16 June Epidemiology and Biostats

Question 1

Describe a cross-sectional study

Answer 1

Used to assess prevalence
Collects information at a specific point in time
Information about an entire population to look for risk factors and possible causes
It is a descriptive study and is observational not longitudinal or experimental
Seeks to answer: What is happening?
Can help identify prevalence and odds ratios

Question 2

How is a cross-sectional study different from a case-control and a longitudinal study?

Answer 2

Cross-sectional looks at an entire population whereas a case-control study looks at a specific group of people with a given characteristic
Cross-sectional looks at one data point at one point in time whereas a longitudinal study looks at multiple data points repeatedly over a longer period of time

Question 3

Describe a case control study

Answer 3

Compares people with disease to those without in an effort to determine what happened.
Helps establish odds ratio by looking for prior exposure or risk factor.
From now looking backwards

Question 4

Describe a cohort study

Answer 4

Compares a group with a given exposure or risk factor to a group without these factors to see if factor affects likelihood of developing disease.
Looks for relative risk to determine who will likely get the disease.
From now looking forwards

Question 5

Describe phase I of drug trials

Answer 5

Healthy volunteers establish whether or not the drug is safe and assess pharmacodynamic traits

Question 6

Describe phase II of drug trials

Answer 6

Patients with the disease test whether or not the drug works and what the dosing should be

Question 7

Describe phase III of drug trials

Answer 7

Large study with many patients randomly assigned to groups to assess whether the new treatment is as good or better than the current standard

Question 8

Describe phase IV of drug trials

Answer 8

After the drug is approved for use, it is monitored for long-term and unseen effects to determine whether or not the drug can stay in use.

Question 9

What phase of drug trials assesses the long-term effects of the drug and looks to see if it can stay on the market?

Answer 9

Phase IV

Question 10

What phase of drug trials is on healthy volunteers to see if the drug is safe and to study the pharmacodynamics?

Answer 10

Phase I

Question 11

What phase of drug trials is on a large group of patients assigned to several groups where the drug is compared to the effectiveness of the current standard to see if it is the same or better?

Answer 11

Phase III

Question 12

What phase of drug trials is on a small group of patients with the disease to see if the drug works at all and to establish dosing and look for adverse side effects?

Answer 12

Phase II

Question 13

What kind of observational study looks at patients with a disease to figure out what risk factors they might have in common to help determine what happened?

Answer 13

Case-control study

Question 14

What kind of observational study looks at people with certain risk factors and follows them to see who gets the disease to help establish who will get a disease?

Answer 14

Cohort study

Question 15

What kind of observational study is used to establish relative risk?

Answer 15

Cohort study

Question 16

What kind of observational study is used to establish odds ratio?

Answer 16

Case-control study

Question 17

What are two observational study types used to understand heritability and influence of environmental factors of a disease?

Answer 17

Twin concordance and adoption studies

Question 18

What properties of a test are fixed and which are variable depending on disease prevalence?

Answer 18

Sensitivity and specificity are fixed and PPV and NPV vary depending on disease prevalence

Question 19

How do PPV and NPV vary with pretest probability?

Answer 19

PPV varies directly with pretest probability

NPV varies inversely with pretest probability

Question 20

How does the relationship between prevalence and incidence change based on the duration of a disease?

Answer 20

In a short-lived disease, prevalence and incidence are about equal
In a chronic disease, prevalence is much larger than incidence

Question 21

Define absolute risk reduction

Answer 21

Take the percent of people without an intervention who develop a disease and subtract the percent of people with the intervention who have the disease to find the difference in percentage between the two groups

Question 22

Define number needed to treat

Answer 22

NNT is the inverse of the absolute risk reduction

Question 23

Define attributable risk

Answer 23

The difference between percentages of those exposed who developed disease minus those not exposed who developed the disease

Question 24

Define number needed to harm

Answer 24

The inverse of attributable risk

Question 25

What is confounding bias?

Answer 25

A factor is related to both the exposure and the outcome, but is not on the causal pathway.
Example: coal miners have higher rates of pulmonary disease than general population, but they also have a higher rate of smoking which is the actual cause of the increased prevalence of lung disease.

Question 26

Describe mean, median, mode and the effect outliers have on them

Answer 26

Mean is the average and is most affected by outliers
Median is the middle value and is the average of the two middle values in an even set
Mode is the most common value and is the least affected by outliers

Question 27

What is a type I and type II error in an experiment?

Answer 27

Type I: false-positive result, you thought there was a difference when there really was not one
Type II: false-negative result, you thought there was not a difference when in fact there was one

Question 28

When is a t-test used?

Answer 28

Checks the differences between means of 2 groups

Remember that tea is meant for 2

Question 29

When is an ANOVA used?

Answer 29

Checks differences between means of 3 or more groups

Stands for analysis of variance (has 3 words)

Question 30

When is a Chi-square test used?

Answer 30

Checks differences between 2 or more percentages or proportions of categorical outcomes
Pronounce Chi-tegorical