Dear Statalisters
A group of us are considering how Kaplan-Meier (KM) plots (as produced with sts graph) might be improved. We have come up with some ideas are are interested in your opinions. Below is a link to a survey which includes initial proposals. I would be grateful for Stata users' opinions on the most useful and most acceptable ones. Also, feel free to pass the survey to colleagues who might be interested.
Please could I ask people to refrain from posting opinions here to avoid influencing others' responses to the survey.
Many thanks, Tim
Problem
The standard way to present time-to-event data, such as survival, is with Kaplan-Meier plots. These are formatted by journals and reported in a number of ways, but commonly lack some key information.
The key problems are:
Suggestion
We have some initial proposals on how to improve Kaplan–Meier plots, and we need your help to know which (if any) would be the most useful and most acceptable to a wide audience.
Invitation to a short survey
Could you take our short survey of 13 meaningful multiple choice questions?
You’ll be asked to compare standard and alternative graphs, using data from one of three RCTs. To start the survey, go to: bit.ly/kmunicate.
The survey will be open until 31-May-2017.
Survey results
The findings will be written up for publication in a peer-reviewed journal.
We are interested to hear from anyone who looks at survival curves and are casting our net as wide as possible. Please forgive us if you have already received an invitation through another means.
If you have colleagues you think would be interested (including clinicians, journal editors, operations specialists, systematic reviewers, regulators, statisticians and trialists), please feel free to forward our invitation and link.
Thanks for your time.
Tim Morris, Chris Jarvis, Will Cragg, Babak Choodari-Oskooei, Patrick Phillips and Matt Sydes
A group of us are considering how Kaplan-Meier (KM) plots (as produced with sts graph) might be improved. We have come up with some ideas are are interested in your opinions. Below is a link to a survey which includes initial proposals. I would be grateful for Stata users' opinions on the most useful and most acceptable ones. Also, feel free to pass the survey to colleagues who might be interested.
Please could I ask people to refrain from posting opinions here to avoid influencing others' responses to the survey.
Many thanks, Tim
Problem
The standard way to present time-to-event data, such as survival, is with Kaplan-Meier plots. These are formatted by journals and reported in a number of ways, but commonly lack some key information.
The key problems are:
- Expressing how many people are contributing data at any point in the graph, including the pattern of censoring
- Expressing that the uncertainty of the estimate increases over time
Suggestion
We have some initial proposals on how to improve Kaplan–Meier plots, and we need your help to know which (if any) would be the most useful and most acceptable to a wide audience.
Invitation to a short survey
Could you take our short survey of 13 meaningful multiple choice questions?
You’ll be asked to compare standard and alternative graphs, using data from one of three RCTs. To start the survey, go to: bit.ly/kmunicate.
The survey will be open until 31-May-2017.
Survey results
The findings will be written up for publication in a peer-reviewed journal.
We are interested to hear from anyone who looks at survival curves and are casting our net as wide as possible. Please forgive us if you have already received an invitation through another means.
If you have colleagues you think would be interested (including clinicians, journal editors, operations specialists, systematic reviewers, regulators, statisticians and trialists), please feel free to forward our invitation and link.
Thanks for your time.
Tim Morris, Chris Jarvis, Will Cragg, Babak Choodari-Oskooei, Patrick Phillips and Matt Sydes
