14. Biostatistics & Epidemiology

Question 1

RR of 0.71 (P=0.001). How can you interpret this?

Answer 1

There is a 0.1% probability of observing the given RR (or more extreme) by chance alone assuming no differences in mortality
The drug decreased the risk of mortality by 29%

Question 2

What type of study retrospectively associates outcomes with specific risk factors.

Answer 2

Case control

Question 3

What study assess exposure and outcome at a single time point?

Answer 3

Cross-sectional

Question 4

What studies analyze exposed and unexposed individuals and either prospectively or retrospectively follow the outcome.

Answer 4

Cohorts
Note: difference bw case control and retrospective cohort studies is the order in which outcomes and risk fators are assessed. Case control determines outcome first and looks for RF, cohorts look at RF and determine outcome.

Question 5

Define RR of 1, >1, <1

Answer 5

RR of 1 means the outcome occurs with equal freq in both groups and that there is no assoc bw the exposure and outcome
RR >1 means the outcome occurs more freq
RR <1 means the outcome occurs less freq

Question 6

What is attributable risk percent? How do you calculate it?

Answer 6

ARP measures the impact of a RF being studied (eg % of colorectal carcinoma in women with high fat consumption that should attribute the dz to diet)
ARP = (RR-1/RR)
RR = relative risk
or ARP = (risk in exposed - risk in unexposed)/risk in exposed

Question 7

What study would be best for determining incidence of a disease?

Answer 7

Cohort study - can calculate RR

Question 8

Loss to follow-up in prospective studies creates a potential for what type of bias? This can overestimate or underestimate the assoc bw exposure and disease

Answer 8

Attrition bias (subtype of selection bias)

Question 9

If the negative predictive value is 96%, what is the probability that the patient has the condition if the test is negative?

Answer 9

4%

Question 10

What do hazard ratios <1 vs >1 indicate?

Answer 10

<1: tx group has a lower event rate than nontx group

>1: tx group has a higher event rate

Question 11

What type of bias occurs when the treatment regimen selected for a patient depends on the severity of the patient’s condition?
What is intention-to-tx analysis?

Answer 11

Susceptibility bias (form of selection bias). Confounding by indication. 
AVoid by randomly assigning tx to minimize confounding. 
ITT: compares initial randomized tx groups regardless of eventual tx

Question 12

The tendency of the study population to affect the outcome since they are aware they are being studied?

Answer 12

Hawthorne effect

Question 13

T/F: PPV and NPV depends on prevalence

Answer 13

True

Note: sensitivity and specificity do not depend on prevalence