Behavioral Science Flashcards
What is phase I?
Is it safe?
What is phase II?
Does it work?
What is phase III?
Is it good or better?
What is phase IV?
Can it stay?
What is incidence?
of new cases/ population at risk
What is prevalence?
of existing cases/population at risk
What is Relative risk reduction?
1-RR
What is precision?
Same as reliability
What is accuracy?
Same validity. Measured against a gold standard
What are the three major selection biases?
Berkinson: looking only at inpatients
Loss to follow up
Healthy worker and voluntary bias: study pop healthier than general population
What is hawthorne effect?
Example of measurement bias. Groups behave differently because they know they’re being studied
What is procedure bias?
Subjects in different groups not treated the same
What are cross over studies for?
To reduce confounding bias
How to reduce lead time bias?
Measure back end survival: adjust survival according to severity of disease time of diagnosis
What is alpha?
Type I error
What is beta?
Type II error
How is power increased?
Increased sample size, expected effect size, precision of measurement
What increases statistical power?
Meta analysis
How many months of life expectancy must you have for hospice care?
What is a cross sectional study?
Measuring exposure and outcome simultaneously (“snapshot”)
What should be considered when evaluating a screening test?
Lead time bias (time for diagnosis to death seems longer because it was diagnosed earlier than normally would’ve)
What is attrition bias?
Form of selection bias: loss to follow up
What is effect modification?
Stratifying the data will show a significant difference?
What is a confounder (vs effect modification)?
Stratifying the data will not show significant difference
What is reflection?
Summarizing what the patient said
What is facilitation?
Asking patient to tell you more
What is number needed to teat?
1/ARR
What is number needed to harm?
1/AR
What is absolute risk reduction?
ARR = mortality rate in control - mortality rate in intervention
What is the equation for confidence interval?
95% CI = +/- 1.96 (SD)/ sqrt (n)
99% CI = +/-2.58(SD)/sqrt (N)
What does it mean if CI crosses 0?
No significant difference and null cannot be rejected
What does it mean if CI between two groups do not overlap?
Significant difference exists
What is the pearson correlation coefficient (r)?
Always between -1 and +1. The closer absolute value of r is to 1, the stronger the linear correlation between two variable. Positive r means positive correlation. Coefficient of determination = r^2
What is beneficence?
Acting in the patient’s best interest (may conflict with autonomy since they have the right to refuse treatment)
What is nonmaleficence?
Do no harm. Proceed with something if benefits outweigh risk.
When is a minor emancipated?
In the military, married, or self-supporting
What is apgar based on?
A: appearance P: pulse G: grimace A: activity R: respiration
What is a good apgar score?
> 7; 4=6 assist and stimulate,
What is presbycusis?
High frequency hearing loss due to destruction of hair cells at cochlear base
Pincer grasp
By 10 mos
Cubes stacked
number = age x 3
Feed self with fork and spoon
~ 2 years
Parallel play
2-3 years
Cooperative play
4 years
1000 words
3 years (remember 3 zeroes)
Copies line or circle, stick figure
4 years
Tricycle
3 years
Hops on one foot
4 years
Uses buttons or zippers, grooms self
5 years
Social smile
2 months
Stranger and separation anxiety
6 and 9 mo respectively
What is RR?
(a/a+b)/(c/c+d)
What can minors consent to?
Prenatal care, STD, contraception, drug/alcohol addiction
What is odds ratio?
ad/bc
If a disease is bad what type of test do you want?
High sensitivity
What is a cross over study?
Patients randomly allocated to sequence of two or more treatments given consecutively with washout in between.
What is latent period?
Initial steps in pathogenesis and/or risk modifiers occurs years before disease clinically manifests; sometimes risk modifier exposure must be continuous to affect clinical outcome
What is an ROC curve (receiver operating characteristic)?
Produced by plotting sensitivity (TP rate) against 1 - specificity (FP rate). The area under the curve represents accuracy of the test. Want the most area under the curve.
What is the difference between randomized clinical trial and open label clinical trial?
In randomized they don’t know what drug they are getting and in open label they do know (it can also be randomized)
What is a case series?
Type of medical research study that tracks subjects with a known exposure, such as patients who have received a similar treatment, or examines their medical records for exposure and outcome
What is correlation coefficient?
Correlation between two variables. The closer it is to 1 the stronger the association.
REMEMBER IT IS NOT THE SLOPE