STATISTICAL METHODS FOR PHARMACOVIGILANCE

A VERY SIMPLIFIED VERSION OF STATISTICAL METHODS FOR PHARMACOVIGILANCE
Prepared for M.Pharm Pharmacology- Clinical Research and Pharmacovigilance

1. Statistical Methods for Pharmacovigilance Sanju Kaladharan 3/16/2019 SANJU KALADHARAN

2. Signal Reported information on a possible causal relationship between an adverse event and a drug, the relationship being unknown or incompletely documented previously. Usually more than a single report is required to generate a signal, depending on the seriousness of the event and the quality of the information. 3/16/2019 SANJU KALADHARAN

3. Disproportionality Though not a definition, it is the case that disproportionality is an issue: reporting of events that are statistically disproportionate (i.e.an O/E ratio exceeding a specified threshold), is a signal of disproportionate reporting.  3/16/2019 SANJU KALADHARAN

4. 3/16/2019 SANJU KALADHARAN

5. 3/16/2019 SANJU KALADHARAN

6. 3/16/2019 SANJU KALADHARAN

7. 3/16/2019 SANJU KALADHARAN

8. 3/16/2019 SANJU KALADHARAN

9. 3/16/2019 SANJU KALADHARAN

10. 3/16/2019 SANJU KALADHARAN

11. Confidence Interval A confidence interval is an interval that will contain a population parameter a specified proportion of the time. The confidence interval can take any number of probabilities, with the most common being 95% or 99%. 3/16/2019 SANJU KALADHARAN

12. Information Component where Pr(R|D) is the posterior probability of observing a specific adverse event R given that a specific drug D is the suspect drug. Pr(R) and Pr(D) are prior probabilities of observing R and D in the entire database. Pr(R,D ) is joint probability that both R and D were observed in the same database coincidentally 3/16/2019 SANJU KALADHARAN

13. 3/16/2019 SANJU KALADHARAN

14. 3/16/2019 SANJU KALADHARAN

15. 3/16/2019 SANJU KALADHARAN

16. 3/16/2019 SANJU KALADHARAN

17. Multi-item-association analysis methods Multi-item gamma poisson shrinker (MGPS) Bayesian confidence propagation neural network (BCPNN) 3/16/2019 SANJU KALADHARAN

18. 3/16/2019 SANJU KALADHARAN

19. BAYESIAN CONFIDENCE PROPAGATION NEURAl NETWORK (BCPNN) The Uppsala Monitoring Centre (UMC) for WHO databases uses BCPNN architecture for SD Neural networks are highly organized & efficient Give simple probabilistic interpretation of network weights Analogous to a living neuron with its multiple dendrites and single axon BCPNN calculates cell counts for all potential R-D combinations in the database, not just those appearing in at least one report Done with two fully interconnected layers One for all drugs and one for all adverse events 3/16/2019 SANJU KALADHARAN

20. THANK YOU 3/16/2019 SANJU KALADHARAN