Intro To Pharmacoepidemiology

Question 1

Define pharmacoepidemiology

Answer 1

The study of the USE and EFFECT of drugs in large populations

Question 2

How is pharmacoepidemiology different from clinical pharmacology?

Answer 2

Clinical pharmacology is the study of drugs in INDIVIDUALS
- it states that therapy should be individualized and tailored to the needs of the individuals

Meanwhile, pharmacoepi deals with drugs in POPULATIONS

Question 3

What is the difference between epidemiology and pharmacoepidemiology?

Answer 3

Epidemiology studies the DETERMINANTS and DISTRIBUTION of a disease (not drug effects on the population with the disease!)

Question 4

Epidemiology (not pharmacoepi) is separated into two basic areas:

Answer 4

Epidemics = study of infectious disease outbreaks
Chronic disease epidemiology

Question 5

Pharmacoepidemiology applies the ________ of epidemiology to the _________ of clinical pharmacology

Answer 5

Pharmacoepidemiology applies the METHODS of epidemiology to the CONTENT of clinical pharmacology

Question 6

What legislation was made in response to excessive adulteration/misbranding of food and drugs?

Answer 6

Pure Food and Drug Act, 1906

Question 7

What legislation was formed after 100+ people died of renal failure from the marketing of a sulfanilamide elixir in diethyl glycol?

Answer 7

Food, Drug and Cosmetic Act, 1938

Question 8

The Food, Drug and Cosmetic Act required manufacturers to do what to release drugs to the public?

Answer 8

Preclinical toxicity testing and continuously gathering clinical data
(BUT no proof of efficacy required yet!)

Question 9

What are some problems with randomized clinical trials?

Answer 9

• expensive
• small population (usually)
• drugs compared to placebo
• exclude certain populations at risk (elderly, pregnant, comorbidities)
• may be unethical
• not timely

Question 10

What legislation required manufacturers to submit post-marketing research at approval of a new drug?

Answer 10

FDA Amendment Act, 2007

Question 11

What is a case report?

Answer 11

A report of an event involving one single patient
Simple and inexpensive, usually for extremely rare events

Question 12

What is a case series?

Answer 12

Collections of patients whom all had a single exposure/single outcome
No control group = no tested hypothesis tested!

Question 13

What is an analysis of secular trends? What does it lack?

Answer 13

Examine trends in an exposure that is the presumed cause with trends in the presumed effect in a disease and compare the correlations.

Lacks individual data! (Only compares trends)

Question 14

What is a case-control study? What is its measure of association?

Answer 14

Compares cases’ outcomes to control group outcomes to look for differences in source exposures.
High potential for bias

Measure of association = odds ratio

Question 15

What is a cohort study? What are its measures of association?

Answer 15

Identifies a cohort of subjects and follows them over time to determine outcomes. Involved an exposure at the beginning with some subjects being exposed and some not. Loss of follow-up is a concern.

Measures of association = risk ratio and attributable risk

Question 16

What is a randomized control/clinical trial?

Answer 16

Studies where investigators control the exposure received by each patient and subjects are randomly assigned to control or exposure groups.

Question 17

What is a pragmatic clinical trial?

Answer 17

Studies where the investigator tests the effectiveness of an intervention under “real world” conditions.
More inclusive, but with so much intersubject variability it is harder to draw conclusions

Question 18

What is the Evidence Paradox?

Answer 18

Even though thousands of RCTs are published yearly, there is not enough evidence to effectively inform clinical decision-making.
We need real-life patients (comorbid, non-adherent etc)

Question 19

What types of studies use Odds ratios and Relative risk/Risk ratios?

Answer 19

ORs are based on prevalent data, used in case-control and cross sectional trials
(Odds of outcome in exposed/odds of outcome in control)

RA’s are based on incidence data, used in cohort and RCTs
(Risk in exposed/risk in non-exposed)

Question 20

When is the null hypothesis significant?

Answer 20

When the null outcome is NOT included in the confidence interval. (When the CI is too wide, it decreases credibility of the results)

Question 21

What are some examples of study result categories? (4)

Answer 21

1. No association
2. Artificial association (chance, biased)
3. Indirect association (confounded)
4. Causal association (true association!)

Question 22

What are the two types of bias errors?

Answer 22

1. Information bias:
   - Interviewer probes some subjects more thoroughly than others OR
   - recall bias
2. Selection bias:
   - controls don’t represent the actual population makeup OR
   - the makeup of the group gets skewed by drop-outs

Question 23

What are confounders?

Answer 23

Variable related to both the exposure and outcome. Confounding occurs when these variables are distributed unequally between groups
Ie. Observing PPI effects on kidney disease —> age affects both of these [fewer elderly on PPIs, but many more elderly have CKD]