If you attempted to attach a file it didn't make it.
Best anyway is to provide a small example data set in the form of data step code so that what you have is conveyed to us accurately. Any attachment of a file in another form such as Excel, csv, or text would require us to make assumptions to create data and we may not make the same decisions since we are not intimately familiar with all the aspects of your project.
First thing I would do is make sure that you have actual SAS date, time or datetime values. These are numeric with matching formats. If this does not sound familiar then ask.
With the date/datetime values for each measurement the next step would be to sort the data by patient and date. This will allow determining the time/duration since the previous measurement in a data step.
Simple code like:
Flagvalue = (Phmeasure > 0.5);
Would create a 1/0 numeric coded result for when the comparison is true.
With example data this could be used to create the requested proportions using the duration as weight .
Or create graphs.
There might be some fiddly bits about how you want to treat the intervals when the measure crosses your boundary value.
Are there interventions taking place that you are interested about before/after behavior?
