Incident Neutropenia in New Users of Cyclooxygenase-Inhibitors: An Observational Cohort Study

Date of Approval
Application Number
Technical Summary

We will conduct a propensity score matched retrospective cohort study to evaluate the absolute risk of neutropenia in new-users of cyclooxygenase-inhibitors (COX-Inhibitors) and the relative risk of neutropenia across different COX-Inhibitors, and of COX-Inhibitors compared to non-use of COX-Inhibitors.
We will include patients aged 2 years or older who were registered in CPRD GOLD at any time between 1990-2018 and who newly used a cohort defining COX-Inhibitor (diclofenac, ibuprofen, naproxen, coxibs, paracetamol). New use will be defined as a prescription of a cohort defining COX-Inhibitor after a period of at least 180 days in which no COX-Inhibitors were prescribed. The date of the first-time prescription will be called cohort entry date (CED). Additionally, we will identify a cohort of randomly selected non-users of COX-Inhibitors. We will exclude patients who had a record of neutropenia, agranulocytosis, HIV, substance abuse, chemotherapy, cancer, or diseases associated with neutropenia prior to CED. We will follow patients from CED until the first of the following occurs: a recorded neutropenia or an exclusion criterion, the end of drug exposure, the end of the study period, the end of data contribution or 365 days after CED, initiation of another cohort defining drug, or death. In each cohort we will calculate incidence rates of neutropenia and the proportion of patients developing neutropenia during follow-up. We will conduct propensity score matched Cox-proportional hazard analyses to calculate hazard ratios with 95% confidence intervals of the risk of neutropenia across COX-Inhibitors and of COX-Inhibitors compared to non-use. We will perform subgroup analyses by sex and age, and sensitivity analyses in which we will restrict the length of follow-up to 90 and 180 days.

Health Outcomes to be Measured

i. Incident neutropenia in new-users of cyclooxygenase-inhibitors (COX-Inhibitors).
ii. All-cause mortality within 30 days after incident neutropenia.
We will not measure sepsis or IV antibiotic use within 30 days of neutropenia as an outcome.


Susan Jick - Chief Investigator - BCDSP - Boston Collaborative Drug Surveillance Program
Stephan Gut - Corresponding Applicant - University of Basel
Christoph Meier - Collaborator - University of Basel
Jan Gaertner - Collaborator - Palliative Care Center Hildegard
Julia Spoendlin - Collaborator - University of Basel
Marlene Rauch - Collaborator - University of Basel


HES Admitted Patient Care