Drug substitution associated with a hospital stay in Belgium: a retrospective analysis of a claims database

This study measures the extent of drug substitution associated with a hospital stay in Belgium.

Data were extracted from the 2006-2007 dataset of the Belgian Agency of Health Insurance Funds on drug use of patients hospitalized in acute hospitals. Reimbursed drugs received in ambulatory care during the 3 months prior to hospitalization were compared with drugs received during the 3 months following hospital discharge. Both a narrow definition and a broad definition were used for drug substitution. Narrow substitution (switches between generic and originator drugs) was computed for 14 drug classes for chronic conditions with the highest public expenditure. Broad substitution (changes between chemical substances within the drug class at ATC level 4, changes in brand name) was calculated for statins and proton-pump inhibitors only.

The database included 17 764 patients (mean age 66 ± 17 years; 60% female). In 71% of cases an originator drug was received prior to and following hospitalization. A generic drug was received prior to and following hospitalization in 25% of cases. Some form of narrow substitution occurred in 4% of cases: a generic drug was replaced by an originator drug in 2% of cases and an originator drug was replaced by a generic drug in 2% of cases. Some form of broad substitution occurred in 25% of cases for proton-pump inhibitors and 13% of cases for statins.

Hospitalization was not a trigger for changes between originator and generic versions of a drug. Broad substitution associated with a hospital stay was relatively limited for statins and proton-pump inhibitors.