Epidemiology/Biostatistics Flashcards
What are the different types of studies?
- Case control
- Cohort
- Cross-sectional
- Twin concordance
- Adoption
What type of study is a Case-control?
Observational & retrospective
What is the design of a case-control?
- Compares a group of people w/ dz to a group w/o dz
- Looks for prior exposure or RF
- Asks, “what happened?”
What do Case-control studies measure?
Odds ratio (OR)
What is an example of a case control study?
Pts w/ COPD had higher odds of a hx of smoking than those w/o COPD had
What type of study is a Cohort study?
Observational & prospective or retrospective
What is the design of a Cohort study?
- Compares a group w/ a given exposure or risk factor to a group w/o such exposure
- Looks to see if exposure inc the likelihood of dz
What do Cohort studies measure?
Relative risk (RR)
What is an example of a Cohort study?
Smokers had a higher risk of dev COPD than nonsmokers had
What type of study is a Cross-sectional study?
Observational
What is the design of a Cross-sectional study?
- Collects data from a group of people to assess freq of dz (& related RF) at a particular point in time
- Asks, “what is happening?”
What does a Cross-sectional study measure?
- Dz prevalence
- Can show RF assoc. w/ dz but doesn’t est causality
What is the design of a Twin concordance study?
Compares the freq w/ which both monozygotic twins or borh dizygotic twins dev sam dz
What do Twin concordance studies measure?
Heritability
What is the design of an Adoption study?
Compares siblings raised by biological vs adoptive parents
What do Adoption studies measure?
Heritability & influence of environmental factors
What is a clinical study?
Experimental study involving humans
What do clinical trials compare?
Therapeutic benefits of 2 or more tx or of tx & placebo
What improves the study quality of a clinical trial?
- Randomized
- Controlled
- Double-blinded
What is a double blind study?
Neither pt nor doctor knows whether the pt is in the tx or control group
What is a triple blind study?
Refers to the additional blinding of the researchers analyzing the data
What is the study sample of a Phase I clinical trial?
Small # of healthy volunteers
What is the purpose of a Phase I clinical trial?
Assess safety, toxicity & pharmacokinetics
What is the study sample of Phase II clinical trial?
Small # of pt w/ dz of interest
What is the purpose of a Phase II clinical trial?
Assess tx efficacy, optimal dosing & adverse effects
What is the study sample of a Phase III clinical trial?
Large # of pt randomly assigned either to the tx under investigation or to the best available tx (or placebo)
What is the purpose of a Phase III clinical trial?
Compares the new tx to the current standard of care
What is the study sample of a Phase IV clinical trial?
Postmarketing surveillance trial of pts after approval
What is the purpose of a Phase IV clinical trial?
Detects rare or LT adverse effects
What is an Evaluation of dx tests?
Uses 2x2 table comparing test results w/ the actual presence of dz
What are the components of evaluating dx tests?
- TP= True Positive
- FP= False Postiive
- TN= True Negative
- FN= False Negative
What are the fixed properties of evaluation dx tests?
Sensitivity & specificity
Which properties vary w/ prevalence or pretest probability?
PPV & NPV
What is Sensitivity (true-positive rate)?
Proportion of all people w/ dz who test positive or the probability that a test detects dz when dz is present
What does a value approaching 100% sensitivity mean?
Desirable for ruling out dz & indicates a low false-negative rate
What is sensitivity used for?
Screening in dz w/ low prevalence
What does sensitivity do?
- PID= Positive In Dz
- SNOUT= SeNsitivity rule OUT
What is Specificity (true-negative rate)?
Proportion of all people w/o dz who test negative or the probability that a test indicates non-dz when dz is absent
What does a value approaching 100% Specificity mean?
Desirable for ruling in dz & indicates a low false-positive rate
What is specificity used for?
Used as a confirmatory test after a positive screening test
What does specificity do?
- NIH= Negative In Health
- SPIN= SPecificty rules IN
What is Positive predictive value (PPV)?
- Proportion of + test results that are +
- Probability that person actually has the dz given a + test result
What does PPV vary w/?
Directly w/ prevalence or pretest probability
What does a high PPV mean?
High pretest probability
What is Negative Predictive value (NPV)?
- Proportion of negative test results that are true negative
- Probability that person actually is dz free given a negative test result
What does a low NPV mean?
High pretest probability
What is Incidence rate?
# of new cases in a specificed time period
Pop at risk druing same time period
What does incidence look at?
New incidents
What is Prevalence?
# of existing cases
Pop at risk
OR
Incidence rate x average dz
What does prevalence look at?
All current cases
Prevalence> incidence for ____ ____.
Chronic Diseases
What are Odds ratio (OR) used for?
Case-control studies
What does Odds ratio mean?
Odds that the group w/ the dz (cases) was exposed to a RF divided by the odds that the group w/o the dz (controls) was exposed
What is Relative risk used for?
Cohort studies
What does Relative risk mean?
Risk of dev dz in th exposed group divided by risk in the unexposed group
If prevalence is low ___=___.
Relative risk= Odds ratio
What is Attributable risk?
The difference in risk b/w exposed & unexposed groups or the proportion of dz occurrences taht are attributable to the exposure
What is Absolute risk reduction (ARR)?
Absolute reduction in risk assoc. w/ a tx as compared to a control
of pts who need to be tx for 1 pt to benefit is calculated as ______________.
1/absolute risk reduction
of pts who need to be exposed to a RF for 1 pt to be harmed is calculated as ___________.
1/attributable risk
What is precision?
Consistency & reproducibility of a test (reliability)
Absence of random variation in a test
What is a Random error?
Reduces precision in a test
Inc precision→ dec ____ ____.
Standard deviation
What is Accuracy?
The trueness of test measurements (validity)
The absence of systematic error or bias in a test
What is Systematic error?
Reduces accuracy in a test
What is Bias?
Occurs when there is systemic error or favor in a particular direction
What is Selection bias?
Nonrandom assignment of participation in a study group
What is Recall bias?
Knowledge of presence of disorder alters recall by subjects; common in retrospective studies
What is Sampling bias?
Subjects are not representative of the general pop; therefore results are not generalizable.
A type of selection bias
What is Late-look bias?
Info gathered at an inappropriate time
What is Procedure bias?
Subjects in diff groups are not tx the same
What is Confounding bias?
Occurs when factor is related to both exposure & outcome, but is not on the casual pathway; factor distorts or confuses effect of exposure on outcome
What is Lead-time bias?
Early detection confused w/ inc survival; seen w/ improved screening
What is Observer-expectancy effect?
Occurs when a researcher’s belief in the efficacy of a tx changes the outcome of that tx
What is the Hawthorne effect?
Occurs when the group being studied changes its behavior owing to the knowledge of being studied
What are the ways to reduce bias?
- Blind studies to limit influence of participants & researches on interpretation of outcomes
- Placebo control groups
- Crossover studies to limit confounding bias
- Randomization to limit selection bias & confounding bias
- Matching to reduce confounding bias
What are measures of central tendency?
What are measures of dispersion?
- Standard deviation (SD)
- Standard error of the mean (SEM)
- Z-score
- Confidence interval
What is normal distribution?
Gaussian, also called bell-shaped
Mean=median=mode
What is a Positive skew?
- Mean>median>mode
- Asymmetry w/ longer tail on right
- Mode is lease affected by outliers in the sample
What is a Negative skew?
- Mean<median>
</median><li>
Asymmetry w/ longer tail on left</li>
</median>
What is Power (1-ß)?
Probability of rejecting null hypothesis when is it in fact false or the likelihood of finding a diff if one fact exists
What does Power (1-ß) increase w/?
- Inc sample size
- Inc expected effect size
- Inc precision of measurement
What is Meta-analysis?
Pools data & integrates results from several similar studies to reach an overall conclusion
Inc statistical power
What is Meta-analysis limited by?
Quality of individual studies or bias in study selection
What is a confidence interval?
Range of values in which a specified probability of the means of repeated samples would be expected to fall
What is a t-test?
Checks diff b/w the means of 2 groups
What is ANOVA?
Checks diff b/w the means of 3 or more groups
What is Chi-square (x2)?
Test checks diff b/w 2 or more % or proportions of categorical outcoomes (not mean values)
What is Pearson’s correlation coefficient (r)?
- r is always b/w -1 & +1
- The closer the absoulte value of r is to 1 the stronger the linear correlation b/w the 2 variables
What is Primary dz prevention?
Prevent dz occurrence
What is secondary dz prevention?
Early detection of dz
What is Tertiary dz prevention?
Reduce disability from dz
What are Medicare & Medicaid?
Federal programs that originated from amendments of the Social Security Act
Who is Medicare available to?
Pts >65yo & <65yo w/ certain disabilities & those w/ end-stage renal dz
Who is Medicaid available to?
Joint federal & state health assistance for people w/ very low income
