In cases when no definitive adjudication could be made on the basis of the above criteria, an adjudication committee consisting of clinicians with expertise in acute paediatric care assigned outcome by consensus, using all available information.
Patients who scored worse than the median - IQR (interquartile range) get assigned outcome 1 for that mental state outcome (e.g. speech and thought); those who scored median +/− IQR got assigned 2; and those who outperformed and scored over the median + IQR got outcome 3.
For the permutation testing, the entire selection procedure was repeated using a dataset with randomly assigned outcomes.
To be included in the model, the selection frequency of a biomarker in the dataset with nonpermuted (true) outcomes had to fall outside the 95% CI of its selection frequency using the dataset with randomly assigned outcomes.
The average follow-up time was 17 months and of those 21 it was only possible to clearly assign outcome in 14.
The strengths of our study include the large number of children studied, detailed review of all clinical and laboratory data to assign outcome, validation of the model in an independent part of the dataset, and inclusion of all children presenting with suspected SBI in a cohort study design, which suggests the predictive value of our SBI score should be generalisable to other similar settings.
The study pharmacist is the only member of the clinic team not blinded to the treatment arm and is not involved in assessing patients or assigning outcomes.
Ideally, one would randomly assign outcomes corresponding to each empirically or theoretically derived higher-order factor across samples to ascertain that their common variance is representative of the target construct.
To determine toolkit effectiveness, Lugtenberg et al's 20 method was adopted to assign outcomes from each toolkit to one of three categories: (1) not effective (if no significant effects were demonstrated); (2) partially effective (if half or less of the outcome measures showed significant effects) or (3) mostly effective (if more than half the outcome measures showed significant effects).
As previously proposed, non-measurable negative outcomes are assigned poor outcome quantitative values and analyzed using rank-based procedures [ 19].
For TB outcomes, the site of TB treatment, date of TB treatment initiation, date of TB treatment completion, date of assigned treatment outcome, and treatment outcome (cured, treatment completed, treatment defaulted, treatment failure, death) will be collected from study site TB registers as well as from the Department of Health's TB outcomes database.
