Estimating Risk I & II Flashcards
Relative Risk // Rate Ratio
- the STRENGTH of the relationship between exposure and outcome
- the probability of an outcome for exposed group and unexposed group
When the relative risk (rr) is 1.0
- no difference between the exposed and unexposed groups
when the relative risk (rr) is > 1
- the exposure “increases” risk – it’s possibly a risk factor
when the relative risk (rr) is < 1
- the exposure variable “decreases” risk – it’s possibly a protective factor
Risk Difference
- that is the excess risk than can be attributed to having had the exposure
Odds Ratio
- the ratio of the probability of occurrence of an event to that of non-occurrence
- tells us how much higher the odds of exposure are among case-patients than among controls
What kinds of studies can odds ratio be used for?
- cohort
- cross-sectional
- case-control
What kinds of studies can relative risk be used for?
- only cohort studies
When odds ratio is < 1.0
- cases are less likely to be exposed than controls
- protective factor
When odds ratio equals 1.0
- odds of exposure are the same in the cases and controls,
- exposure has no influence whatsoever on the disease outcome
When odds ratio is > 1.0
- cases are more likely to be exposed than the controls
- risk factor
Point Estimate
Confidence Intervals
- “best guess” number based on analysis of the data
- CI = data around the calculated odds ratio
Statistical Significance
- is determined using statistical tests such as the Х2 (chi-square) test.
P-value
- the probability that a finding could have occurred by chance alone – of less than 0.05
Sample size influence confidence intervals by…
- yielding narrower confidence intervals to reveal statistically significant associations
Attributable Risk
- tells how much of the health burden due to a specific disease can be alleviated through the elimination of a risk factor for that disease
- tells us the potential for prevention
Target Population
- those most susceptible to the illness
Etiologic fraction
- attributable risk in the exposed, in the hope that the excess risk characterized by that number will MOTIVATE BEHAVIOURAL CHANGES.
Population Attributable Risk (PAR) = Population Etiologic Fraction
- the benefit of investing in things like prevention should be HIGH for the entire population for it to command political attention
Absolute Risk
- the incidence of a disease in a population
- is implicit, but doesn’t tell us any associations of risk
Rarity Assumption
- when the disease being studied does not occur frequently