EPIDEMIOLOGY - EVERYTHING
Measure of association between an exposure and an outcome.
Calculated as A/B.
Used in Case Control Studies.
The probability of an event occurring in an exposed group vs the probability of it occurring in a non-exposed group.
Calculated as A/(A+B).
Used in Cohort studies.
The proportion of new people with a disease, divided by the total number of people in the beginning of the measurements.
The proportion of people with a disease, divided by the total number of people being measured.
NUMBER NEEDED TO TREAT
The number of patients who need to be treated in order to prevent one additional bad outcome.
Calculated as: NNT=1/ARR
Where, ERc are the control/placebo group, and ERt are the treatment group.
The ability of the test to detect the true positive rate - correctly identify those people with a disease.
Use the table, and go down: TP/(TP+FN)
The ability of the test to correctly identify those without the disease
Use the table, and go up: TN/(TN+FP)
POSITIVE PREDICTED VALUE
The proportion of positive results that are true positives.
Use the table, and go to the right: TP/(TP+FP)
NEGATIVE PREDICTED VALUE
The proportion of negative results that are true negatives.
Use the table, and go to the left: TN/(TN+FN)
The probability of observing a result as least as extreme as the one that was observed, given that the null hypothesis is true.
Used to determine whether there is a statistically significant difference in the means of two groups.
Used to determine whether there is a statistically significant difference between the observed frequencies vs the expected frequencies.
Used for categorial variables.
Immunisation, legally banning hazardous activities and products, education on how to be healthy
Regular exams and screening tests, daily aspirins, exercise, to prevent further heart attacks/strokes, subtly modified jobs so unhealthy individuals can return to work.
Chronic disease management groups, support groups, treatment itself
INTENTION TO TREAT ANALYSIS
Is an analysis of every member in a randomised control trial group, without taking into consideration adherence, non-compliance, protocol deviations, withdrawal. Therefore, the estimate of treatment effect from these studies is conservative.