Systematic reading, coding and re-categorisation were performed by the two first authors.
The tumour diagnoses were verified using histological methods, and pathological categorisations were performed according to the World Health Organisation (WHO) classification system.
For this reason an initial, naïve categorisation of data was performed without applying the coding scheme and special attention was paid to the passages in the text that could not be categorised according to the theory.
This notes review process and categorisation of results was performed exclusively by DNS for the first audit data, and exclusively by SK for the second audit data; both investigators used the same criteria mentioned above for result categorisation and identification of information from the notes.
Dichotomous categorisation of the independent variables was performed in order to carry out logistic regression analyses.
A qualitative categorisation of the participants' work-related activity descriptions was performed.
This selection was preceded by checking the linearity assumption for all potential predictors and if necessary categorisation (three groups) or dichotomization of potential predictors was performed.
The categorisation of the body site of the primary melanoma was performed as described before by studies based on the AJCC database (Soong et al, 2010).
As a control, additional training was performed on the same matrix providing a column with a random categorisation.
Subsequently, a multi-dimensional analysis was performed which cross-referenced the key subject areas with an earlier categorisation of the same errors using GEMS.
