This repository contains code to accompany the paper "Pre-operative echocardiography for predicting perioperative cardiorespiratory adverse events in children with idiopathic pulmonary arterial hypertension undergoing cardiac catheterization"
Files included in this repository:
Identifies which covariates are associated with the occurence of an adverse event and combines them in a multivariable model. Multiple graphics are produced during the bootstrapped samples developing a model. Lastly, uses the median coefficients over the bootstrap to develop a scoring system which can be used clinically
Worker file to analyse all the input data and establish the data type for all columns of data from the input file ("Spreadsheet.xlsx").
Installs and loads the relevant R packages. Defines functions which are then used in the analysis. Loads the data iincluding the raw fields, column types, number of data rows and TAPSE reference ranges to allow conversion to Z scores. Formats the inputted data. Defines the complications and escalations of care and groups of variables relevant to the analysis.
Note: original data is not available due to privacy reasons. Code is provided to demonstrate analysis methods solely.
For more details please contact Tim Dawes at Timothy.Dawes [at] gosh.nhs.uk
Code to calculate necessary sample size to achieve 90% power, alpha = 0.05 based on effect size from Taylor et al, Risk of cardiac catheterization under anaesthesia in children with pulmonary hypertension. BJA, 2007.