Statistics Flashcards
If prevalence of a disease increases what happens to the positive predictive value and negative predictive value of a given test?
PPV increases
NPV decreases
In a low prevalence population what are the characteristics of PPV and NPV
PPV low
NPV high
Formula for relative risk
RR = Incidence (exposed)/Incidence (unexposed)
Odds ratio formulae using the two by two table
OR = AD/BC
A = exposed cases B = exposed non-cases C = not exposed cases D = Not exposed non cases
Attributable risk (reflects the additional incidence of disease related to an exposure taking into account the background rate of disease)
Incidence of disease in exposed persons - incidence of disease in non-exposed persons
Population attributable risk formula
PAR = attributable risk x prevalence of exposure
PAR % = Cases caused by exposure/ All cases
Number needed to treat formula
Reciprocal of the ARR
Absolute risk reduction formula
ARR = AR Control = AR Treatment
AR control = events in control/ total in control group
AR treatment = event in treatment/total in treatment group
Type 1 statistical error
Inappropriately reject the null hypothesis i.e find a difference that doesn’t exist. This is due to the p-value
Type 2 statistical error
The null hypothesis is inappropriately accepted. i.e a difference that existed was not found. This is usually because the study was underpowered
Purpose of an intention to treat analysis
Groups patients according to the treatment they were randomised to. Gives a more realistic (and thus conservative) estimate of the effectiveness of the treatment
Sensitivity definition
Proportion of people who has the disease who test positive
Specificity definition
Proportion of people who do not have the disease who test negative
