(094) A first step in drug utilisation and safety studies: identifying indications for drug use in administrative data sources. A pilot study with gabapentinoids in a French healthcare data source.
A contribution from the ConcePTION project
Postdoctoral researcher Toulouse University Hospital, CERPOP-SPHERE Team, Inserm UMR 1295, Toulouse University, Toulouse, France Toulouse, France
Background: For drug safety studies in pregnancy, it is important to distinguish the effect of drug from the effect of maternal disease. This is because maternal conditions may result in different risks to the pregnancy and the fetus/newborn, independent of the medications dispensed. However, prescribing and dispensing data do not necessarily contain structured information on the indication for administration. This is problematic for drugs with multiple indications, such as antiepileptic agents and analgesics.
Objectives: To describe an algorithm for disentangling the different indications of gabapentinoids (pregabalin/gabapentin), approved in Europe for epilepsy, neuropathic pain, and generalized anxiety disorder.
Methods: Design. Cohort study. Setting. The algorithm was developed in seven European healthcare data sources covering six European countries. This pilot study used data from EFEMERIS (a French cohort of pregnant women) from 2006 to 2019. Analyses. The algorithm identified data components used as markers of prescribing: indication, prescriber specialty, primary care and/or specialized care diagnoses, procedures, reimbursement status, and prescribing/dispensing data. Each component estimated one, several or no reason to prescribe. Results from all components were then aggregated to estimate the indication for each pregnancy. Sensitivity analyses explored alternative assessment windows, diagnostic code values and alternative uses such as off-label use.
Results: We present results on 238 pregnancies in which women received 461 gabapentinoid dispensing before and/or during pregnancy. We had 37 gabapentin-exposed pregnancies and 204 pregabalin-exposed pregnancies. The algorithm indicated that the reason for prescribing was pain in 34% of pregnancies, anxiety in 7%, and epilepsy in 2 %. Multiple indications were identified in 18% of pregnancies, with 80% for pain and anxiety. However, in almost 39% of pregnancies, no reason for prescribing the drug could be detected. The results were similar for pregnancies exposed to gabapentin and pregabalin.
Conclusions: The results demonstrate the possibility to use of the richness of large administrative healthcare data sources to improve evidence on reasons for prescribing. It will be essential to consider these in estimating drug use and safety in the context of maternal illness. Our algorithm will be tested in six other large population data sources in the coming months.
