Week 1 Flashcards
sample size can be bigger than population size
if redundant samples are taken
probability/random sampling
simple random sampling, systematic sampling, stratified sampling, cluster random sampling
Non random sampling
snowball sampling, quota sampling, purposive sampling
Proportionate stratified sampling
same proportion from each stratum
Disproportionate stratified sampling
different proportion from different strata
ensures minorities are adequately covered
HALE
Health adjusted life expectancy
DALYs
Disability adjusted life years
Attack rate
Percentage of people who were exposed and then got disease
Relative risk
risk of disease in those who were exposed relative to non exposed
Attack rate exposed and disease / attack rate non exposed and disease
Sufficient cause
factor (or combination of factors ) that will inevitably produce disease
presence of it alone is enough to result in the disease
Component cause
factor that contributes towards disease causation but is not sufficient to cause disease on its own
Necessary cause
any agent (or component cause) required for the development of a given disease disease is never present when the factor is not present
temporality
exposure must precede disease development
strength of association
stronger an association is (as described by the relative effect), the less likely it is to be due solely to either bias or confounding
consistency
higher consistency, more likely to be causal rather than association
dose-response relationship
if a factor does cause a disease, then the risk of developing the disease is likely to be related to the amount or ‘dose’ of exposure; level and duration of exposure
reversibility
if you remove the exposure, you remove the outcome
Bradford hill criteria
plausibility consistency reversibility strength of association temporality dose-response
data transformation can occur only from
numerical to categorical
systematic sampling
divide population size by sample size
round decimal down to give k
select x between 1 and k
x is first number, then every kth number after that
reduced risk/lower odds: none
0.83-1
reduced risk/lower odds: weak
0.67-0.83
reduced risk/lower odds: moderate
0.33-0.67
reduced risk/lower odds: strong
0.10-0.33
increased risk/higher odds: none
1.0-1.2
increased risk/higher odds: weak
1.2-1.5
increased risk/higher odds: moderate
1.5-3.0
increased risk/higher odds: strong
3.0-10.0
upstream work
looking at source and how to prevent disease
downstream work
treatment after exposure to disease
absolute requirement- bradford hill criteria
temporality
strength of association described by
relative effect, OR or RR
