Methods
This was a retrospective study in which we collected the data of children diagnosed with cystic fibrosis from January 2010 to December 2020 at All India Institute of Medical Sciences, New Delhi. An institutional ethics committee approved this study. We included all patients with cystic fibrosis age less than 18 years under follow-up care of Pediatric chest clinic. We diagnosed CF by clinical phenotype and two sweat chloride values of >60 mEq/L or identifying two diseases causing mutations. Patients with incomplete data, precisely the diagnostic test results and whose outcome was not known, were excluded.
Data on demographic variables, clinical features, family history, type of mutations, compliance to therapy, age of airway colonisation with bacteria, lung function parameters, number of hospitalisations and duration of hospitalisation were extracted on a pre-designed proforma. Shwachman and Kulezycky (SK) severity scores were used to calculate and grade the severity of illness (7).
Statistical analysis : We entered data in a Microsoft excel sheet and exported it to Stata for analysis. Stata 15.0 [Stata Corp., College Station, TX] statistical software was used for data analysis. Comparison of all variables was carried out between those who died and survived. Continuous variables were compared using a t-test or Mann-Whitney test as appropriate. We reported categorical data as frequency (%) and compared using chi-square or fisher-exact test. Kaplan Meier analysis followed by the log-rank test was used to analyse the bivariable association of various potential factors with time to death as an outcome. Subsequently, stepwise Cox regression analysis with potential risk factors and time to death as the outcome was performed with inclusion and exclusion criteria as p (0.05) and p (0.1), respectively. In this study, a p-value of <0.05 was considered statistically significant.