Epidemiology/Biostatistics Flashcards
(96 cards)
1
Q
What are the different types of studies?
A
- Case control
- Cohort
- Cross-sectional
- Twin concordance
- Adoption
2
Q
What type of study is a Case-control?
A
Observational & retrospective
3
Q
What is the design of a case-control?
A
- Compares a group of people w/ dz to a group w/o dz
- Looks for prior exposure or RF
- Asks, “what happened?”
4
Q
What do Case-control studies measure?
A
Odds ratio (OR)
5
Q
What is an example of a case control study?
A
Pts w/ COPD had higher odds of a hx of smoking than those w/o COPD had
6
Q
What type of study is a Cohort study?
A
Observational & prospective or retrospective
7
Q
What is the design of a Cohort study?
A
- 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
8
Q
What do Cohort studies measure?
A
Relative risk (RR)
9
Q
What is an example of a Cohort study?
A
Smokers had a higher risk of dev COPD than nonsmokers had
10
Q
What type of study is a Cross-sectional study?
A
Observational
11
Q
What is the design of a Cross-sectional study?
A
- Collects data from a group of people to assess freq of dz (& related RF) at a particular point in time
- Asks, “what is happening?”
12
Q
What does a Cross-sectional study measure?
A
- Dz prevalence
- Can show RF assoc. w/ dz but doesn’t est causality
13
Q
What is the design of a Twin concordance study?
A
Compares the freq w/ which both monozygotic twins or borh dizygotic twins dev sam dz
14
Q
What do Twin concordance studies measure?
A
Heritability
15
Q
What is the design of an Adoption study?
A
Compares siblings raised by biological vs adoptive parents
16
Q
What do Adoption studies measure?
A
Heritability & influence of environmental factors
17
Q
What is a clinical study?
A
Experimental study involving humans
18
Q
What do clinical trials compare?
A
Therapeutic benefits of 2 or more tx or of tx & placebo
19
Q
What improves the study quality of a clinical trial?
A
- Randomized
- Controlled
- Double-blinded
20
Q
What is a double blind study?
A
Neither pt nor doctor knows whether the pt is in the tx or control group
21
Q
What is a triple blind study?
A
Refers to the additional blinding of the researchers analyzing the data
22
Q
What is the study sample of a Phase I clinical trial?
A
Small # of healthy volunteers
23
Q
What is the purpose of a Phase I clinical trial?
A
Assess safety, toxicity & pharmacokinetics
24
Q
What is the study sample of Phase II clinical trial?
A
Small # of pt w/ dz of interest
25
What is the purpose of a Phase II clinical trial?
Assess tx efficacy, optimal dosing & adverse effects
26
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)
27
What is the purpose of a Phase III clinical trial?
Compares the new tx to the current standard of care
28
What is the study sample of a Phase IV clinical trial?
Postmarketing surveillance trial of pts after approval
29
What is the purpose of a Phase IV clinical trial?
Detects rare or LT adverse effects
30
What is an Evaluation of dx tests?
Uses 2x2 table comparing test results w/ the actual presence of dz
31
What are the components of evaluating dx tests?
* TP= True Positive
* FP= False Postiive
* TN= True Negative
* FN= False Negative
32
What are the fixed properties of evaluation dx tests?
Sensitivity & specificity
33
Which properties vary w/ prevalence or pretest probability?
PPV & NPV
34
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
35
What does a value approaching 100% sensitivity mean?
Desirable for ruling out dz & indicates a low false-negative rate
36
What is sensitivity used for?
Screening in dz w/ low prevalence
37
What does sensitivity do?
* PID= Positive In Dz
* SNOUT= SeNsitivity rule OUT
38
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
39
What does a value approaching 100% Specificity mean?
Desirable for ruling in dz & indicates a low false-positive rate
40
What is specificity used for?
Used as a confirmatory test after a positive screening test
41
What does specificity do?
* NIH= Negative In Health
* SPIN= SPecificty rules IN
42
What is Positive predictive value (PPV)?
* Proportion of + test results that are +
* Probability that person actually has the dz given a + test result
43
What does PPV vary w/?
Directly w/ prevalence or pretest probability
44
What does a high PPV mean?
High pretest probability
45
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
46
What does a low NPV mean?
High pretest probability
47
What is Incidence rate?
_# of new cases in a specificed time period_
Pop at risk druing same time period
48
What does incidence look at?
New incidents
49
What is Prevalence?
_# of existing cases_
Pop at risk
OR
Incidence rate x average dz
50
What does prevalence look at?
All current cases
51
Prevalence\> incidence for ____ \_\_\_\_.
Chronic Diseases
52
What are Odds ratio (OR) used for?
Case-control studies
53
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
54
What is Relative risk used for?
Cohort studies
55
What does Relative risk mean?
Risk of dev dz in th exposed group divided by risk in the unexposed group
56
If prevalence is low \_\_\_=\_\_\_.
Relative risk= Odds ratio
57
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
58
What is Absolute risk reduction (ARR)?
Absolute reduction in risk assoc. w/ a tx as compared to a control
59
of pts who need to be tx for 1 pt to benefit is calculated as \_\_\_\_\_\_\_\_\_\_\_\_\_\_.
1/absolute risk reduction
60
of pts who need to be exposed to a RF for 1 pt to be harmed is calculated as \_\_\_\_\_\_\_\_\_\_\_.
1/attributable risk
61
What is precision?
Consistency & reproducibility of a test (reliability)
Absence of random variation in a test
62
What is a Random error?
Reduces precision in a test
63
Inc precision→ dec ____ \_\_\_\_.
Standard deviation
64
What is Accuracy?
The trueness of test measurements (validity)
The absence of systematic error or bias in a test
65
What is Systematic error?
Reduces accuracy in a test
66
What is Bias?
Occurs when there is systemic error or favor in a particular direction
67
What is Selection bias?
Nonrandom assignment of participation in a study group
68
What is Recall bias?
Knowledge of presence of disorder alters recall by subjects; common in retrospective studies
69
What is Sampling bias?
Subjects are not representative of the general pop; therefore results are not generalizable.
A type of selection bias
70
What is Late-look bias?
Info gathered at an inappropriate time
71
What is Procedure bias?
Subjects in diff groups are not tx the same
72
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
73
What is Lead-time bias?
Early detection confused w/ inc survival; seen w/ improved screening
74
What is Observer-expectancy effect?
Occurs when a researcher's belief in the efficacy of a tx changes the outcome of that tx
75
What is the Hawthorne effect?
Occurs when the group being studied changes its behavior owing to the knowledge of being studied
76
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
77
What are measures of central tendency?
78
What are measures of dispersion?
* Standard deviation (SD)
* Standard error of the mean (SEM)
* Z-score
* Confidence interval
79
What is normal distribution?
Gaussian, also called bell-shaped
Mean=median=mode
80
What is a Positive skew?
* Mean\>median\>mode
* Asymmetry w/ longer tail on right
* Mode is lease affected by outliers in the sample
81
What is a Negative skew?
* Mean
Asymmetry w/ longer tail on left
82
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
83
What does Power (1-ß) increase w/?
* Inc sample size
* Inc expected effect size
* Inc precision of measurement
84
What is Meta-analysis?
Pools data & integrates results from several similar studies to reach an overall conclusion
Inc statistical power
85
What is Meta-analysis limited by?
Quality of individual studies or bias in study selection
86
What is a confidence interval?
Range of values in which a specified probability of the means of repeated samples would be expected to fall
87
What is a t-test?
Checks diff b/w the means of 2 groups
88
What is ANOVA?
Checks diff b/w the means of 3 or more groups
89
What is Chi-square (x2)?
Test checks diff b/w 2 or more % or proportions of categorical outcoomes (not mean values)
90
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
91
What is Primary dz prevention?
Prevent dz occurrence
92
What is secondary dz prevention?
Early detection of dz
93
What is Tertiary dz prevention?
Reduce disability from dz
94
What are Medicare & Medicaid?
Federal programs that originated from amendments of the Social Security Act
95
Who is Medicare available to?
Pts \>65yo & \<65yo w/ certain disabilities & those w/ end-stage renal dz
96
Who is Medicaid available to?
Joint federal & state health assistance for people w/ very low income
