Biostats Flashcards
Case Control Study
Looks for prior exposure or risk factor and asks ‘what happened?’ observational and retrospective (odds ratio)
Cohort Study
compares a group with a given exposure or risk factor to a group without. Can be prospective or retrospective relative tisk
What do cohort studies look to answer?
does an increased exposure increase the likelihood of disease? ‘Smokers has a higher risk of developing COPD than nonsmokers’
Cross-sectional study
collects data from a group of people to assess frequency of disease (and related risk factors) * at a particular point in time* (disease prevalence)
T or F. Cross-sectional studies can show risk factor association with disease but does not establish casualty
T.
What is the study sample in a phase I clinical trial?
small no. of healthy volunteers
What is the purpose of a phase I clinical trial?
assesses safety, toxicity, and pharmacokinetics (ADME)
What is the study sample in a phase II clinical trial?
small no. of patients with the disease of interest
What is the purpose of a phase II clinical trial?
assesses treatment efficacy, optimal dosing, and adverse effects
What is the study sample in a phase III clinical trial?
large no. of patients randomly assigned
What is the purpose of a phase III clinical trial?
compares new treatment to the current standard of care
What is the study sample in a phase IV clinical trial?
post-market surveillance
What is the purpose of a phase IV clinical trial?
detects rare or long-term AEs
2x2 Table
What is sensitivity?
The proportion of all people with disease who test positive (or the probability that a test detects disease when it is present)
=TP/(TP+FN)
=1-false negative rate
T or F. A Sensitivity approaching 1 is good for RULING OUT disease
T. SNOUT
What is specificity?
Proportion of all people without disease who test negative
=TN/(TN+FP)
What is specificity good for determining?
ruling IN disease. SPIN
It is used as a confirmatory test a positive screening test
Ex. HIV Testing. Screen with ELISA (sensitive, high false-positive rate though). Follow up with a Western Blot (specific)
What is positive predictive value?
Proportion of positive test results that are true positives
=TP/(TP+FP)
What is negative predicitive value?
Proportion of negative test results that are real negatives
=TN/(TN+FN)
Which predictive value (positive or negative) increases with increased disease prevelance?
Positive
NOTE: Unlike specificity and sensitivity, PPV and NPV vary with prevelance. If prevelance is low, even tests with high sensitivity or specificity will have low PPV)
What is the difference between prevalence and incidence?
Prevalence is the no. of total cases in a population/population whereas incidence is the no. of NEW cases/population
Shortcut eqn for prevalence
=incidence X disease duration
Eqn for odds ratio (for case-control studies)
OR= (a/b)/(c/d) or ad/bc
Eqn for relative risk (for cohort studies)
RR= a/(a+b)/(c/(c+d))
What is attributable risk?
the probability of getting a diseasein the exposed group divided by the probability of getting a disease in the unexposed group (e.g. smoking causes 1/3 cases of pneumonia)
= a/(a+b) - c/(c+d)
What is absolute risk reduction?
reduction in risk associated with a treatment compared to a placebo
= c/(c+d) - a/(a+b)
Number needed to treat and No. needed to harm eqns
Number needed to treat= 1/absolute risk reduction
No. needed to harm= 1/attributable risk
T or F. A systematic error reduces precision (consistency and reproducibility) in a test
F. It reduces accuracy (trueness of the measurements) in a test
What is late-look bias?
gathering info at an inappropriate time (e.g. survey in a fatal disease where only the living can participate)
What is Berkson’s bias?
A form of selection bias that causes hospital cases and controls in a case control study to be systematically different from one another because the combination of exposure to risk and occurrence of disease increases the likelihood of being admitted to the hospital.
What is lead-time bias?
When early detection is considered extended survival even though the timeline/progression of the disease is the same
What is a pygmalion effect?
When a researcher’s belief in the efficacy of a treatment changes the outcome of the treatment
What is the Hawthrone effect?
Occurs when the group being studes changes its behavior owing to the knowledge of being studies
Dr. Hawthrone is watching you
What is the eqn for SEM (standard error of the mean)?
stand. dev/ sqrt(n)
SEM decreases as n. increases
Positive skew of normal Gaussian distribution
Wave to the right
mean > mode > median
What is a type 1 error (a)?
stating that there is an effect or difference when none exists (mistakenly accepting the hypothesis and not the null) (you sAw a difference that wasn’t really there)
NOTE: If p is less than 0.05 then there is less than 5% chnges that the data will show something that is not really there
What is a type II (beta) error?
Accepting the null inappropriately
you were Blind to a difference that did exist
What is Power?
the probability of rejecting the null when it is in fact false
What things affect power?
Sample size. Increased sample size= increased power
Interpreting CI
If a 95% CI for a mean difference between 2 variables includes 0, then there is no significant difference and Ho is not rejected
If a 95% CI for odds ratio or relative risk includes 1, then there is no significant difference and Ho is not rejected
When is parental consent not required?
- Patient is emancipated (married, has children, or is in the military)
- emergencies
- when giving contraceptives
- treatment involving STDs, during pregnancy, or in the management of drug addiction
Surrogates can act in the interest of a patient that is incapacitated or has not given an oral or written advance directive. What is the order of priority in family members?
- Spouse
- Adult children
- Parents
- Siblings
- Other
What are some exceptions to confidentiality?
- Potential harm to others or self
- Suspicion of abuse
- Impaired automobile drivers
- Serious communicable diseases
What is the Tarasoff decision?
law requiring physicians to directly inform aand protect potential victims from harm
