Statistics

Question 1

What is the p-value tell us?

Answer 1

Type of assessment of error

Tells us the probability that the association is due to chance

Statistically Significant if P-value <0.05 = low probability that association is due to chance

REMEMBER: Don’t rely SOLELY ON P-value ALSO LOOK AT 95% CI

Question 2

What is the confidence interval and what does it tell us?

Answer 2

Interval around the point of estimate

95% CI → if study were repeated → 95% of the 95% CI will contain the true value (the new 95% CI would NOT contain the true value ~1/20 times)

90% CI and lower = contain the true value ~9/10 times etc

REMEMBER: Don’t rely SOLELY ON P-value ALSO LOOK AT 95% CI

Question 3

Interpret this: Relative Risk: 1.5 (95CI: 1.2, 1.9)

Answer 3

True RR has a 95% probability (high likelihood) of being between 1.2 and 1.9

Since 95% CI does not include a RR of 1 (RR = 1 → means no relationship → same risk in both exposed and non-exposed groups) → the RR is regarded statistically significant

True value like to be in the middle of CI and less likely near ends → think normal distribution curve

Question 4

what does it mean if the CI includes 1?

Answer 4

Having a 95% CI of RR / OR which includes 1 → statistically insignificant

As there is a chance the RR or OR is 1 –> as it’s in the 95% CI

and having a RR or OR = 1 - means there is no difference in risk between the exposed and non-exposed for developing the outcome

Question 5

What is prevalence?

Answer 5

The percent of a certain health outcome present in a population

= Number of cases / total population * 100

Question 6

What is incidence? and its benefit vs prevalence

Answer 6

Percentage of NEW cases of an outcome over a period of time

= number of NEW CASES / Person-years

Helps understand if epidemic is getting better/worse