Brainscape's Knowledge GenomeTM

Browse over 1 million classes created by top students, professors, publishers, and experts.


See full index

EPIB 601 Fundamentals > Pre-midterm > Flashcards

Pre-midterm Flashcards

1
Q

What distinguishes a ratio?

A
  • A/B

- Dimensionless or dimensions

How well did you know this?
1
Not at all
2
3
4
5
Perfectly
2
Q

What distinguishes a proportion?

A
  • A/(A+B)
  • 0, 1
  • Dimensionless
How well did you know this?
1
Not at all
2
3
4
5
Perfectly
3
Q

What distinguishes a rate?

A
  • N/T (instantaneous change over time)
  • Has dimensions
  • E.g., speed
How well did you know this?
1
Not at all
2
3
4
5
Perfectly
4
Q

Two categories of incidence measures?

A
  1. Cumulative incidence/risk/incidence proportion/attack rate
  2. Incidence density/incidence rate
How well did you know this?
1
Not at all
2
3
4
5
Perfectly
5
Q

Synonyms of incidence proportion

A
  • Cumulative incidence
  • Risk
  • Attack rate
How well did you know this?
1
Not at all
2
3
4
5
Perfectly
6
Q

Formula of incidence proportion

A

new cases during followup/#people followed-up

How well did you know this?
1
Not at all
2
3
4
5
Perfectly
7
Q

What are the assumptions of incidence proportion?

A
  • Conditional on no competing risks
  • Timing is everything for the interpretation (over a year? 2 years?)
  • Fixed cohort with no exits
How well did you know this?
1
Not at all
2
3
4
5
Perfectly
8
Q

Interpretation of incidence proportion?

A

Risk/probability of developing a disease

How well did you know this?
1
Not at all
2
3
4
5
Perfectly
9
Q

Formula of incidence density

A

new cases during follow-up/total person-time at risk during follow-up

How well did you know this?
1
Not at all
2
3
4
5
Perfectly
10
Q

Unit of incidence density

A

Time^(-1)

How well did you know this?
1
Not at all
2
3
4
5
Perfectly
11
Q

Incidence density is a…

A

Rate (like speed)

How well did you know this?
1
Not at all
2
3
4
5
Perfectly
12
Q

What is incidence density useful for?

A
  • Dynamic cohorts

- Dealing with competing risks

How well did you know this?
1
Not at all
2
3
4
5
Perfectly
13
Q

Problem with incidence density?

A

Person-years are considered equivalent, but they are not, making it hard to compare studies

How well did you know this?
1
Not at all
2
3
4
5
Perfectly
14
Q

Interpretation of incidence density?

A

If in a steady population:

  • Put over 1 year
  • Take reciprocal (1/xx)
  • Interpret as average time until event occurs
How well did you know this?
1
Not at all
2
3
4
5
Perfectly
15
Q

Risk Incidence rate?

A

Risk = (Incidence Rate)*Time

  • Good if risk is < 20%
  • Becomes less good as we go through years, it becomes an overestimate
How well did you know this?
1
Not at all
2
3
4
5
Perfectly
16
Q

When are survival analyses good?

A
  • Dynamic cohort

- Changing incidence rates

How well did you know this?
1
Not at all
2
3
4
5
Perfectly
17
Q

Assumptions of survival analyses?

A
  • No changes in survivorship over calendar time

- Uniform withdrawal/events over interval

How well did you know this?
1
Not at all
2
3
4
5
Perfectly
18
Q

Formula for prevalence proportion?

A

existing cases/#total population

How well did you know this?
1
Not at all
2
3
4
5
Perfectly
19
Q

Incidence Prevalence

A

If steady state:
- P/(1-P) = (Incidence rate)Duration of disease
If rare disease (P < 0.10)
- P = (Incidence rate)Duration of disease

How well did you know this?
1
Not at all
2
3
4
5
Perfectly
20
Q

Direct standardization

A

Apply rates from population of interest to the standard population

How well did you know this?
1
Not at all
2
3
4
5
Perfectly
21
Q

Standardized mortality ratio formula

A

= observed events/expected events x 100

How well did you know this?
1
Not at all
2
3
4
5
Perfectly
22
Q

7 steps of natural history of disease?

A
  1. Biological onset
  2. Pathological evidence
  3. Symptoms
  4. Medical care sought
  5. Diagnostic
  6. Treatment
  7. Outcome (cure, control, disability, death)
How well did you know this?
1
Not at all
2
3
4
5
Perfectly
23
Q

According to the natural history of disease, what is the preclinical phase? and the clinical phase?

A

Preclinical: from biological onset to symptoms
Clinical: from symptoms to outcome

How well did you know this?
1
Not at all
2
3
4
5
Perfectly
24
Q

According to the natural history of disease, when does primary prevention takes place and what is it?

A
  • Before biological onset

- Attempt to prevent development (e.g., vaccination)

How well did you know this?
1
Not at all
2
3
4
5
Perfectly
25
Q

According to the natural history of disease, when does secondary prevention takes place and what is it?

A
  • Between biological onset and symptoms appearance

- Screen to improve prognosis

How well did you know this?
1
Not at all
2
3
4
5
Perfectly
26
Q

According to the natural history of disease, when does tertiary prevention takes place and what is it?

A
  • Between appearance of symptoms and outcome

- Management to diminish impact of disease

How well did you know this?
1
Not at all
2
3
4
5
Perfectly
27
Q

What is sensitivity and what does it result in?

A
  • P(T+ | D+)

- Higher sensitivity means fewer false negatives

How well did you know this?
1
Not at all
2
3
4
5
Perfectly
28
Q

Formula for sensitivity?

A

= True positives / all D+

How well did you know this?
1
Not at all
2
3
4
5
Perfectly
29
Q

What is specificity and what does it result in?

A
  • P(T- | D-)

- Higher specificity means fewer false positives

How well did you know this?
1
Not at all
2
3
4
5
Perfectly
30
Q

Formula for specificity?

A

= True negatives / all D-

How well did you know this?
1
Not at all
2
3
4
5
Perfectly
31
Q

What happens when you require T+ on two tests to be considered D+, such as in sequential testing?

A

Less Sensitivity = Sensitivity1 x Sensitivity2 (you end up with more false negatives)
More specificity = Specificity1 + Specificity2 - (Spec1 x Spec2) (you end up with less false positives)

How well did you know this?
1
Not at all
2
3
4
5
Perfectly
32
Q

What happens when you require T+ on one of two tests to be considered D+, such as in simultaneous testing?

A

More sensitivity = Sensitivity1 + Sensitivity2 - (Sens1 x Sens2) (you end up with less false negatives)
Less specificity = Specificity1 x Specificity2 (you end up with more false positives)

How well did you know this?
1
Not at all
2
3
4
5
Perfectly
33
Q

Positive predictive value formula

A

P (D+ | T+) = True positives/all T+

How well did you know this?
1
Not at all
2
3
4
5
Perfectly
34
Q

Negative predictive value formula

A

P (D- | T-) = True negatives/all T-

How well did you know this?
1
Not at all
2
3
4
5
Perfectly
35
Q

Relation between predictive value and prevalence?

A

Increased prevalence leads to increased PPV

How well did you know this?
1
Not at all
2
3
4
5
Perfectly
36
Q

Relation between Sensitivity/specificity and predictive value?

A

Specificity impacts PPV

Sensitivity impacts NPV

How well did you know this?
1
Not at all
2
3
4
5
Perfectly
37
Q

Formula for PPV with sensitivity and specificity?

A

PPV = (sens x prev)/[(sens x prev) + (1-spec)*(1-prev)]

How well did you know this?
1
Not at all
2
3
4
5
Perfectly
38
Q

Formula for NPV with sensitivity and specificity?

A

NPV = (spec x 1-prev)/[(spec1-prev)+(1-sens)prev)]

How well did you know this?
1
Not at all
2
3
4
5
Perfectly
39
Q

What is a likelihood ratio?

A

Likelihood that a test result would be expected in a person D+
Compared to
The likelihood that the same result would be expected in a person D-

How well did you know this?
1
Not at all
2
3
4
5
Perfectly
40
Q

How is LR+ calculated?

A

P(true positives)/P(false positives) = sens/(1-spec)

How well did you know this?
1
Not at all
2
3
4
5
Perfectly
41
Q

How is LR+ interpreted?

A

The higher the LR+, the more likely the test is positive in D+ person compared to D- person (rules in the disease)

How well did you know this?
1
Not at all
2
3
4
5
Perfectly
42
Q

How is LR- calculated?

A

P(false negatives)/P(true negatives) = (1-sens)/spec

How well did you know this?
1
Not at all
2
3
4
5
Perfectly
43
Q

How is LR- interpreted?

A

The lower the LR-, the less likely the test is negative in a D+ person compared to a D- person (rules out the disease)

How well did you know this?
1
Not at all
2
3
4
5
Perfectly
44
Q

What are significant numbers for LR?

A

LR = 1, NS
LR+ > 1 –> associated with disease (good from > 10)
LR- < 1 –> associated with no disease (good from < 0.1)

How well did you know this?
1
Not at all
2
3
4
5
Perfectly
45
Q

Overall percent agreement formula?

A

agreed/#total ratings

How well did you know this?
1
Not at all
2
3
4
5
Perfectly
46
Q

Corrected percent agreement formula?

A

yes-yes/(#total ratings - #no-no)

How well did you know this?
1
Not at all
2
3
4
5
Perfectly
47
Q

Kappa statistics formula?

A

(% agreement observed - % agreement expected)/(100 - % agreement expected)

How well did you know this?
1
Not at all
2
3
4
5
Perfectly
48
Q

Prognosis is susceptible to which bias?

A

Lead time bias

- Prognosis usually from diagnosis, ideally from onset, but better screening may inflate survival

How well did you know this?
1
Not at all
2
3
4
5
Perfectly
49
Q

What is a causal mechanism?

A

Joint action of multiple component cause

  • Causal pie
  • Is a sufficient cause
How well did you know this?
1
Not at all
2
3
4
5
Perfectly
50
Q

What is a sufficient cause?

A

The smallest set of conditions and events that inevitably produces the disease (a complete pie)

How well did you know this?
1
Not at all
2
3
4
5
Perfectly
51
Q

What is a component cause?

A

A piece of the pie

How well did you know this?
1
Not at all
2
3
4
5
Perfectly
52
Q

When are component cause necessary?

A

When they appear in all causal pies

How well did you know this?
1
Not at all
2
3
4
5
Perfectly
53
Q

What do we need to know about the strength of a component cause?

A
  • Cannot be determined for individual cases - they were all equal
  • At the population level, a component cause is stronger when it takes par in a larger proportion of disease cases (e.g., smoking) - depends on the prevalence
How well did you know this?
1
Not at all
2
3
4
5
Perfectly
54
Q

What is the sum of all component causes?

A

Infinite, due to interactions

How well did you know this?
1
Not at all
2
3
4
5
Perfectly
55
Q

What is induction time?

A

The time between the action of a component cause and the completion of a sufficient cause (pie).
- For the last component cause, induction time is always 0.

How well did you know this?
1
Not at all
2
3
4
5
Perfectly
56
Q

What is the latent period?

A

Period between the action of the last component cause/disease initiation and the clinical manifestation
- The latent period is reduced with screening

How well did you know this?
1
Not at all
2
3
4
5
Perfectly
57
Q

What is the empirical induction period?

A

It is used when the specific causal mechanisms are unknown and that the induction and latent period may not be distinguished
- Latent + induction = empirical induction period

How well did you know this?
1
Not at all
2
3
4
5
Perfectly
58
Q

What are 4 of Hill’s causal guidelines, and which one is necessary(*)?

A

*Temporality
Strength of association (strong associations are less likely to be confounded, but sometimes causal is weak)
Specificity (rare and unnecessary)
Biological gradient (dose response relation, linear or not)

How well did you know this?
1
Not at all
2
3
4
5
Perfectly
59
Q

What is the counterfactual when talking about causality?

A

Causality can be established when the outcome on the same person, treated and untreated, is different. But only one of these events may be observed. The other is counterfactual.
- Counterfactual refers to treatment/no treatment simultaneously, and in the same population

How well did you know this?
1
Not at all
2
3
4
5
Perfectly
60
Q

What is the principle of exchangeability?

A
  • When the risk outcome for tx group is equal to the risk of the outcome in the control group had they been treated
How well did you know this?
1
Not at all
2
3
4
5
Perfectly
61
Q

What is the goal of a RCT?

A

Simulate the counterfactual comparison by having exchangeable groups thanks to randomization. Protects against the confounding, even if not measured or unknown, without having to collect tons of info.

How well did you know this?
1
Not at all
2
3
4
5
Perfectly
62
Q

3 prerequisites for RCT?

A
  • Clinical equipoise ($/ethics)
  • Modifiable exposure, and modifiable by the investigator
  • Common and early outcomes ($)
How well did you know this?
1
Not at all
2
3
4
5
Perfectly
63
Q

What is efficacy and effectiveness?

A

Efficacy: how well an intervention works under ideal conditions
Effectiveness: how well an intervention works in field conditions

How well did you know this?
1
Not at all
2
3
4
5
Perfectly
64
Q

What is the Hawthorne effect?

A

When watched people change their behaviour

How well did you know this?
1
Not at all
2
3
4
5
Perfectly
65
Q

What is the goal of the control group?

A
  • Control for Hawthorne effect

- Control for Secular trends

How well did you know this?
1
Not at all
2
3
4
5
Perfectly
66
Q

4 types of control?

A
  • Nothing
  • Placebo
  • Active Alternative
  • TAU
How well did you know this?
1
Not at all
2
3
4
5
Perfectly
67
Q

When do you use nothing as a control, and what problems does it cause?

A
  • When you have no accepted competitor to the treatment

- Participants are not blind, and no causal inference is possible

How well did you know this?
1
Not at all
2
3
4
5
Perfectly
68
Q

What is a placebo, and how useful is it for control groups?

A
  • Sensorily similar to experimental intervention, without the active ingredients
  • Good for blinding, allows straightforward conclusions
  • Good to assess side effects of tx
How well did you know this?
1
Not at all
2
3
4
5
Perfectly
69
Q

What is an active alternative, and how useful is it for control groups?

A
  • Another treatment (good for ethics and relevance)

- Often used for equivalence or noninferiority RCT, but conclusions drawn may be ambiguous

How well did you know this?
1
Not at all
2
3
4
5
Perfectly
70
Q

What is TAU, and why is it used, what are the problems?

A
  • Treatment as usual
  • For when the current practice varies and is hard to standardize
  • You get little control over what happens in this group
How well did you know this?
1
Not at all
2
3
4
5
Perfectly
71
Q

How can you protect internal validity? At what cost?

A

With tighter exclusion criteria - it decreases generalizability

How well did you know this?
1
Not at all
2
3
4
5
Perfectly
72
Q

You’ll need a bigger sample size if you want…(4 items)

A
  1. smaller a (more certainty)
  2. bigger 1-B (more power)
  3. be able to detect a smaller difference
  4. you have more variation in your sample
How well did you know this?
1
Not at all
2
3
4
5
Perfectly
73
Q

4 RCT designs?

A
  1. Two-arm (parallel) RCT
  2. Crossover RCT
  3. Factorial RCT
  4. Cluster-randomized RCT
74
Q

What is a crossover RCT? Assumptions?

A
  • Each participant is their own control, they get tx/placebo and get assessed, then there is a wash-out period, then they swith
  • Assumption: no residual carryover
75
Q

What is a factorial RCT? Assumptions?

A
  • When you want to look at the outcome of 2 distinct treatments, so you do a 2x2 design (possibility to test effect modification)
  • Assumptions: 1) outcomes for both tx are different; 2) modes of action are independent
76
Q

What is a cluster-randomized RCT? Concerns? What does it need?

A
  • When you randomize groups (towns, hospitals, schools) instead of individuals
  • Concerns: contamination (groups talking with each other)
  • Requires larger samples and fancy stats to handle the clustering of individual units within groups
77
Q

When is simple randomization good? What is the problem?

A

Good: for large samples, simple and easy
Problem:
- If small sample, equal chance criteria unmet
- Could lead to unequal number of participants in each arms, decreasing the power and increasing confounding

78
Q

When is block randomization good? What is the problem?

A
  • Good: to ensure balance between treatment arms, especially in smaller samples
  • Problems: increases predictability, which leads the way to selection bias (but you can vary the block size at random to palliate)
79
Q

When is stratified randomization good? What is the problems?

A

Good: to control for a covariate and ensure balance between arms (prevents type I error and improves power)

Problem:

  • all subjects must be identified before the assignment starts
  • many covariates may get complicated, so should use only those known to have a strong effect
80
Q

When is covariate adaptive randomization good? What is the problems?

A

Good: to balance covariates, when you don’t want to wait for all participants to be recruited before assigning
- Especially good for small samples with many covariates
Problems:
- only the first participant is truly randomly assigned
- predictability
- control only for known confounders

81
Q

What is allocation concealment?

A
  • Not knowing what the next assignment is going to be
  • ALWAYS feasible (both for subjects and investigators)
  • Prevents selection bias (sicker patients getting tx, for example)
82
Q

What is the problem with stopping a RCT early for benefit?

A

Size effects are often overestimated

83
Q

What are the advantages/problems of the intention-to-treat analysis compared to the as-treated analysis?

A
  • Advantage: Preserves exchangeability
  • Problem: Diluted treatment effect due to non compliance and unplanned cross-over

Bottom line:

  • Good for effectiveness, as it is a good reflection of the real world
  • Conservative estimate of efficacy
84
Q

2 problems of subgroup analyses

A
  1. Multiple comparison problem (more Type I error, need more stringent alpha)
  2. Lack of power (more Type II error, due to smaller n)

Bottom line: subgroup analyses should ideally be planned in advance, but it is not always possible

85
Q

Major differences between RCT and observational study?

A

Observational study:

  • Tx assignment no longer under control of investigator
  • Absence of random assignment, so cannot assume exchangeability of groups
86
Q

What is a cohort?

A

A study population, drawn from a source population, with goal to make inference to a reference population

87
Q

What is the main difference between prospective and retrospective cohort studies?

A

Prospective
- Exposure assessed at beginning of study or as it occurs during study
- Outcomes assessed in future
Retrospective
- Exposure assessed from historical records
- Outcomes assessed when study is begun

88
Q

What are ambispective cohorts?

A

Mixture of propsective and retrospective cohorts

89
Q

What is the difference b/w prospective and retrospective cohort studies in terms of non-differential exposure?

A

Prospective: nature forces non-differential exposure measurement
Retrospective: mask to induce non-differential exposure measurement

90
Q

Closed/stationary cohort?

A

Fixed membership - no one can enter, but people can leave by loss to follow-up and competing risks

91
Q

Open/dynamic cohort?

A

People can come in and come out

92
Q

If we say that the study design should follow the etiology - what do we mean?

A

Baseline should take place before the biological onset to assess exposure of interest

93
Q

What are fixed (vs time-varying) exposures?

A

Fixed: Stable regardless of the disease process (e.g., genotype, race, ever smoked)
Time-varying: varies within individuals (e.g., health behaviour, meds use, biomarker levels, SES)

94
Q

What is immortal time bias?

A

“Period of follow-up during which, by design, death or outcome cannot occur”

  • When entering into one group (e.g., oscar winners) is contingent upon survival
  • Time during which the participant is “immortal” should be considered as non-exposed (e.g., non oscar winner)
  • Excluding this time would also introduce bias (selection bias)
95
Q

Internal vs external validity in terms of population?

A

Internal validity: result from study population is equal to the true effect in the source population
External validity: result from study population is equal to true effect in reference population

96
Q

Formula for risk difference (RD)? Range?

A

= Risk1 - Risk0
*range = [-1, 1] (0 = null)

Reminder that risk = cumulative incidence = (#who develop disease during period)/(#at risk during period)

97
Q

Formula for relative risk/risk ratio (RR)? Range?

A

= Risk 1/Risk 0
*range = [0, infinity] (1 = null)

Reminder that risk = cumulative incidence = (#who develop disease during period)/(#at risk during period)

98
Q

Formula for incidence rate difference (IRD)? Range?

A

= rate1 - rate0
*range = [infinity, infinity] (0 = null)

Reminder that rate = (#who develop disease during period)/(total person time at risk during period)

99
Q

Formula for incidence rate ratio (IRR)? Range?

A

= rate1/rate0
*range = [0, infinity] (1 = null)

Reminder that rate = (#who develop disease during period)/(total person time at risk during period)

100
Q

Absolute vs relative measures?

A

Absolute: difference measures

Relative: ratio measures (RR, IRR) - may be very large even for exposures with little public health consequences

101
Q

What is incidence hazard?

A

Instantaneous failure (hazard) rate at time t given that individual has survived up to time t (may be derived from cumulative survival)

102
Q

What is a cumulative hazard? How is it measured?

A

Accumulation of hazard from time zero to time t

By Cox proportional hazards models

103
Q

As N increases, what happens to random error, precision, and bias/systematic error?

A

Random error: decreases
Precision: increases
Systematic error: stable

104
Q

What is selection bias?

A

Bias regarding the way in which the study participants have been selected

  • Differential selection of exposed vs non-exposed
  • may be immigrative (selection into study)
  • may be emigrative (selection out of study)
105
Q

What is information bias?

A

Bias regarding the way the study variables are measured

- Collecting different quality and extent of information from exposed and non-exposed groups

106
Q

What is confounding bias?

A

Incompletely controlled factors

107
Q

What happens if losses to follow-up are non-differential with respect to disease? And differential?

A

Non-differential: unbiased estimate

Differential: biased estimate

108
Q

4 ways to limit selection bias in cohort studies?

A
  1. Random draw from source population (for immigrative selection bias; by design)
  2. Limit loss to follow-up (for emigrative selection bias; by design)
  3. Contact sample of non-respondents to assess if differential and according to which characteristics (for emigrative selection bias; by design)
  4. Sensitivity analyses
109
Q

Synonym for information bias?

A

Measurement error

110
Q

Possible consequence from information bias?

A

Misclassification bias (erroneous info results in individual being placed in a different category)

111
Q

When is misclassification non-differential, vs differential?

A

Non-differential: errors on variable in question do not depend on value of other variables (e.g., disease status)

Differential: error on variable depend on value of other variables

112
Q

What would happen if there is information bias (non-differential/differential) without confounding and selection bias? With?

A

Without:

  • Non differential: toward the null
  • Differential: either direction

With:
- Anything goes

113
Q

7 ways to minimize information bias in cohort studies?

A
  1. Improve measurement properties of tools
  2. Repeat assessments
  3. Ensure parallel assessments in exposed vs unexposed groups
  4. Blind assessors
  5. Validate sub-samples
  6. Correct for known measurement error
  7. Sensitivity analyses for unknown measurement error
114
Q

3 properties of a confounder?

A
  1. Must be associated with exposure
  2. Must be associated with outcome
  3. Must not be an effect of the outcome or of the exposure
115
Q

Can the level of confounder vary over time?

A

Yes - you adjust with regular regressions

116
Q

What is confounding by indication?

A

When association between confounder and exposure is expected to be in a predictable direction (e.g., sicker individuals have more radical tx)

117
Q

5 ways to minimize confounding bias?

A
  1. Randomize
  2. Restrict study population
  3. Adjust (when confounders no affected by prior exposure)
  4. Causal methods (when time-varying confounders are affected by prior levels of exposure)
  5. Sensitivity analyses
118
Q

What are the two classification criteria of Pearce (2012) classification scheme?

A

1) outcome under study (incidence vs prevalence)

2) sampling based on the outcome or not

119
Q

What is a study base?

A

Either aggregate of total population-time in which cases occur OR members of source population during time period when cases are identified

120
Q

4 principles of the study base?

A
  1. Sample controls from study base in which cases arose
  2. Controls are proxy/sample for complete study base
  3. Controls should be representative of study base’s exposure distribution
  4. Controls should be selected independently of the exposure
121
Q

Can incidence case-control studies be conducted in a defined cohort or in a dynamic population?

A

Both.

  • In a defined/closed cohort, cases accrue over the course of follow-up in the exposed and unexposed subcohorts
  • In a dynamic population, in a steady state, the exposed and unexposed person-time is assumed to be relatively constant over time
122
Q

3 types of case-control studies conducted within a fixed cohort?

A
  1. Density/risk-set sampling (controls selected longitudinally from risk set available each time case arises)
  2. Cumulative incidence/exclusive sampling (controls from those who never experience outcome)
  3. Case cohort/inclusive sampling (controls from entire source population - those at risk at beginning of follow-up)
123
Q

Control sampling in density (nested) case-control studies?

A

Density sampling or risk-set sampling: Select controls longitudinally based on risk set available each time a new case arises
Goal: to represent the distribution of exposed person-time vs unexposed person-time in the source population

124
Q

Problem with density sampling?

A

Often infeasible because it relies on having info about everybody’s disease status, updated regularly

125
Q

In density sampling, is the risk set static or dynamic?

A

Dynamic - changes from one case to the next

126
Q

In density sampling, what influences the probability that an individual control will be selected?

A
  • Contribution in person-time at risk (the more time contributed, the more likely to be selected)
  • Eligible to be selected multiple times, as long as at risk (i.e., may serve as control for multiple cases)
127
Q

In density sampling, what happens if a control becomes a case?

A

Will be included in the study both as control and as a case

128
Q

Primary measure of association in case-control study?

A

Odds ratio

129
Q

Formula for odds ratio?

A

= [(E+D+)(E-D-)]/[(E-D+)(E+D-)]

130
Q

What formula should be true if controls are selected independently of the exposure?

A

E+D-/E-D- = PT1/PT0

131
Q

In incidence density case-control studies, the Odds Ratio is equals to…?

A

OR = Incidence Density Ratio = Rate1/Rate0

- If controls are selected independently of exposure

132
Q

Sampling method in Cumulative incidence (“exclusive”) case-control studies?

A

Controls are selected from those who do not experience the outcome at any point during the follow-up (survivors)

133
Q

Cumulative case control studies correspond to which cohort studies?

A

Those that follow a closed cohort and measure risks

134
Q

What is the main problem with cumulative case control studies?

A

Controls selected from the survivors do not represent the experience of the entire source population, because it ignores the contribution of cases

135
Q

In cumulative case control studies, the Odds ratio estimates…

A

OR approximates Risk Ratio, and ONLY FOR RARE DISEASES (< .05)

136
Q

Why is the rare disease assumption necessary for cumulative case control studies?

A

If the disease is rate, the experience of cases will be a very small part of the overall experience of the source population

137
Q

What happens if you don’t satisfy the rare disease assumption in cumulative case control studies?

A

The odds ratio will be an overestimate of the risk ratio

138
Q

What is the sampling method in case-cohort (or inclusive sampling) studies?

A

Controls are selected from the entire source population (those at risk - disease free - at beginning of follow-up)

139
Q

What happens to the probability that a participant will be included as control in case-cohort studies?

A
  • Every participant has an equal chance of being included as a control
140
Q

Why is there no need for the rare disease assumption in case-cohort studies?

A
  • Controls are selected from the entire source population, so the distribution of exposure is equal to the source population
141
Q

In case-cohort studies, odds ratio is equal to…

A

Risk ratio

142
Q

How are controls to be selected in dynamic case-control studies?

A

If steady state with constant level of exposure over study period: estimate distribution of PT1 and PT0 in the source population from a representative sample of controls without the outcome

If no steady state: controls sampled at midpoint of the study period will represent average distribution of PT0 and PT1, and OR=IDR

143
Q

2 types of study bases?

A
  1. Primary study base = fixed cohort
    - Base defined by experience to be investigated, usually well-defined
    - Cases are subject within the study base who develop disease, and all cases are identifiable but not necessarily used
  2. Secondary study base = dynamic population
    - Cases are defined before study base is identified
    - Study base is defined as source of the cases (controls are those that would have been recognized as cases had they develop the disease)
144
Q

Pros/cons of primary vs secondary study base

A
  • Primary study bases are easier to sample for controls (bc base is well defined), but otherwise limited
  • Hard to control sample for secondary study bases (who would have become cases had they developed the disease?)

Overall, secondary study bases are more practical and more common

145
Q

What is a population roster/sampling frame?

A
  • Census lists, birth certificates, electoral rolls, etc.

- Makes sampling easier

146
Q

3 other approaches when no roster is available?

A
  • Random digit dialing
  • Neighborhood controls
  • Hospital controls
147
Q

Optimal number of controls per case?

A
  • Usually 1:1, but you gain precision (not validity) for up to 1:4
  • Beyond that, marginal added value
148
Q

What happens to exchangeability in case-crossover design?

A

Since each case serves as their own control, the confounding by stable and slow-varying characteristics is eliminated (measured or not), thus increasing exchangeability

149
Q

Where is the control period in the unidirectional case-crossover?

A

Before

150
Q

When is bidirectional case-crossovers acceptable?

A

When outcome cannot influence subsequent exposure

151
Q

What happens in case-crossovers to cases with identical exposures (concordant pairs) during analysis?

A

They do not contribute

152
Q

What is the measure used in case-crossovers and what is estimated?

A
  • Exposure odds ratio

- Unbiased estimate of the IRR

153
Q

What are cross-sectional studies? Synonym?

A

Synonym: prevalence studies

Snapshot of the distributions of exposure and outcome in a population

154
Q

When are cross-sectional studies practical?

A
  • Disease onset are to measure
  • Lengthy follow-up
  • Not enough resources
155
Q

What are the uses of cross-sectional studies?

A

Public health, burden of disease

156
Q

Most common measure used for cross-sectional studies?

A

Odds ratio, unless the outcome is common (>10%) then best to estimate prevalence ratio

157
Q

Formula of Prevalence odds ratio for cross-sectional studies

A

= [(E+D+)(E-D-)]/[(E-D+)(E+D-)]

158
Q

Formula for prevalence ratio in cross-sectional studies?

A

= [E+D+/all E+]/[E-D+/all E-]

159
Q

Prevalence ratio is an approximation of

A

IRR

160
Q

What is the ecological fallacy?

A

Apply population data to draw conclusions on the individual level

161
Q

What are multilevel studies?

A

Any study with multilevel or nested data (e.g., individuals clustered within areas or groups)

162
Q

Common example of a multilevel study?

A

The effect of ecological exposure on individual-level outcome

163
Q

What is cross-level interaction?

A

When individual-level exposures vary depending on class-level exposure

164
Q

What is a measure of potential impact?

A

An estimate of the effect of removing the exposure

165
Q

What are the two measures of potential impact at the individual level?

A
  • Attributable risk (risk difference)

- Attributable fraction

166
Q

What are the two measures of potential impact at the population level?

A
  • Population attributable risk

- Population attributable fraction

167
Q

What is background risk?

A

Incidence not due to exposure, or risk in the unexposed group.
- explained by the fact that there can be several sufficient causes for a given disease (i.e. consistent with multicausality)

168
Q

Formula for attributable risk?

A 
AR = risk1 - risk0 
(= to risk difference)
169
Q

Assumptions to attributable risk:

A
  • Effect of exposure is causal

- Sum does not add up to 100% due to interaction

170
Q

Interpretation of attributable risk, e.g. of 0.7%:

A

Among 1000 babies who regularly sleep prone, there are 7 excess cases of SIDS attributable to prone sleeping

171
Q

What is the attributable fraction?

A

Proportion by which the incidence rate of the outcome among those exposed would be reduced if the exposure were eliminated

172
Q

Formula for attributable fraction?

A

= (risk1 - risk0)/risk1
= AttribRisk/risk1
= (RR-1)/RR

173
Q

What is the interpretation of attributable fraction, e.g. of 0.67:

A

Among the prone sleeping babies, 67% of the cases of SIDS are attributable to the prone sleeping posture

174
Q

What is the population attributable risk?

A

How much of the burden of disease is due to the exposure in the entire population?

175
Q

Formula for population attributable risk?

A

= risktotal - risk0

= AR*probability of exposure

176
Q

Interpretation for population attributable risk, if 2.4 per 1000?

A

Among every 1000 babies in a population, there are 2 excess cases of SIDS attributable to prone sleeping (if all babies were made to sleep on their backs, then 2 SIDS cases can be averted for every 1000 babies

177
Q

Goal and assumptions of population attributable risk?

A

Goal: provide measure of public health impact of exposure on entire population (answers to question about value of prevention)
Assumptions
- Effect of exposure on outcome is causal
- Depends on prevalence of exposure and strength of the effect

178
Q

What is the population attributable fraction?

A

The proportion by which the incidence rate of the outcome in the entire population would be reduced if the exposure were eliminated (proportion of disease in pop attributable to exposure)

179
Q

Formula for population attributable fraction?

A

= PAR/risk total

= [ExpPrev(RR-1)]/[ExpPrev(RR-1) + 1]

180
Q

Interpretation of population attributable fraction, e.g., 41%

A

Making all babies sleep on their back would eliminate 41% of all cases of SIDS in the population

EPIB 601 Fundamentals (6 decks)

Brainscape helps you reach your goals faster, through stronger study habits.
© 2024 Bold Learning Solutions. Terms and Conditions