A Retrospective Observational Cohort Study Describing Epidemiology, Patient Characteristics, Treatment Patterns and The Potential Burden Of Inhaled Short Acting Beta-2 Agonist (SABA) Over-prescription And Inhaled Corticosteroids (ICS) Under-prescription A

Date of Approval
Application Number
Technical Summary

Firstly, we will describe the current asthma population in England in people aged 4-17 in terms of SABA and ICS prescriptions, and the temporal patterns seen during the study period. Patients will be categorised according to their SABA and ICS use in terms of British asthma guidelines and Global Initiative for Asthma (GINA) guidelines. Next we will use a cohort study design to compare the effects of different patterns of SABA prescriptions on health outcomes (main outcome will be exacerbations; using Cox models) and healthcare resource utilisation (main outcome will be GP visits; using Poisson or negative binomial models, depending on distribution) (dataset will be all eligible current asthma patients in CPRD aged 4-17 eligible for linkage with HES-ONS). SABA prescriptions will be the main exposure, derived from 6 months of prescription data before the study start; the categories used will depend on the findings of the descriptive analysis. Models will be adjusted for multiple potential confounders, in particular ICS prescriptions. If the descriptive analysis shows patients change between SABA or ICS categories over time then this will be taken into account; firstly the regression models will censor patients once they change to a different exposure category, secondly, in sensitivity analyses we will use extensions to the regression models to take into account the time-varying exposures. If numbers allow we will investigate mortality as an outcome. We will also determine total costs (healthcare and environmental (CO2 emission) ). We will explore the use of other asthma treatments (leukotriene receptor antagonists (LTRA) and oral corticosteroids. We will also include co-existing conditions (allergies, asthma phenotypes, depression) in the models as appropriate.

Health Outcomes to be Measured

Incidence; prevalence; treatment patterns; exacerbations; health resource utilisation; mortality; costs; total CO2 emission


Ekaterina Maslova - Chief Investigator - Astra Zeneca Ltd - UK Headquarters
Jennifer Quint - Corresponding Applicant - Imperial College London
Ann Morgan - Collaborator - Imperial College London
Constantinos Kallis - Collaborator - Imperial College London
Graham Roberts - Collaborator - University of Southampton
Ian Sinha - Collaborator - Alder Hey Children’s NHS Foundation Trust
Ralf van der Valk - Collaborator - Astra Zeneca Ltd - UK Headquarters
Sofie Arnetorp - Collaborator - Astra Zeneca R&D Molndal Sweden
Trung Tran - Collaborator - Astra Zeneca Inc - USA


HES Accident and Emergency;HES Admitted Patient Care;HES Outpatient;ONS Death Registration Data;Patient Level Index of Multiple Deprivation;Practice Level Index of Multiple Deprivation