Uptake of self-management education programmes for people with type 2 diabetes in primary care through the embedding package: a cluster randomised control trial and ethnographic study

Table 3 Intervention impact in the intention-to-treat population (64 practices; 35,155 patients; 92,977 observations)

Abbreviations: CI, Confidence Interval; IQR, Interquartile range
^a The intention-to-treat population was used for the analysis with multiple imputation to impute missing values. However, means for the intention-to-treat population could not be generated using multiple imputation, so the complete-cases population was used for the summary statistics. The summary data are crude data and do not account for factors included in the model (imputation of missing data, nested random effects to account for non-independence of data, adjustment for covariates), therefore the crude estimates and model estimates are not directly comparable, which is why some effect sizes are in the opposite direction to the summary data
^b Pre-specified primary outcome

ISSN: 2731-4553