AFP interview Flashcards
Absolute risk
Probability individuals will experience the outcome during a specific period
Absolute risk reduction
Absolute difference between two cohorts
Bias
An underlying factor that produces a systematic change in results of study
Confidence interval
The range in which the true result is likely to exist
Confounder
Variable which is not the one that you are interested in but which may affect results
Hazard ratio
Hazard is that rate at which events occur and is time specific. Hazard ratio is the comparison hazard in intervention vs control group p(t)/p(c)
Incidence
The number of new cases during a specific time period in a given population
Intention to treat analysis
Patients are analysed in the groups in which they were randomly allocated regardless of the treatment the ultimately received
Negative predictive value
If the test is negative, what are the chances you do not have the disease
Number needed to treat
Number of patients needed to treat to prevent the occurrence of one adverse event
Odds ratio
Odds is a ratio of those who get the event over those that have the event vs those that don’t. The odds ratio of this in the intervention group vs control group.
P value
The likelihood the difference in results are likely to have occurred simply through chance and hence can reject the null hypothesis
Per protocol analysis
Only those who received or completed treatment (or not) are analysed
Positive predictive value
If the test is positive, what are the chances of you having the disease in question
Power
Ability of the study to find a statistically significant difference under set experimental conditions
Prevalence
The baseline risk of a disorder in the population of interest
Relative risk
Risk a proportion i.e the proportion that get the outcome vs not. Relative risk is a ratio of proportions in intervention and control group
Relative risk reduction
Difference between risks of cohorts relative to the control group
Sensitivity
The percentage of people with the disease that will be picked up
Specificity
The percentage of people without the disease that will be correctly labelled as disease free
T-test
Used to determine if there is a significant difference between the means of two groups (ANOVA for >2 groups)
Allocation bias
None random allocation of patients to groups
Confounding bias
When the 2 groups aren’t similar at baseline
Information/observational bias
Measurement error - often results from no blinding
Attrition bias
Unequal patient loss from the groups
