We use a test with a high level of this early on in the disease to rule diseases out.
What is sensitivity
100
This is the degree to which the information collected accurately answers the question asked.
What is Validity
100
The proportion of patients surviving a defined period of time from a point in the course of the disease
What is survival rate
100
These studies are more rigorous than observational studies
What are experimental studies
100
This helps to reduce selection bias
What is random assignment
200
Tests to rule a disease in have high levels of both of these.
What is Specificity and PPV
200
This is a measure of how much a risk factor contributes to disease incidence at the population level.
What is Population Attributable Risk
200
These are subjects lost to follow up
What are censored subjects
200
If a risk factor increases the likelihood of disease, its removal can be used to prevent disease
What is prevention
200
This helps to reduce information bias
What is blind assessment
300
This method is used when 2 or more different tests are run at the same time and only one test has to be positive. The effect is to increase sensitivity!
What is Parallel testing
300
A precision problem, related to sloppiness, causes these.
What are random errors
300
This type of study uses assigned treatments which are best done randomly and blindly.
What is an experimental study
300
exposure and outcome may be assessed at the same time
What is a cross sectional study
300
These 2 biases are found in the assignment phase.
What is sampling bias and selection bias
400
This method of testing increases specificity
What is serial testing
400
A validity problem, caused by bias, is responsible for these.
What are systematic errors
400
Increasing sample size is the remedy for problems with this.
What is reliability
400
Exposed and unexposed animals are identified and followed for a period of time until the outcome does or does not occur.
What is a cohort study
400
This bias is addressed most often in the analysis phase
What is confounding bias
500
A disease control program is most likely to use this type of testing.
What is serial testing
500
This is the fraction of disease occurrence in a population associated with a particular risk factor.
What is Population Attributable Fraction
500
The most "valuable" of the study designs
What is an experimental design
500
Animals with and without the outcome are chosen and then assessed for whether or not they were previously exposed
What is a case-control study
500
Information bias is addressed most often in this phase.