Statistics, Study Design, Public Health

Question 1

Which value represents a study’s ability to detect a difference when one exists i.e., the probability of finding a true relationship?

Answer 1

Statistical power, (1 - beta)

Question 2

What is the 68-95-99.7 rule for the percentage of values that lie within a band around the mean in a normal distribution?

Answer 2

• 68% of all observation lie within 1 SD of the mean, with half above and half below
• 95% of all obervations lie within 2 SD of the mean
• 99.7% of all observations lie within 3 SD from the mean

Question 3

An arrangement in which a payor (individual, employer, or government entity) pays a fixed, predetermined fee to cover all the medical services required by a patient is termed what?

Answer 3

Capitation; payment structure underling HMO provider networks

Question 4

When an insurer pays a provider a single payment to cover all the expenses associated with an incident of care i.e., elective surgery; this is known as what?

Answer 4

Global payment

Question 5

How do you calculate attributable risk percent in the exposed?

Answer 5

• ARP_exposed = [(risk in exposed - risk in unexposed) / (risk in exposed)] x 100
• ARP_exposed = [(RR - 1)/(RR)] x 100

Question 6

Matching is a method generally used in the design stage of case-control studies in order to control which type of bias?

Answer 6

• Confounding
• Initial step involves selecting variables that could be confounders i.e., age, race, gender, and/or smoking status

Question 7

How can confounding bias vs. effect modification be differentiated in terms of effect on exposure and outcome?

Answer 7

• Confounding: when the exposure-disease relationship is muddled by the effect of a confounding variable, associated with both the exposure and outcome
• Effect modification: external variable positively or negatively impacts the observed effect; only related to outcome