Reading...
Play button
Play button
Use LEFT and RIGHT arrow keys to navigate between flashcards;
Use UP and DOWN arrow keys to flip the card;
H to show hint;
A reads text to speech;
25 Cards in this Set
- Front
- Back
|
What is a Case Report/Case Series?
|
Study that tells a "story"
Observational, descriptive For the purpose of educating others, stimulating further inquiry |
|
|
What is a one-tailed hypothesis? Example?
|
Make assumption that expected difference in sample mean can only occur in one direction
Ex: Drug X results in higher HDL than placebo. |
|
|
What defines a Cross Sectional Study?
Weaknesses? |
Subjects sampled at a point in time
Observational but analytic (data!) Most useful for chronic diseases (HTN, high chol), common diseases (obesity) Can be used to estimate prevalence of disease process Commonly done with surveys or questionnaires Weaknesses: Not good for rare diseases/exposures Recall Bias Sampling bias |
|
|
What is a two-tailed hypothesis? Example?
|
Expected difference in sample mean could occur in either direction (higher/lower, better/worse)
Ex: Drug X results in lower or higher serum cholesterol than placebo. |
|
|
What defines a Case Control Study?
Weaknesses? |
Persons with condition (cases) are identified, suitable comparison subjects (controls) are identified, and the two groups are compared with respect to prior exposures
Subjects selected based on dz status Retrospective Observational, analytic Good for studying RARE outcomes; efficient in resources and times Weaknesses: Selection bias (cases or controls may not be adequately representative) Recall bias Difficult to prove cause preceded effect |
|
|
T-Test:
Use Strengths Weaknesses |
Used to compare means between two samples/groups when samples have normal distribution
Strengths: good for determining differences between 2 groups, esp w/small sample size Weakness: can only compare 2 groups |
|
|
What defines a cohort study?
Weaknesses? |
People without diseases of interest are followed to see if they develop the disease. Disease incidence in persons with a characteristic compared with incidence in persons without the characteristic.
Can be retrospective or prospective Observational, analytic Can study multiple diseases Can estimate incidence based on new cases that develop. **Better for rare exposures Weaknesses: $$$, time consuming if dz is rare or slow to develop Loss to follow-up (attrition) may lead to selection bias Relatively statistically inefficient unless dz is common |
|
|
ANOVA:
Use Strengths Weaknesses |
Analysis of Variance
Used when want to compare more than 2 means (more efficient than doing a bunch of t-tests)--does so by comparing variance BETWEEN the samples to variance WITHIN the samples Strengths: can compare more than 2 sample means at once Weaknesses: If one mean not equal to others, test doesn't tell you which one is 'off'; just tells you they're not equal to each other |
|
|
What groups do prospective/retrospective cohort studies contain?
|
Prospective cohorts:
Group everyone by EXPOSURE and see if they develop the outcome (smoking could be an exposure) |
|
|
When is an ANOVA unlikely to show difference between groups?
|
When between group variance = within group variance
|
|
|
What defines a randomized control trial?
Weaknesses? |
Experimental
Gold standard for study design People are randomized into groups Always prospective Ideally blinded or double-blinded (study participants and practitioners) **Most like an experiment, provides STRONGEST evidence for causality in relation to temporality and control Weaknesses: Expensive, time consuming, ethically undoable sometimes May not be able to generalize results (patients tend to be very motivated) |
|
|
What is odds ratio?
|
Estimation for relative risk, when you can't calculate relative risk (often for case control studies--when you don't know the exposure, when you're working with the outcome)
=odds that case was exposed/odds control was exposed |
|
|
This test employs an f-statistic.
|
ANOVA
|
|
|
What does an odds ratio of 0.2 mean?
|
Ex with bacterial infections:
Children that had a bacterial infection had 0.2 times (or 1/5 times) the odds of having taken antibx. |
|
|
What defines Systematic Reviews and Meta-Analyses?
|
Conclusion from pooled data from several different studies
Highest statistical power |
|
|
What does an odds ratio of 1 mean?
|
Odds of having exposure = Odds of not having exposure
|
|
|
Chi-Square:
Use Example |
Tests proportions (not means); basis is expected distribution vs observed distribution
Ex: Flip a coin 100 times; expect 50 heads, 50 tails, but observe 30 heads, 70 tails. Chi-Square can tell you if this occurred by chance or not. |
|
|
Cohort Study vs Case-Control Study vs Cross-Sectional Study
|
Cohort: Exposure-->Outcome
Case-Control: Outcome-->Exposure Cross-Sectional: Outcome and Exposure measured at same time |
|
|
Prevention:
Primary Secondary Tertiary |
Primary: prevention of dz in entirety, prevention of complications (most cost-effective); ex: IMMUNIZATIONS, FLUORINATED DRINKING WATER, DIET/EXERCISE
Secondary: aimed at early dz detection, and preventing progression of dz. EX: newly diagnosed diabetic: MONITOR BLOOD SUGAR, HgA1C measurement, ROUTINE EYE EXAMS Tertiary: dz well established, attempting to minimize complications of dz and restore fn. Ex: Patient recently had stroke: SMOKING CESSATION, DAILY ASA, CHOLESTEROL CONTROL, REHAB TO RESTORE FN |
|
|
How would you interpret a p-value of <.001 comparing mean days ill between children with and without pertussis?
|
If there is no difference in mean days of illness, then the probability of the difference found here is 0.1%. Very unlikely to have occurred by chance!
|
|
|
What statistical test would you use to measure if there was a difference in apnea events in children with pertussis vs. children without pertussis?
|
Chi Square (proportion)
|
|
|
What statistical test would you use to measure if there was a difference in mean WBCs in children with pertussis vs. children without pertussis?
|
T-test
|
|
|
What statistical test would you use to measure if there was a difference in mean WBCs in children with pertussis vs. children without pertussis vs. children with influenza?
|
ANOVA
|
|
|
Review PPV/NPV
Sensitivity, Specificity |
xxx
|
|
|
How would PPV and NPV be affected if there were decreased prevalence of a disease being tested?
|
NPV go up
PPV go down |
