Social and Behavioral Sciences Flashcards
What is lead-time bias?
The time between detection of a disease by a new screen and detection of the disease by usual screens. Might give the impression that patients live longer with the new screening but it is actually just that you know about the disease longer. No actual change in prognosis.
What are the standard deviation correlations to % of a gaussian distribution?
- 1 to +1 = 68
- 2 to +2 = 95
- 3 to +3 = 99.7
What is the difference between ecological studies and cross-sectional surveys?
Both are “snapshots in time” among wide groups of people. Ecological studies look at the POPULATION level and cross-sectional surveys look at INDIVIDUALS.
How do you calculate number needed to treat (NNT)?
NNT = 1/ARR
ARR (absolute risk reduction) = control event rate - treatment event rate
How do you calculate an Odds Ratio?
Odds of developing event in exposed group/Odds of developing event in unexposed group.
^Unlike probability Odds Ratios DO NOT USE the total number of people exposed to calculate odds.
Draw out the 4x4 and calculate (a/b) / (c/d)
In insomnia treatment, what things qualify as “stimulus control”?
The goal of stimulus control therapy is to dissociate the bedroom from any stimulating activities that do not involve sleep and the fear of not sleeping.
Use bed only for sleep and sex
Go to bed only when sleepy
Leave bed when unable to sleep and go to another room
Maintain a fixed wake up time, incl. weekends
With confounding there is usually no significant difference when _______.
The data are stratified
Negative Predictive Value =
Number of true negatives/ total number of negative tests
How do you calculate relative risk reduction (RRR)?
RRR = (ARcontrol - ARtreatment) / ARcontrol
Schizoaffective disorder = ?
Schizophrenia with CONCURRENT manic or major depressive episode
How is schizoid personality disorder characterized?
Aloof and socially detached but no paranoid delusions.
How do you calculate number needed to harm (NNH)?
NNH = 1/Absolute risk INCREASE
