Pharmaco-epidemiology Flashcards
Pharmaco-epidemiology
- is the study of use and effect of drugs in a large number of people
- is the application of principles of epidemiology to drug effects and drug use
- comprises of two components: ‘pharmaco’ (meaning drug) and ‘epidemiology’ (meaning the study of distribution and determinants of disease in a population.
-examines the relationship between drug exposure and health outcome in a defined population
Post- marketing surveillance
- refers to the period after drug enters the market
- Adverse Event Reporting Systems
- Assists with pharmacovigilance (i.e., identification, assessment & prevention of adverse drug effects in medication)
- Multiple reports = more pharmaco-epidemiological investigations
- Label changes or even withdrawal from the market
Drug approval process
preclinical
phase 1: Safety
phase 2: efficacy
phase 3: dosing/ population studies
post-marketing
PE through drug development and post-marketing, however most PE studies focus on the period after the drug enters the market
Drug Recalls
Most effective way to protect the public from a defective or potentially harmful product.
Recall can be voluntary (i.e., made by the manufacturing company) or by FDA request.
355 entries on the FDA website from 2017-2022
Drug Recall Classification
class I: a dangerous or defective product that could cause serious health problems or death
class II: a product that might cause a temporary health problem, or pose slight threat of a serious nature.
class iii: a products that is unlikely to cause any adverse health reaction but that violates FDA labelling or manufacturing laws
Clinical trial limitations
- Certain groups like children and pregnant women may be excluded from clinical trials
- Phase 3 clinical trials only have a few 1000’s of people enrolled making it difficult to detect some adverse events that may be rare
- Short duration of clinical trials
Reasons for conducting pharmaco-epidemiology studies
Clinical: hypothesis generating
Regulatory: to obtain approval for marketing and/or as a response to question by regulatory agency
Marketing: if a study illustrates the adverse effects of a drug, it is easier to market.
Legal: to protect from future liabilities regarding the drug.
Association and Causation
- Pharmaco-epidemiology methods are used to evaluate causal relationships between exposure and outcome
- Describing drug use, identifying associations or relationships with drug use and determining causal relationships
- Associations: is there an association between the medication and outcome of interest?
- Causal relationships: did the medication cause the outcome of interest?
Sources of pharmaco-epidemiology data
- spontaneous reporting of adverse events
- global drug surveillance
- prescription event monitoring
- automated databases
Spontaneous reporting of adverse events
New or rare adverse events
- Rare or new AEs may be highlighted that may not have been discovered during clinical trials
Hypothesis generating
- To explore possible explanations for the adverse event in question
Limitation
- No control group to compare
- Factors other than the drug may be contributing to the outcome
Global drug surveillance
Combination of:
-set definitions
- homogenous way to store data
- homogeneous terms to describe AEs
Prescription event monitoring
A cohort of users of a particular drug is defined from their prescriptions and followed up for a predefined period of time, with the purpose of identifying all side effects associated with the drug of interest.
Types of automated databases
Medical record databases
Claims databases
Medical record databases
- Longitudinal, anonymised patient record databases that are used by health care providers for routine patient care
- Wealth of info for research as lifetime’s accumulation of health care information on patients
- growing use to assess un-intended and intended drug effects
Medical record databases Pros and Cons
Pros:
- high validity
- test results
- lifestyle information may be available
Cons:
- sometimes incomplete
- various coding systems
-prescribed information available, not dispensed
Claims databases
- contain medical, pharmacy and dental info of patients who have made a claim for their treatment
- info such as age, sex, race, income and mortality
eg. prescribed or dispenses (filled) prescriptions
Desired qualities of a database
- Representative
- Large
- Timely (i.e., up to date)
- Continuity: individual observations, calendar time
- Linkage on unique identifier
- Access
Countries with high quality data sources: UK, Sweden, Belgium, The Netherlands
Application of the studies
Estimate of drug use risks
- PE studies can be used to estimate the risk of drug use. For e.g., when there were case reports of triazolam (anti-depressant) induced psychiatric disturbances, it was removed from the market. Issue averted by recommending a lower dose.
Aid public health policy decisions
- Regulatory agencies may impose restrictions on specific drugs based on the results of PE studies, including assessment on whether a drug should be withdrawn
Aid therapeutic guidelines and discovery of new indications for the drug
- In addition to examining the effectiveness of drugs in elderly/patients with comorbidities, PE studies may also help discover new indications of an already marketed drug
Pharmacovigilance
- aims to improve patient care and safety by collecting and managing data on the safety of medicines.
- source of data for PE studies.
Three measures of therapeutic effects
Absolute risk reduction (ARR)
Relative risk reduction (RRR)
Number needed to treat (NNT)
Absolute Risk Reduction aka absolute risk difference (ARR)
Is the simplest measure of therapeutic effect.
Defined as the absolute value of difference in the event rates between exposed and unexposed
ARR = (Re - Ru)
Relative Risk Reduction (RRR)
Measures the extent an exposure (therapy) reduces a risk in comparison with individuals in the unexposed group.
Example:
exposure vs no exposure
treatment vs no treatment
RRR= Ru - (Re / Ru)
Number Needed to Threat (NNT)
- number of individuals who would have to receive the treatment for one of them them to benefit from the treatment over a specified period of time.
Reciprocal of ARR: 1/ARR
NNT= 1/ (Re - Ru)
Medication adherence
The term adherence is used to describe a patient’s failure to consume medication according to the prescriber’s direction.
Studies looking into patterns of adherence usually report patients’ medication-taking behaviour as a dichotomous outcome (i.e. adherent vs non-adherent).
Calculating Adherence
MPR = (Sum of days’ supply for all fills in period) / Number of days in period) * 100%
Most studies report adherence to be > 80%
Ethical Considerations
- privacy
- confidentiality
- data sharing
- informed consent
