Module 11

Question 1

Evaluating epi associations (key questions 1-3)

Answer 1

1. Could association have been observed by chance?
   - Determined through statistical tests
2. Could association be due to bias?
   - Bias refers to systematic errors (ex: sample selection, data analysis)
3. Could other confounding variables have accounted for the observed relationship?

Question 2

Evaluating epi associations (key questions 4-5)

Answer 2

4. To whom does this association apply?
   - Representativeness of sample
   - Participation rates
5. Does the association represent a cause-and- effect relationship?
   - Considers criteria of causality.

Question 3

Statistical power

Answer 3

The ability of a study to demonstrate an association if one exists. Determined by:
– Frequency of the condition under study
– Magnitude of the effect
– Study design
– Sample size

Question 4

How to know study is valid

Answer 4

Eliminate alternative explanations
o Bias (systematic error)
o Confounding
o Random error

Question 5

Internal validity

Answer 5

The appropriate measurement of exposure, outcome, and association between exposure and disease
– Proper selection of study groups
– Lack of error in measurement

Question 6

External validity

Answer 6

The ability to generalize beyond a set of observations to some universal statement

Question 7

Random errors

Answer 7

Reflect fluctuations around a true value of a parameter because of sampling variability. Contributing factors:

1. Poor precision
2. Sampling error
3. Variability in measurement

Question 8

Poor precision

Answer 8

• Occurs when the factor being measured is not measured sharply
• Precision can be increased by increasing sample size or the number of measurements

Question 9

Sampling error

Answer 9

• Occurs when the sample selected is not
  representative of the target population
• Increasing the sample size can reduce the
  likelihood of sampling error

Question 10

Variability in measurement

Answer 10

The lack of agreement in results from time to time reflects random error inherent in the type of measurement procedure employed

Question 11

Factors that contribute to systematic errors

Answer 11

1. Selection bias
2. Information bias
3. Confounding

Question 12

Bias

Answer 12

Systematic error that leads to incorrect/invalid estimate of association (easier to avoid than to remove or fix)
Types: selection, information

Question 13

How to evaluate for bias

Answer 13

o Identify source
o Estimate magnitude
o Assess direction

Question 14

Selection bias

Answer 14

Error due to systemic differences in characteristics between those selected for study and those not
o Case control—if different criteria related to exposure
o Retrospective cohort—if selection of exposed or unexposed group related to outcome

Question 15

Preventing selection bias

Answer 15

o Define study population independent of disease not after cases appear (prior to follow-up)
o Get same information from cases and controls
o Don’t let disease influence the availability of information
o Don’t let disease influence loss of subjects to follow-up

Question 16

Information bias

Answer 16

The means of obtaining information about subjects is
inadequate/ incorrect. Sources:
1. Misclassification bias
2. Bias in abstracting records
3. Bias in interviewing
4. Bias from surrogate interviews
5. Surveillance bias
6. Recall bias
7. Reporting bias

Question 17

Confounding

Answer 17

Mixing of effects between exposure, outcome,
and third variable (confounder)
Factors other than exposure that ↑↓ risk of disease

Question 18

Criteria for confounders

Answer 18

To be a confounder, an extraneous factor must satisfy the following criteria:

1. Be a risk factor for the disease
2. Be associated with the exposure
3. Not be an intermediate step in the causal path between exposure and disease

Question 19

Confounding magnitude and direction

Answer 19

Magnitude of confounding = (RR crude – RR adjusted)/RR adjusted
o Size bias depends on degree of association
o Effect of multiple confounders may be large
Direction of confounding
o Exaggerate (positive confounding)
o Hide (negative confounding)

Question 20

Methods to control confounding

Answer 20

Prevention strategies—attempt to control confounding through the study design itself
In designing the study: Individual and Group matching
In analysis of data: Stratification and Adjustment

Question 21

Matching

Answer 21

Matches subjects in the study groups according to the value of the suspected/known confounding variable to ensure equal distributions.

• Frequency matching: the # cases with particular match characteristics is tabulated (Group matching)
• Individual matching: the pairing of one or more controls to each case based on similarity in sex, race, or other variables (matched pairs)

Question 22

Analysis strategies to control confounding

Answer 22

1. Stratification: analyses performed to evaluate the effect of an exposure within strata (levels) of the confounder
2. Multivariate techniques: use computers to construct mathematical models that describe simultaneously the influence of exposure and other factors that may be confounding the effect

Question 23

Risk factors

Answer 23

Exposure that is associated with a disease.

Due to the uncertainty of “causal” factors the term risk factor is used.

Question 24

Criteria for risk factors

Answer 24

1. The frequency of the disease varies by category or value of the factor (ex: light vs. heavy smokers)
2. The risk factor precedes onset of the disease
3. The observation must not be due to error

Question 25

Koch’s Postulates for disease causation

Answer 25

Criteria to prove an organism caused disease:
– The organism is always found with the disease
– The organism is not found with any other disease
– The organism isolated from the diseased host can produce disease when introduced to another susceptible host
*Not as useful for non-infectious diseases

Question 26

Modern concepts of causality

Answer 26

• 1964 Surgeon General’s Report - Five criteria for causality: strength of association, time sequence, consistency upon repetition, specificity, coherence of explanation
• Sir Austin Bradford Hill expanded list to include: biologic gradient, plausibility, experiment, and analogy