Statistical Signal Process in R Language in the Pharmacovigilance Programme of India

Abstract

Background:

The Ministry of Health & Family Welfare, Government of India, initiated the Pharmacovigilance Programme of India (PvPI) in July 2010. The purpose of the PvPI is to collect data on adverse reactions due to medications, analyze it, and use the reference to recommend informed regulatory intervention, besides communicating the risk to health care professionals and the public. The goal of the present study was to apply statistical tools to find the relationship between drugs and ADRs for signal detection by R programming.

Methods:

Four statistical parameters were proposed for quantitative signal detection. These 4 parameters are IC₀₂₅, PRR and PRR_lb, chi-square, and N₁₁; we calculated these 4 values using R programming. We analyzed 78,983 drug-ADR combinations, and the total count of drug-ADR combination was 4,20,060. During the calculation of the statistical parameter, we use 3 variables: (1) N₁₁ (number of counts), (2) N_1. (Drug margin), and (3) N_.1 (ADR margin).

Results:

The structure and calculation of these 4 statistical parameters in R language are easily understandable. On the basis of the IC value (IC value >0), out of the 78,983 drug-ADR combination (drug-ADR combination), we found the 8,667 combinations to be significantly associated.

Conclusions:

The calculation of statistical parameters in R language is time saving and allows to easily identify new signals in the Indian ICSR (Individual Case Safety Reports) database.