Use LEFT and RIGHT arrow keys to navigate between flashcards;
Use UP and DOWN arrow keys to flip the card;
H to show hint;
A reads text to speech;
65 Cards in this Set
- Front
- Back
Categorical vs Continuous Date types? (what are the sub categories?)
|
Categorical: Nominal (hair color) and Ordinal (pain scale, high med., low)
Continuous: Interval (rank/order, temp, but no true zero), Ratio (has true zero, eg mass in kg) |
|
What is dichotomos data?
|
two mutually exclusive possibilities
|
|
what is discrete data?
|
integers only
|
|
When use chi test? (ie what data type?)
|
categorical - hair color, etc
|
|
When use t-test?
|
with continuous data (blood pressure levels, etc)
|
|
This is calculated as c/(a+c)
|
False-negative error rate
|
|
This is the ability of a test to detect a disease when it is present
|
Sensitivity
|
|
This defines normal and abnormal test results
|
cutoff point
|
|
this is the tendency of a measure to be correct on average
|
accuracy
|
|
this is calculated as a/(a+c)
|
sensitivity
|
|
this is the ability of a test to exclude a disease when it is absent
|
specificity
|
|
this is a nondifferential error
|
random error
|
|
the closer this is to the upper left corner of an ROC curve, the better it is
|
cutoff point.
|
|
This is an analytic study in which the outcomes were deterimed prior to the exposure being determined
|
Case-control
|
|
this is an analystical study in which the outcomes were determined some time after the exposure/intervention was determined.
|
cohort/follow-up
|
|
an analytical study in which the population is surveyed at a single point in time for risk factors and disease outcomes
|
cross-sectional
|
|
an analytical study in which the intervention is randomly allocated
|
randomized control trial
|
|
What measure of association do you use in case-control studies?
|
Odds Ratio
|
|
What is performance bias?
|
It is systematic differences in care the provided to intervention and control group.
|
|
What is co-intervention?
|
When members of comparison group change habits that change risk of outcome.
|
|
What is effect size?
|
The difference in outcomes observed between treatment/control groups.
|
|
What is the difference between reliability and validity?
|
Reliablity is the degree to which measurements are reproducible. Validity is the degree to which results = true values.
|
|
What is internal validity? External?
|
Internal = the freedom from bias, random error, confounding variables
External = the generalizability of the measurements (relevance/applicability) |
|
What is retrieval bias?
|
Investigator conducting review selects studies that support hypothesis (or are otherwise biased)
|
|
What is reporting bias?
|
Investigator of original study only report data that supports view (e.g. drug sponsored). The tendency to under-report unexpected or undesirable experimental results while being more trusting of expected/desirable results.
|
|
What is publication bias?
|
Publishing only studies with statistically significant results make it to the journals.
|
|
Can confounding be transmitted into a meta analysis?
|
Yes
|
|
What's up with the funnel plot? Why does it narrow at the top?
|
Bc the y-axis is the size of the study and as you go up, variance decreases as you improve statistical power.
|
|
What is systematic error?
|
When variations in measurement are predictable/biased (i.e. there is consistent variablility between 'true' and 'study' samples). Results in reduced accuracy.
|
|
What is random error?
|
It is 'noise' that reduces precision of measurements (fluctuations due to sampling variability around a true value).
|
|
What is measurement bias?
|
When analysis of outcomes is influenced by knowledge of treatment group
|
|
What is confounding?
|
It is the influence of variables that may be responsible for association/effect.
|
|
What is the Hawthorne Effect?
|
When people act differently when they know they are being watched.
|
|
What is allocation bias?
|
When investigators choose non-random method of assignment
|
|
What is recall bias?
|
people are more likely to think/recall previous risk factors associated with adverse event than those who are unaffected
|
|
What is healthy use bias?
|
When pts who adhere to preventive therapies may be more likely to engage in a broad spectrum of healthy lifestyle - not representative of general population.
|
|
How is Kappa Coefficient used to reduce error?
|
It's an estimate of inter-rater reliability
|
|
What is the definition of risk?
|
The proportion of persons who are unaffected at beginning of a study period but undergo risk event during study period (proportion affected of susceptible known population over time)
|
|
What are two examples of probability?
|
Inicidence (risk) and prevalence
|
|
What is the equation for odds?
|
p/(1-p)
The probablity of event divided by probability of no event. |
|
What is an example of a study that uses odds ratios?
|
Case-control
|
|
What is the equation for risk ratio (relative risk)?
|
(incidence in exposed)/(incidence in unexposed)
|
|
What is the equation for odds ratio?
|
(odds of exposure in cases)/(odds of exposure in controls)
|
|
Set up a table and calculate risk ratios and odds ratios
|
Now
|
|
How do you calculate relative risk difference (RRD)?
|
(EER - CER)/(CER)
|
|
How do you calculate Absolute Risk Difference?
|
CER-EER
|
|
How does interval data differ from ratio data?
|
Interval data does not have a true zero - e.g. celsius
Ratio data has a true zero - e.g. kilograms. |
|
How do you calculate Beta-error? What is Beta usually set at?
|
It's power; Simply, Power = 1-Beta. Usually set at 0.2 or 20%
|
|
When calculating specificity and sensitivity, is it on left or right column?
|
1. Sensitivity is on the left
2. Specificity is on the rt |
|
Set up a table and calculate sensitivity and specificity
|
do it
|
|
What is the definition of sensitivity? How is it used in screening methods?
|
Its the probability that a person with disease will have a positive result. Used to screen IN diseases with low prevalence.
|
|
What is the definition of specificity? How is it used in screening methods?
|
It's the probability that a person without disease will have a negative result. Used for confirmatory test after a positive screen test.
|
|
What is the difference between a screening test and a diagnostic test?
|
Screening = detect disease in asymptomatic individuals
Diagnostic test = confirm or rule out suspected disease in symptomatic individuals |
|
The likelihood of disease given a positive or negative test depends on:
|
Disease prevalence and the test's properties
|
|
What the definition of a positive likelihood ratio?
|
It is how many times more (or less) likely a positive test result is found in diseased, as compared to non-diseased, individuals:
(True positive rate)/(false positive error rate) |
|
What is the def of a negative likelihood ratio?
|
How many times more (or less) likely a negative test result is found in non-diseased, as compared to diseased, individuals (false neg error rate)/(true neg rate)
|
|
What is the def of PPV?
|
Positive predictive value is the probability of disease given a positive test resutl
|
|
What is NPV?
|
The negative predictive value is the probability of no disease given a negative test result.
|
|
With respect to PPV and NPV, as prevalence goes up, does PPV and NPV go up or down?
|
PPV goes up and NPV goes down
|
|
What is primary prevention?
|
Measures/activities to prevent disease from occurring
|
|
What is secondary prevention?
|
Early interventions/screens of disease to prevent progression or reduce disability
|
|
What is tertiary prevention?
|
Treatment/rehabilitation once disease has occurred to minimize sequelae or recurrence.
|
|
What is spectrum bias?
|
Study population has higher disease prevalence and/or severity than typical population
|
|
What is lead time bias?
|
Detecting disease earlier appears to increase duration of survival fictitiously.
|
|
What is length time bias?
|
Screening more likely to catch case that progresses slowly from disease onset to symptoms.
|