Ethics And Stats Flashcards
What are the 4 tenets of medical ethics
Autonomy (requires that pt has mental capacity to evaluate options)
Non-maleficence (avoid deliberate/unintentional harm)
Beneficence (it is our duty to act in the best interest of the pt)
Justice (fairness to individuals)
Pediatric trauma. Needs transfusion but parents jehovah’s witness. What is the right thing to do?
Based on which tenet?
Tell family you are obligated to transfuse even against their will.
Not based on autonomy.
Based on beneficence
Which tenet is the informed consent based on?
Autonomy
What is the mode of the following series:
4, 5, 8, 9, 11, 11
- Mode is the number that appears most often
Which depends on the sample size?
a) standard deviation
b) variance
c) standard error
Standard error
What’s the difference between continuous, ordinal, and nominal
Continuous: real numbers, not integers
Ordinal: categories mean something. Young/middle aged/old or pain scale
Nominal: categories that don’t have order significance
What study design starts with the outcome and examine the exposure?
Case control study
Reason to do: if the outcome is rare
Find 10 ppl who have medulloblastoma and ppl who don’t have it then compare
You use odds ratio in what type of study?
Case control study. Think: you’re studying an odd disease that not very many ppl have. Odds ratio
How to calculate specificity/sensitivity
For sn/sp you start with everyone who has disease
And see how good the test is.
Positive PID (positive in disease) is a sensitive issue
What % will be positive among those who have the disease?
Specificity: negative in health (NIH) NIH is specific
How to calculate positive/negative predictive value
You start with the test and ask how good is the test
Sensitivity/specificity/PPV/NPV
which is affected by the prevalence of the disease?
PPV and NPV
What statistic test do you use for nominal variables?
Chi square
What do you use for ordinal variables?
Wilcox on rank sum. Mann whitney.
Odds ratio vs relative risk
Relative risk is probability 1/probability 2
Odds ratio is p1/(1-p1) divided by p2/(1-p2)
For kaplan meier curves what is log rank test and cox?
Log rank compare survival between groups
Cox adjusts for confounders
What is the most common indication for surgical intervention in older adults >65
Biliary tract disease
What type of surgical procedure is at the highest risk of postop delirium?
Ortho according to passmachine
Cardiac according to truelearn
Difference between adverse events, sentinel events, and close calls
Adverse events: shit happens
Close calls: better to be lucky than good
Sentinel events: should never happen. Wrong site surgery, foreign body, etc
What is the most commong cause of wrong site surgery?
Poor communication
What % needles under 10mm are seen on x-rays?
<25mm?
29%
84% of needles are seen if <25mm
Does sleep deprivation affect the speed or accuracy of a given surgical task?
Slows down.
Accuracy is fine
Technical errors are more likely to happen during routine or complex operations?
Routine operations in complex pt
What is the most common reason for lawsuits in nursing home?
Falls and pressure ulcers
What are formal/informal advance directives?
What is a durable power of attorney?
Formal: written document
Informal: not a written document. Verbal statement given to friend or family by the pt regarding medical wishes
Durable power of attorney: it’s a formal advance directive that designates a surrogate decision maker for the oatient
What happens to systemic vascular resistance with increasing age?
It increases with age
What happens to oxygen consumption with increasing age?
O2 consumption decreases with age
What’s the order of priority for a incapacitated pt?
Siblings, parents, adult children, divorced wife, non-divorced wife
Non-divorced wife > adult children > parent > adult sibling
What is the relationship between type II error and power?
Power = 1 - beta (type II error)
Blinding vs randomization
Which one is used to minimize treatment bias introduced by the pts and clinicians
Blinding
Blinding vs randomization
Which one is used minimize unrecognized variables that may affect outcomes?
Randomization
1,000 total pts. 200 with diabetes 800 without
SSI: 120 (40 with diabetes, 80 without)
What is the odds ratio of SSI occurring in diabetes
Odds of diabetics having SSI / odds of non-diabetics having SSI
40/160 = 0.25 80/720 = 0.111
0.25/0.111 = 2.25
How is power of a study related to null hypothesis?
Power = rejecting null hypothesis when in fact null hypothesis is false.
Accepting null hypothesis when in fact null hypothesis is false = beta. Type II error
Chi square vs Mann whitney U. What variables are used for each test?
Chi square: nominal variables
Mann whitney: ordinal variable
