A prediction model for future exacerbation risk in children

Study type

• Study topic:
  Disease /health condition
• Study type:
  Non-interventional study
• Scope of the study:
  Assessment of risk minimisation measure implementation or effectiveness

  Disease epidemiology
• Data collection methods:
  Secondary data collection

Non-interventional study

• Non-interventional study design:
  Cohort

Study drug and medical condition

• Medical condition to be studied:
  Asthma

Population studied

• Short description of the study population:
  Patients with following criteria were included:; • Valid blood eosinophil count expressed as a numeric value ≤5000 blood eosinophils/µl, recorded at least 1 year prior to the end of available data; • Aged 5-12 years at date of last valid blood eosinophil count; • An asthma diagnosis (at any time); • 2 years of continuous data (one year pre/ one year post date of last valid blood eosinophil count).
• Age groups:
  Children (2 to < 12 years)

  Adolescents (12 to < 18 years)
• Special population of interest:
  Other
• Special population of interest, other:
  Asthma patients
• Estimated number of subjects:
  5000

Study design details

• Main study objective:
  The aim of this study is to create a tool to predict which paediatric patients are at risk of future exacerbation.
• Outcomes:
  1. Exacerbations:An exacerbation is defined as the occurance of the following: • Respiratory-related hospital attendance / admission AND/OR• Respiratory-related A&E attendance AND/OR• An acute oral corticosteroids course, 2. Blood eosinophil count3. Percent Predicted Peak Flow 4. Number of GP consults for lower respiratory tract infections5. Acute oral steroid usage6. Hospital in-patient admissions
• Data analysis plan:
  Univariable logistic regression models will be used to identify baseline measures of disease severity, patient demographics and comorbidities predictive of future exacerbations. The dichotomous variable indicating an exacerbation during the outcome period (YES/NO) will be used as the dependent variable with each measure of disease severity, patient demographic and comorbidity as an explanatory variable. Those variables which show an association (p < 0.05) with future exacerbation will be entered into a multivariable model and step-wise reduced to produce a final list of non-collinear predictors of one or more future exacerbations. Results will be presented as odds ratios (OR) with 95% confidence intervals (95% CI).