Lectures 1-4 Flashcards
Definition of EBM:
- medicine combining the best current research evidence with clinical expertise and patient values.
The five outcomes of disease:
- death
- disease
- discomfort
- disability
- dissatisfaction
Prevalence =
(total #sick)/(total population)
- how many people are sick
- a single time point
- Prevalence can be used to estimate the likelihood of a diagnosis before any “tests”
Cumulative incidence:
(new cases)/(total population at risk at study start)
- how many people are getting sick (NEW CASES)
- incidence is a rate, it happens over a time interval
Incidence density =
- the incidence rate in a dynamic, changing population in which people are under study and at risk for varying periods of time.
- “person-years” or “person-days”
Incidence equation in steady state:
prevalence / duration
Duration equation in steady state:
prevalence / incidence
Prevalence equation in steady state:
(incidence) X (average duration)
Apparent incidence depends on:
- intensity of effort to identify cases, not just the true underlying incidence.
Epidemic:
- an increase in incidence of a disease in a community or region.
- Outbreak is sometimes used to mean a small epidemic in a limited location.
- An epidemic curve is a plot of the distribution of cases over time.
Pandemic:
- an epidemic that crosses many international boundaries.
Endemic:
- the constant presence of a disease or infectious agent within a given geographic area or population group.
Random samples:
- Every person has an equal probability of being sampled
Probability sample:
- Every person has a known (but not necessarily equal) probability of being selected.
- Can weight the sample towards some low-frequency groups of interest.
“grab” samples:
- convenient
- susceptible to bias
alpha =
- the probability you’ll make a false hit.
- Arbitrarily, p = 0.05
- false hit = type 1 error
- 5% of the time, we’ll make an error
beta =
- probability you won’t find a difference when one actually exists (false miss)
- false miss = type 2 error
- probability of missing a reality
power =
1 - beta
- power of study to pick up a study when it actually does exist
- Low power is a common reason for type II errors.
Low power is a common reason for what types of errors?
type 2
Power can be reduced by:
BLAMSE
- beta too lax (~0.2)
- low prevalence of outcome in population
- alpha too stringent (~ 0.01)
- multiple comparison adjustments
- sample size (small)
- effect size (small)
A confidence interval is:
- the range of values you get for repeated/multiple samples
- 95% CI means that 95% of values fall within 2SD
2 X 2 tables for hypothesis testing:
(alpha, beta, and power)

2 X 2 tables for hypothesis testing:
(type 1 and 2 errors)

2 X 2 tables for hypothesis testing:
(false hits and false misses)
