(276) Dynamic Relationship Among Immediate Release Fentanyl Use and Cancer Incidence: A Multivariate Time-Series Analysis Using Vector Autoregressive (VAR) Models

Saturday, August 27, 2022

8:00 AM – 6:00 PM CEST

Location: Hall C

Publication Number: 950

Presenting Author(s)

RAQUEL GUTIERREZ MACHIN
– Agencia Espanola de Medicamentos y Productos Sanitarios-AEMPS, MADRID, Madrid, Spain

Co-Author(s)

Marta Monreal
– Agencia Espanola de Medicamentos y Productos Sanitarios-AEMPS, Madrid, Madrid, Spain
Diana González
– Spanish Agency of Medicines and Medical Devices, Spain
Alfonso Rodríguez
– Spanish Agency of Medicines and Medical Devices, Spain
Pilar Rayón
– Spanish Agency of Medicines and Medical Devices, Spain
Dolores Montero
– Spanish Agency of Medicines and Medical Devices, Spain
Consuelo Huerta
– Assistant Professor Epidemiology and Public Health, Faculty of Medicine. Department of Public Health. Complutense University of Madrid, Spain

Background: Recent studies have reported a substantial increase in the incidence of immediate release fentanyl (IRF) use in Spain from 2012 to 2017. Modelling IRF use and its dynamic associations with cancer incidence, will allow to put in context this increase in IRF use and future assessments on the effectiveness of risk minimization measures put in place by the regulatory authorities.

Objectives: To estimate the dynamic relationship among immediate release fentanyl use and cancer incidence

Methods: Monthly cumulative incidence for IRF and cancer were extracted from a nationwide electronic healthcare record database (BIFAP). A Vector autorregresive (VAR) model was constructed using the following step procedure:1) split data into training data for modelling and test for validation 2) assessing for time series stationarity; 3) selecting lag-length; 4) building the VAR model; 5)assessing residual autocorrelation; 6) checking stability of the VAR system; 7) evaluating Granger causality; 8)impulse response analysis and forecast error variance decomposition 9)prediction performance with validation data.

Results: The analysis showed a strong linear correlation between IRF and cancer (Pearson correlation coefficient: 0.6 (95% CI: 0.5-0.7). Two VAR models, VAR (2) and VAR (11) were selected and compared. All tests performed for both models satisfied assumptions for stability, predictability and accuracy. Granger causality revealed cancer incidence is a good predictor for IRF use. VAR (2) seemed to be slightly more accurate, according to the RMSE of the test data.

Conclusions: This study demonstrates that a robust and structured VAR modelling approach is useful to estimate dynamics associations involving IRF use and cancer incidence. Such information could be useful to put into context drug use and unsuccessful regulatory interventions, identifying enablers and barriers to implementation.