The SHMI methodology uses 142 statistical models, each corresponding to a different diagnosis group. Each of these models is constructed using a 3 year dataset and includes the following risk-adjustment variables: age, gender, method and month of admission, Charlson comorbidity index, year index and birthweight (for perinatal diagnosis groups only).
Statistics relating to the model fit for each of the 142 statistical models (model fit statistics) and model coefficients (model predict statistics) are published for the purposes of transparency. Details of the statistics provided in each of the files are available in the 'SHMI statistical model data definitions' file.
Notes:
1. Further information on data quality can be found in the SHMI background quality report, which can be downloaded from the 'Resources' section of the publication page.