Fig. 5.

The unbiased tree replaces the composite organisational measure for the primary features of the organisation of the process of stroke care. The variable appears as the most predictively significant factor, followed by median weekly wage and the year dummy. The highest quality score belongs to hospitals with a high organisational score and with the neurophysiology variable taking a value located in wealthier regions. Variables day case theatres, nurse-to-bed, neurology, neurosurgery, teaching, FT, admissions, stork mortality and ‘no qualifications’ fail to appear in the tree