Biostats Flashcards
Study that is determining the frequency of disease at a particular point in time (prevalence), purely observational
cross-sectional study
Study that compares a group of people with disease to a group without disease, looks for prior exposures or risk factors, retrospective and observational
Case-control study
Study that compares a group with a given exposure exposure or risk factor to a group without that exposure, determines whether exposure affects likelihood of disease
Cohort study
What study type determines the odds ratio?
case-control (group of people with dz vs. without dz)
What study type determines the relative risk?
cohort study (group with exposure to risk factor vs. group without exposure)
What type of study always involves an intervention?
clinical trial
Phase I drug trial
Is this drug safe?
-test drug on healthy volunteers
Phase II drug trial
Is this drug effective?
-test drug in group of pts with disease
Phase III drug trial
Is this drug as good or better than what already exists?
-large number of pts randomly assigned to treatment or placebo group
Phase IV drug trial
Can the drug stay on the market?
-postmarketing surveillance of rare or long-term side effects
Sensitivity
proportion of people with disease who test positive
TP / TP + FN
Best type of screening test has high (sensitivity or specificity)
sensitivity
Best type of diagnostic test has high (sensitivity or specificity)
specificity
Specificity
proportion of people without dz who test negative
TN / TN + FP
What is PPV?
proportion of positive test results that are true positives
PPV = TP / TP + FP
What is NPV?
proportion of negative test results that are true negatives
NPV = TN / TN + FN
How does an increase in disease prevalence affects PPV and NPV?
increase in PPV, decrease in NPV
What is incidence?
# of new cases / # of people at risk -incidence looks at new cases in a given period of time
What is prevalence?
# of existing cases / total # of people in a population -looks at all current cases
What happens to incidence and prevalence for a disease with short duration (i.e. flu)?
prevalence ~ incidence
What happens to incidence and prevalence for a disease with long duration (i.e. COPD)?
prevalence > incidence
Equation for Confidence intervals
CI = Mean +/- Z-score x SEM
Z = 2 for 95% confidence intervals
Equation for SEM
SEM = SD / square root of N
What is Type I error (alpha)?
false-positive error - stating there is a difference in an effect when none exists
What is a Type II error (beta)?
false-negative error - stating there is no difference when one exists
What is power?
probability of rejecting the null hypothesis when it is false
What increases power?
increase sample size, increase in expected effect size, increase in precision of measurement
calculate by taking 1 - beta error
Equation for odds ratio
a/c / b/d
odds that group with dz was exposed to risk factor divided by odds that group without dz was exposed to risk factor
Relative risk
a/(a+b) / c/(c+d)
risk of developing a dz if you are exposed to a risk factor divided by risk in unexposed group
Attributable risk
a/a+b - c/c+d
difference in risk between exposed and unexposed (how much of dz occurrences are attributable to exposure)
Relative risk reduction
RR - 1
proportion of risk reduction attributable to intervention as compared to control
Absolute risk reduction
c/c+d - a/a+b
difference in risk attributable to the intervention as compared to a control
Number needed to treat
NNT = 1/ARR
Number needed to harm
NNH = 1/AR