Betty Raman (Division of Cardiovascular Medicine, Radcliffe Department of Medicine, University of Oxford)
14:05 Statistical challenges – structured small group discussion
14:25 Coffee and networking
14:35 Summary and discussion across all groups
SMALL GROUP DISCUSSION OUTLINES
- Participants will be invited to join break out rooms within zoom for small group discussions
- Participants can choose their preferred group
- We request small groups to nominate one person to feedback to the main group
Which patients does this study aim to research? How representative are patients in this type of study? What sorts of bias could this introduce into the study? What about the selection of controls? How could you design a follow up study to improve the representativeness of the sample?
What level of evidence does this study offer? Are these results purely exploratory and descriptive, or are some elements confirmatory? How would you design a follow-up study to address some of the limitations? Are there other data that could be collected or analyses that could be done to obtain stronger causal conclusions?
This was an exploratory study that looked at a large number of outcomes. What strategies could be used to minimise the multiple testing in future studies like this one that look at imaging outcomes? Are there other statistical challenges posed by holistic imaging approaches?
Discuss the importance of communicating the results, making interpretations of the findings in the scientific context while considering the statistical limitations of the sampling design, data and the methods. Are you aware of examples where Covid-related research has been communicated particularly well (or poorly?) in the public sphere?
SCIENCE IN THE TIME OF COVID
Discuss the challenges of conducting studies in the time of an epidemic. Will some patients be less likely to visit hospital? Impact on recruitment, protocol adherence, missing data, potential for bias, other issues? What were the major challenges you faced as statisticians during 2020? Were there advantages to working remotely?