Behavioral Sciences Flashcards
What is an advanced directive?
Set of instructions given by patient in anticipation for the need of a medical decision, varies by state law.
Forms include: Oral - based on what they have repeatedly said (might be a wife or someone's word) Written (living will is a common form) Medical power of attorney Do not resuscitate
What is a living will and how does it differ from a medical power of attorney?
Living will - document which assigns which procedures they would be willing to accept or reject for a critical illness. Quite inflexible - if you have one, this is consulted first.
Medical power of attorney - A specific person is assigned to make all of your medical decisions. Since they are a living person, it is more flexible than living will.
If a patient has no decision-making capacity and no written advanced directive / power of attorney, who makes end of life decisions?
Next of kin makes decisions, in the following order:
Spouse -> adult Children -> Parents -> Siblings
the “spouse ChiPS” in.
What is the definition of pica? What are the three types?
Craving non-food or non-staple food substances which are not culturally acceptable for greater than 1 month. Often leads to nutritional deficiencies, and is more common in pregnancy.
- Earth/soil-rich substances
- Raw starch (i.e. flour or cornstarch)
- Ice
UWorld #8893
What is capitation vs global payment?
Capitation - physicians receive a set amount per patient per time, regardless of how much the patient uses the healthcare system (a lump sum of capital)
Global payment - insurance company pays for all things associated with a particular procedure, typically elective surgery, in one lump sum. I.e. this amount of money should cover the surgery as well as pre-op/post-op expenses.
How does point of service (POS) differ from health maintenance organization (HMO)?
Both require PCP referrals to see a specialist, but point of service is allowed to see doctors outside of the preferred network, where HMO patients must stay in network
How does PPO differ from EPO?
EPO = exclusive provider organization, patients must stay within a network but do not need a PCP referral
PPO = patients do not need PCP referral and can see out of network
EPO is to PPO as HMO is to POS.
When is hospice care available to patients?
Whenever their prognosis is <6 months.
What is the difference between a cross-sectional study and an ecological study?
Cross-sectional - observes frequency of disease and frequency of exposures among INDIVIDUALS at a single timepoint
Ecological study - observes frequency of disease and exposures among POPULATIONS at a single time point -> an even more general iteration of a cross-sectional study.
What is the purpose of cross-sectional / ecological studies?
They cannot prove causality, but are used to formulate hypotheses regarding diseases and their associated risk factors. Can only show correlations, not causation.
In order to better prove causality, you need case-control or cohort studies.
What is the utility of a cohort study vs a case-control study?
Cohort - For studying rare exposures. Have a group of people with a certain exposure -> see if they develop disease
Case-control - For studying rare diseases. Have a group of people with the disease -> see what they are exposed to vs your control group.
What is the measure of likelihood in a cohort study vs a case control study?
Give this in an explanation of COPD vs smokers.
Cohort study - What is the relative risk (RR) of developing COPD given you are a smoker
Case-control study - Odds ratio - What is the ratio of: the odds that you got COPD if you were a smoker divided by the odds that you did not get COPD if you were not a smoker all divided by the odds you got COPD if you were a nonsmoker divided by the odds you did not get COPD if you were a nonsmoker.
What is the formula for odds ratio?
Assuming the normal four-quadrant square, with disease on top and test on the side
OR = (a/b) / (c/d)
What is the formula for relative risk?
Dealing with probabilities so.
Chance of disease given you had exposure / Chance of disease given you did not have exposure.
a/(a+b) / c/(c+d)
Remember this a cohort study
What is a twin concordance study and how do you tell if something is heritable or not?
Twin concordance - With the same parents, chances of developing the same condition with monozygotic vs dizygotic twins. Greater difference in concordance between monozygotic and dizygotic twins = highly heritable. Less of a difference in concordance between MZ and DZ = more environmental.
What is an adoption study?
Measures concordance of developing a disease among monozygotic twins raised by biological vs adoptive parents, or them being split.
There are multiple study designs
What is meant by a triple blind study?
Double blind - The patient and doctors analyzing the data are not aware of who is being treated
Triple blind - The researcher analyzing the data is also not aware of who was treated
What are the four phases of clinical trials?
Does the drug SWIM:
- Safe - Assess toxicity in healthy volunteers
- Work - Assess treatment efficacy in a small number of patients with disease of interest
- Improvement - Large RCT which assesses new treatment vs best treatment available (or placebo)
- Market - Post-marketing surveillance determines longterm adverse effects. Can result in treatment withdrawal.
What measure is (1 - false positive rate)? How do you calculate?
Specificity!
True negative rate = TN / (TN + FP)
(No disease and test is negative) / (No disease and test is negative + No disease and test is positive)
= d / b+d
What measure is (1 - false negative rate)? How do you calculate?
Sensitivity!
True positive rate = TP / (TP + FN)
(Have disease and test is positive) / (Have disease and test is positive + Have disease and test is negative)
= a / (a+c)
What is the positive predictive value and how does it change with population prevalence?
Gives you the usefulness of the test if the test is positive. The test is more likely to have a good PPV if the disease is more prevalent in the population -> more true positives relative to false positives overall
Population prevalence = pretest probability, btw
How do you calculate positive predictive value? What is the range of possible values?
TP / (TP + FP)
Range of values: 0% to 100%. Obviously, it will vary with prevalence because the more true positives you have, the more likely a positive is to be meaningful (numerator is greater relative to denominator)
-> #1233 illustrates this nicely
What is the negative predictive value and how does it change with population prevalence?
Gives you the usefulness of a negative test result (higher value = more likely that your negative test result is significant). The test is more likely to be relevant if the disease is less prevalent in the population -> more true negatives relative to false negatives overall.
How do you calculate negative predictive value?
TN / (TN + FN)
When are likelihood ratios useful and what determines if something is a very useful diagnostic test with these ratios?
Useful when the prevalence is unfavorable for dealing with PPV / NPV -> tells you the utility of just the test alone.
+LR > 10 is a useful test to rule in
-LR < 0.1 is a useful test to rule out
How do you calculate positive likelihood ratio?
LR+ = sensitivity / (1-specificity) = true positive rate / false positive rate.
Notice that this is very similar to the calculation of PPV, just doesn’t depend on prevalence in the population. It is the RATE of true positives / RATE of false positives.
