Study sets, textbooks, questions
Upgrade to remove ads
Step 2 Ethics/Social/Biostats
Terms in this set (31)
Almost always, except when doing so poses a threat to public health (like meningococcal meningitis or TB)
When do pts have the right to refuse tx?
Prognosis of less than 6 mos
What is required for hospice care?
When it might physically harm them (like homocidal ideations or HIV). you CAN encourage the pt to discuss his diagnosis with his spouse (i.e. cancer dx... will affect her)
When do you talk to a spouse about a diagnosis?
Can be removed from vent when brain death is determined in most states. Some states have other regulations in place in case the declaration of death is in violation of individual religious beliefs
If a pt is confirmed to be brain dead, and there is no advanced directive or power of attorney... when to take them off ventilator?
If emergent, when treating STIs, when it's a prenatal visit, and regarding substance abuse.
OR is pt is emancipated--homeless, if they are a parent, married, military, financially independent, HS graduate
When do minors not need parental consent?
If it's nonemergent, but fatal (like leukemia). If emergent-->go ahead and treat/stabilize. If not fatal/life threatening--comply with parents' wishes (i.e. routine vaccines)
When to get a court-order to treat a pediatric patient?
Give the shot. It's an urgent case, with high risk of tetanus infection, so you can go ahead and give it. Only one parent (custody) consent is needed
If pt just stepped on a rusty nail and only mom gives consent (not dad) for tetanus shot, what do you do?
Only if permission is obtained and documented from the family, or the patient prior to death. Must be part of a structured training and under close supervision
When can you perform procedures on a recently deceased patient?
Only non monetary gifts of minimal value and MUST directly benefit the patient (e.g. unbiased educational material and drug samples)
what kind of gifts are acceptable from drug companies/third parties?
Find out why--dissatsifaction with the care you're providing? experiencing bad side effects?
When a patient wants alternative therapy, what do you do?
INCREASE sensitivity (catch more positives/fewer false negatives) and decrease specificity (have more false positives)
Typically, lowering the cutoff point for a test will do what to sensitivity and specificity?
Odds ratio is close to the approximation of the relative risk (rare disease assumption)
If an outcome is uncommon in the population, what does that mean for the odds ratio and the relative risk?
Odds Ratios--relative risk can be estimated with Odds ratio IF the disease is RARE--"rare disease assumption"--that means that the disease isn't everywhere, popping up for lots of reasons
Can you calculate relative risk or odds ratio from case-control studies?
NNT = 1/ARR (absolute risk reduction). Look at rates of both groups, calculate difference. That's ARR. Then take the inverse of that. That's NNT. So you say 16% of these pts had an event vs. only 12% of the other group, so that's a 4% difference. Then you go 1/0.04 to get 25-->thats how many NNT to prevent one adverse event
How do you calculate the number needed to treat (NNT) to prevent one adverse event?
AKA external validity. Being able to apply the study to other populations. Internal validity is if the results are valid because you measured what you wanted to
What is generalizability?
Selection bias. Any bias 2/2 inappropriate selection OR poor retention of study subjects (sampling bias, nonresponse biase, bearskin bias (hospitalized pts), prevalence bias)
What type of bias is attrition bias?
Inaccurate measurement or classification of disease, exposure, or other variable (recall bias, observer bias, reporting bias, surveillance/detection bias aka paying more attention to the "exposed" group)
What are some examples of observational bias?
If it doesn't cross the null, it means that p is < 0.05 (5%). If it DOES cross the null, it means that the p is > 0.05. (Same goes for 99% CI that doesn't cross the null--> p is < 0.01
What does a 95% confidence interval mean?
RR >1 means Positive association between risk factor and outcome. RR < 1 means negative association between risk factor and outcome. Farther than 1 in either direction, stronger the association
What does RR > 1 mean? RR <1?
Matching controls for confounding. --if you control for potential confounders (age, race, etc) then you reduce confounding variables
In case-control studies, what issues does matching (cases and controls) address?
Effect modification is when an external variable affects the exposure/risk factor and makes the outcome more likely (e.g., family history makes OCP users more likely to have breast cancer, but doesn't make those without family history more likely) #3947
Effect modification vs. Confouding?
That the test will give similar results on repeat measurements
What does reliability mean?
Exposure Odds Ratio (RR and relative rate are incidence measures, so you report those in cohort studies-->prospective)
What measure of association is used in a case control study?
Subjects are aware that they're being studied so they change their behavior
What is the Hawthorne effect?
A variable that is independently associated with both exposure and outcome (i.e. cigarette smoking is a confounding factor when studying maternal ETOH use and low birth weight babies because it's associated with both alcohol use and low birth weight babies independently) **don't see an effect when you control for smoking
What are confounding factors?
Proportions of a categorized outcome (e.g. proportions of obese and nl pts with high fat diets)
What does a chi-square test compare?
Means--on average was this group bigger/better than the next?
What does a t-test compare?
Attributable risk percent--
ARP = (RR-1)/RR answers a question like "How much of their colon cancer risk was attributable to the high fat intake?"
How do you calculate the excess risk in a population with an exposure to a particular risk factor?
Tends to decrease (as long as the disease prevalence is about 50% in a given group--standard graph/nl distribution) *Decreases because there are more false positives
Shifting a cutoff point to the right (raising it) does what to PPV?
Matching, restriction, randomization, stratified analysis, statistical modeling
How do you control for confounding variables?
X-linked recessive. All daughters born to a father with hemophilia A and an unaffected mom will be carriers (they have to get dad's X). If mom's a carrier, half of her sons will be affected (they get half of mom's x's and dad's Y)
What's the inheritance of hemophilia A?
Sets found in the same folder
Step 1 - Behavioral Science (Ethics/Development/Ph…
First Aid Behavioral Science
Med Ethics - Kaplan
Sets with similar terms
Step 2 CK Biostats
UWORLD Biostats USMLE Step 2 CK
UWORLD Biostats USMLE Step 2 CK
Other sets by this creator
Infections of the Urinary Tract
Other Quizlet sets
biol 104 lecture 8
Practice Interview Questions w/Answers (Clinical P…
college business ch 3 test