I have a data set containing a daily measurement recorded from 20 participants for 60 days. I am trying to develop methods for predicting/estimating decline in long-term monitoring studies, i.e. can measurement of a parameters on a daily basis be used to detect/predict functional decline by identification of negative trends or abberant measurements. I hope to be able to generate an index of decline by examining the trends (using some combination of sensor derived parameters and referencing these to baseline clinical data (pre-trial clinical assessment including clinical scales meauring frailty and balance) and a daily health questionaire. 60 days was chosen as the maximum practical length of time where we could expect to see changes, also we do not have ethical approval to collect beyond this date.
What is the most appropriate method to define a baseline measurement for each participant and how do I best predict/detect negative trending or abberant behaviour in unseen data?
'Decline' in this case is somewhat difficult to define and in fact it is one of the research questions of this study (i.e. how do you identify decline relative to a baseline clinical assessment using daily measurements) however our intention would be to obtain statistically significant metrics that imply some clinical significance.
I have been using intra-class correlation coefficients and repeated measures ANOVA to examine the stability of each measurement over time. I am fitting a linear regression model (with a time trend) to investigate trends.
I wondered if I could get some other opinions on the soundness of this approach?