I am currently doing an Msc in Mental Health Studies at King's College London (Denmark Hill). For my thesis I am looking at change in therapy after patients have received Cognitive Behavioural Therapy for Psychosis (which is operating in the new demonstration site within the 'Increasing Access to Psychological Therapies' government run incentive).
i am wanting to measure change across therapy using a measure which will be taken at each therapy session for each patient (so there will be a lot of data points). The measure that I will be using will be recorded at weekly time points for the weekly form of the measure and 3 monthly for that version of the measure.
I have been previously advised that use of TSE will
'depend on how many data points you have per each phase (baseline, treatment, follow-up, or whatever other phases you've defined). 3 data points per phase is the minimum that you want, and 5+ is preferred. If I'm correct in assuming that your measurements will be weekly, I'd guess it's unlikely that you'll have this many data points--particularly if you have a pretreatment baseline, which would also have to have these 3+ data points and I guess it's unlikely that you have 3+ weeks of baseline measures.'
I have little knowledge regarding the use of time series analysis but I think it would be appropriate in this instance in my research, as I would really like to pinpoint where change is occuring across
therapy for each individual patient. I know that I could use mixed-model analysis, or indeed descriptive statistics, but I am aware that these methods are inferior in measuring change. In regards to the quotation this is what i have been advised but i was wondering if this was true and common consensus?
Any help would be much appreciated
Thanks
Alice
You may leave the list at any time by sending the command
SIGNOFF allstat
to [log in to unmask], leaving the subject line blank.
|