Model | Analysis method | Stata code for analysis |
---|---|---|
A | Primary analysis: The PPPASI will be analysed using a linear mixed effects model, with fixed effects for time, treatment group, time-by-treatment group interaction and baseline PPPASI. The model will include a random intercept for participants and centre of recruitment. | mixed pppasi treat##i.time base_pppasi /// || centre: || id:, covariance(unstructured) reml |
B | If Model A fails to converge: Model A without the random-effect for centre, centre will instead be included as a fixed effect | mixed pppasi treat##i.time base_pppasi i.centre /// || id:, covariance(unstructured) reml |
C | If Model A fails to converge and > 1 site has ≤3 randomised: Primary analysis model excluding centre as a random or fixed effect | mixed pppasi treat##i.time base_pppasi /// || id:, covariance(unstructured) reml |