Unit 3 Flashcards
The difference in risk (incidence) between exposed and unexposed groups:
AR(e) or RD (attributable risk exposed or risk difference)
Attributable risk or risk difference applies to ___________ animals only.
exposed
Ex: For every 100 animals exposed, expect an additional 20 cases.
Risk difference. Over baseline due to exposure only.
The difference in risk (incidence) between the total population and unexposed group
Population attributable risk
This expresses the amount of disease in the total population attributable to exposure.
PAR
The proportion of incidence in exposed that is attributed to the exposure.
Attributable Proportion (exposed)
AP(e) tells you what fraction (%) of disease in exposed is really due to:
exposure
How do you calculate AP(e)?
(RR-1)/RR
The proportion of incidence in population attributed to exposure
PAP (population attributable proportion)
What fraction (%) of disease in the population is due to Exposure?
PAP
How do you calculate PAP (population attributable risk)?
(incidence in population - incidence in E-)/incidence in population
Which two measures of impact reference the exposed?
AR(e)/RD, AP
Which measure of impact references the TOTAL population?
population attributable risk
Which measure of impact references the population?
population attributable fraction
Identify Koch’s postulates:
- it is present in all cases of the disease
- it does not occur as a fortuitous or non-pathogenic parasite
- it is isolated in pure culture from a case
- organism can be is repeatedly passaged, and induced the same disease in other animals
What are the difficulties with Koch;s postulates in defining “cause”? Hint: There are seven.
- multiple etiologic factors
- multiple effects of a single cause
- asymptomatic carriers (cholera)
- non agent factors such as age
- immunologic processes as cause of disease
- host-agent, host-environment interactions
- noninfectious causes of disease
Identify how surgeon general changes the causation:
- consistency
- strength of association
- specificity of outcome, time, place, population
- temporality
- coherence
Identify how hills criteria changes the causation:
surgeon general criteria plus:
- dose-response relationship
- plausible biologic mechanism
- experimental evidence to compliment
Identify how Surgeon general/Hills criteria changes the focus of causation, and how they rely on factors apart from agent description for disease causation.
- acknowledged that clinically healthy individuals frequently tolerate pathogens quite well
- required proof of associated between hypothesized causal factor and disease
- accommodated non-infectious factors
- mandated temporal sequence
- dose-response relationship adds credibility
the proportion (or fraction or probability) of cases of disease within a population
apparent prevalence
Are results of the study applicable to the real population?
external validity
Correct assessment of exposure and outcome in study groups
internal validity
External validity requires __________, but internal validity does not require ______________.
internal validity; external validity
a systematic error that results in an incorrect estimate in the relationship between exposure and disease:
bias
If uncorrected, renders the results worthless:
bias
Concerns the animals who participate and those that do not:
selection bias
assessment of exposures and outcomes:
information bias
To prevent this type of bias, comparisons must be drawn from the same population:
selection
What are the three types of information bias?
observer, recall, misclassification
differences in way information collected from E+ vs E- or D+ vs D-:
observer bias
Those D+ may tend to recall exposures in greater detail than those in D-
recall bias
When either E or D is not properly classified:
misclassification bias
A “mixing” of effects:
confounding
An extraneous variable that can wholly or partly account for an apparent association between an exposure and outcome:
confounding variable (confounder)
correlated with the explanatory and response variable:
confounding variable
restrict study subjects to avoid known confounding factors:
restriction
Match pairs on a confounding variable so that confounding is distributed evenly between comparands:
matching
When all classes, groups, or categories of a variable (whether exposure, outcome, or covariate) have the same error rate or probability of being misclassified for all study subjects.
non-differential misclassification
Occurs when the proportions of subjects misclassified differ between the study groups
differential misclassification
