1
Q
The purpose of screening is to do what?
A
Identify disease early in its development so it can be stopped and have less effect
2
Q
What is the proportion of cases of a condition accurately identified by the screening test?
A
Yield
3
Q
Two characteristics of a good disease to screen for:
A
Significant public health problem
Treatment is readily available
4
Q
Are screening tests diagnostic?
A
No
5
Q
Who should be targeted for screening?
A
Those at increased risk
6
Q
What are 6 potential harms of screening?
A
Physical harm
False-negative result
False-positive result
Over-diagnosis
Psychological harm
Incidental findings
7
Q
What issue occurs if a screening produces a false-negative result?
A
Patient does not receive the care that they should
8
Q
What issue occurs if a screening produces a false-positive result?
A
Causes unnecessary stress and anxiety on the patient
9
Q
How often a test correctly identifies people who have a disease and others who do not. What is being described?
A
Accuracy
10
Q
The ability of a test to distinguish between those who have a disease and those who don’t.
A
Internal validity
11
Q
What is defined as the degree to which repeated testing results in the same results?
A
Precision
12
Q
Differences in measurement between observers.
A
Inter-observer variation
13
Q
Differences in measurement between one observer.
A
Intraobserver
14
Q
__________ is the ability to get correct results.
A
Accuracy
15
Q
___________ is the ability to get repeatable results.
A
Precision
16
Q
Is accuracy or precision more important?
A
Accuracy
17
Q
The ability of a test to correctly identify people with a disease.
A
Sensitivity
18
Q
Sensitivity equation.
A
true positives / (true positives + false negatives)
19
Q
A test with poor sensitivity will have a high number of _______________.
A
False negatives
20
Q
Percentage (or rate) of false negatives = _________________
A
1 - sensitivity
21
Q
If sensitivity increases, the false negative rate ____________.
A
Decreases
22
Q
If sensitivity decreases, the false negative rate ____________.
A
Increases
23
Q
The ability of a test to correctly identify people without a disease.
A
Specificity
24
Q
Specificity equation.
A
of true negatives / # of true negatives + # of false positives
25
A test with poor specificity will have a higher number of ___________________.
False positives
26
Percentage of false positives = _________________
1 - specificity
27
If specificity increases, false positive rate _____________.
Decreases
28
If specificity decreases, false positive rate _____________.
Increases
29
When should sensitivity be emphasized over specificity?
When the consequences are severe for missing the diagnosis
## Footnote
When more suspicious of the disease
30
When should specificity be emphasized over sensitivity?
When the treatment involves risk and costs
## Footnote
When less suspicious of the disease
31
A highly valid test will have high ___________ and high ___________.
Sensitivity and specificity
32
Confrontation visual fields has a high ___________ and low ___________.
Specificity
## Footnote
Sensitivity
33
Automated threshold visual fields has a high ___________ and low ____________.
Sensitivity
## Footnote
Specificity
34
What curve plots sensitivity vs. specificity?
Receiver Operating Characteristic curve
35
Better tests have ROC curves that crowd toward where?
Upper left corner of plot
36
What does a larger area under the ROC curve indicate?
A more accurate test
37
Do sensitivity and specificity take into account disease prevalence?
No
38
Predictive values account for _____________, ____________ and ___________ of the disease in the population tested.
Sensitivity, specificity and prevalence
39
__________________ is the probability of the patient having the disease given a positive test result.
Positive predictive value (PPV)
40
Positive predictive value equation.
# of true positives / # of all positives
41
How does PPV change as prevalence increases?
PPV increases
42
__________________ is the probability of the patient not having the disease given a negative result.
Negative predictive value (NPV)
43
Negative predictive value equation.
# of true negatives / # of all negatives
44
How does NPV change with prevalence of a disease?
NPV increases with low prevalence; decreases with increased prevalence
45
If the sample provides enough evidence against the claim that there's no effect on the population (null hypothesis) -> _________ the null hypothesis.
Reject
46
If the sample does NOT provide enough evidence that there's no effect on the population (null hypothesis) -> _________ the null hypothesis.
Fail to reject
47
When does a type 1 error occur?
When the null hypothesis is rejected incorrectly
48
When does a type 2 error occur?
When you incorrectly fail to reject the null hypothesis
49
What error occurs when you wrongly reject a null hypothesis?
Type 1 error
50
What error occurs when you fail to reject a false null hypothesis?
Type II error
51
What error occurs when we think treatments are NOT different but they actually ARE different?
Type II error
52
What error occurs when we think treatments ARE different but treatments are NOT different?
Type I error
53
P values indicate the chance of a ____________ error occurring.
Type I error
54
P < ________ is considered statistically significant.
0.05
55
P > 0.05 means what?
No effect was observed (fail to reject null)
56
_________ of a study represents the probability of finding a difference that exists in a population.
Power
57
Power = 1 - probability of making a ___________ error.
Type II
58
What indicates the level of uncertainty around a measure of effect?
Confidence interval
59
Confidence level = 1 - ___________.
P value
60
A more narrow confidence interval = a more ________ result.
Precise
61
The number of events in a treated or control group, divided by the number of people in that group.
Absolute risk
62
Relative risk equation.
Disease incidence in exposed group / disease incidence in non-exposed group
63
Odds ratio equation.
Odds of event in treatment group/ odd of event in control group
64
Odds ratio > 1.00 indicates what?
Exposure results in greater risk
65
Odds ratio < 1.00 indicates what?
Exposure results in less risk
66
Odds ratio = 1.00 indicates what?
Exposure results in no risk
67
Odds equation.
probability / (1 - probability)
68
If the probability of occurrence is 10%, the odds = __________.
1 to 9
69
When the probability is small, the odds are ___________________.
Identical to the probability
70
If these are the probabilities, what are the odds that the horse will win the race?
Odds = 1.5
71
What are the odds of dying if the probability of mortality is 30%?
3/7
72
What are the odds of dying if the probability of mortality is 50%?
1/1 or for every 1 person that dies, 1 will live
73
What are the odds of dying if the probability of mortality is 80%?
4/1 or for every 4 people that die, 1 will live
74
If multiple independent things happen at once, how would you calculate the probability of both happening?
Multiply by one another
75
What does likelihood ratio measure?
The performance of a diagnostic test
76
____________________ is used when a test result is positive (abnormal outcome, disease present).
Positive likelihood ratio
77
_____________________ is used when a test result is negative (normal outcome, disease absent).
Negative likelihood ratio
78
Likelihood ratio > 1 indicates what?
Presence of disease
79
Likelihood ratio < 1 indicates what?
Absence of disease
80
What specifies the number of persons needed to be treated, on average to prevent one more event?
Number needed to treat (NNT)
81
NNT is the reciprocal of.
Absolute risk reduction (ARR)
82
Absolute Risk Reduction (ARR) equation.
Absolute risk of events in control group / absolute risk of events in treatment group
83
NNT =.
1/ARR
84
Parametric tests assume _______________ and that the data are ___________________.
Equal variance
## Footnote
Normally distributed
85
Do non-parametric tests have assumptions?
No
86
What test is used on discrete categorical data?
Chi-square test
87
Is Chi-Square parametric or Non-Parametric?
Non-parametric
88
A T-test is used with what type of data?
Continuous data
89
What type of test determines if a sample mean is different than a specified value?
One sample T-test
90
What type of test determines if a sample means of two groups are different from each other?
Two-sample T-test
91
What test compares means from the same subjects over time to determine if a within-person difference exists?
Paired T-test
92
Comparing IOP before and after glaucoma drops are administered is an example of what test?
Paired T-test
93
______________ provides a test of whether the means of three or more unrelated groups are equal.
Analysis of variance (ANOVA)
94
One-way ANOVA compares what?
One independent variable and one dependent variable
95
Two-way ANOVA compares what?
Two independent variables
96
What statistical test is used to predict the value of a variable based on the value of another variable?
Linear regression
97
Determining if smoking status is associated with lung cancer diagnosis is an example of what type of test?
Chi-squared test
98
What test assesses the association between two categorical variables?
Chi-squared test
99
Determining if average blood pressure in a sample is different from 120mmhg is an example of what test?
One-sample T-test
100
What 3 tests compare means of continuous variables?
One-sample T-test
## Footnote
Two-sample T-test
Paired T-test
101
Looking at the BMI difference in smokers vs. non-smokers is an example of what test?
Two-sample T-test
102
Looking at the weight difference before and after a diet program is an example of what test?
Paired T-test
103
Assessing cholesterol levels in patients with different diet types is an example of what test?
One-way ANOVA
104
What two tests compare means across more than two groups?
One-way ANOVA
## Footnote
Two-way ANOVA
105
Testing the effect of diet type and gender on cholesterol level is an example of what test?
Two-way ANOVA
106
Predicting systolic blood pressure based on age, BMI, and smoking status is an example of what test?
Linear regression
107
What test compares the relationship between a continuous dependent variable and one or more independent variables?
Linear regression
108
When should a chi-squared test be used?
When comparing proportions
109
When should a one-sample T-test be used?
When comparing sample mean to a known value
110
When should a Two-sample T-test be used?
When comparing means between two independent groups
111
When should a Paired T-test be used?
When comparing the means of related groups
112
When should a One-way ANOVA test be used?
When comparing one categorical independent variable with 3+ groups
113
When should a Two-way ANOVA test be used?
When comparing two categorical independent variables
114
When should a Linear regression test be used?
When predicting or adjusting for multiple variables
Block 10
flashcards
Decks in class (32)
# Cards
-
mono: psychophysics73
-
mono: functional retina105
-
pharm: intro to ANS62
-
pharm: cholinergic agonists46
-
mono: parallel processing65
-
mono: striate cortex66
-
pharm: cholinergic agonists 280
-
micro: mites, lice, bugs42
-
micro: fungi101
-
epi: ethics and landmark studies91
-
mono: electrodiagnostics74
-
micro: protists94
-
micro: helminths76
-
micro: gram + pathogens123
-
mono: extrastriate cortex68
-
mono: basics of color vision33
-
mono: CIE color system19
-
epi: testing and stats114
-
micro: gram - cocci26
-
pharm: adrenergic agonists80
-
pharm: nicotine antagonists and skeletal muscle relaxants47
-
pharm: indirect agonist and adrenergic antagonist52
-
micro: gram - non-enteric rods51
-
micro: gram - enteric rods53
-
micro: atypical bacteria 157
-
mono: genetics of color vision36
-
micro: patience’s question review119
-
mono: color labels and color opponency43
-
mono: color vision anomalies57
-
micro: atypical pathogens 257
-
epi: study design and evidence based practice111
-
pharm: adrenal and bone86
