I have longitudinal data for patient-reported disease activity in multiple individuals. I want to quantify variability (or stability) over time and then look for baseline predictors. What is the best approach?Thanks!
How exactly was disease activity measured, how often over which period was
it measured, how large is your sample size?What kind of baseline predictors will
you use, and how many of them are there, and do you already have an idea
which predictors could be particularly useful? And could you elaborate a bit
what you mean by variability in the present context, and why it is useful
to know about it?
Thanks. It is a measure of disease activity in rheumatoid arthritis (RAID). We have longitudinal data over 18 months. I want to be able to describe variability or stability over time. For instance, how many patients who start in remission or low disease activity stay there? Beyond giving this as a numeric figure, e.g. 50% remain in remission over 12 months, is there a better way of describing this and comparing this stability with patients in different ranges of disease activity at baseline. At the moment we only have 100 patients but expect to have another 500 after a year. Predictors would include clinical variables such as blood tests, baseline demographics and individual components of the RAID which is a multidimensional 7 domain questionnaire.
How exactly did you measure disease activity? Remission or low activity seems to indicate an ordinal masurement. How often did you measure it, per hour, daily, weekly, monthly?
What proportion of patients are lost during the course of the study?
Why do you analyse these data, is it for immediate use in clinical practice?