PUBH 6003 Final Important Terms and PUBH 6003 Weeks 1-10 exam review Latest Update 2024-2025 with 350+ Questions and Verified Correct Answers Guaranteed A+
- criteria for confounding - CORRECT ANSWER: associated with exposure, associated
with outcome, not on causal pathway
- methods to assess for confounding - CORRECT ANSWER: 1. evaluate the 3 criteria
- stratification
- compare adjusted to crude models
antagonistic effect - CORRECT ANSWER: the effect of the two factors combined is
less than would be expected based on the combination of their independent individual effects
attributable risk among exposed - CORRECT ANSWER: excess risk of disease due to
the exposure among those who are exposed
berkson bias (a type of *selection* bias) - CORRECT ANSWER: controls are selected from a hospital population
association between exposure and disease weakens
hospital controls do not represent the prevalence of exposure in the community from which cases of the disease arise
calculating reliability - CORRECT ANSWER: Cohen's kappa or percent agreement
Characteristics of a good screening test - CORRECT ANSWER: simple, rapid,
inexpensive, safe, acceptable
Controlling confounding through analysis - CORRECT ANSWER: stratification,
adjustment
Controlling confounding through design - CORRECT ANSWER: randomization,
restriction, matching
Controlling Selection Bias - CORRECT ANSWER: it cannot be fixed once it occurs, but you can design and conduct your study in ways that will minimize the likelihood of
selection bias:
- use similar criteria for selecting cases and controls (and exposed and unexposed) 1 / 4
2. for hospital-based controls: consider diagnostic and referral patterns
- develop protocols to ensure high participation/ follow-up rates
diagnostic versus screening tests - CORRECT ANSWER: diagnostic tests confirm the disease while screening tests look for the diseases regardless of symptoms
ex: mammograms are screening
direct/ proximal cause - CORRECT ANSWER: cause is sufficient and the last step in a causal chain
Directions of confounding - CORRECT ANSWER: negative confounding- toward the
null
positive confounding- away from the null
Disability Adjusted Life Year (DALY) - CORRECT ANSWER: A measure of burden of
disease, one DALY equals one year of healthy life lost due to premature death and time lived with illness, disease or injury.
Diseases appropriate for screening - CORRECT ANSWER: the disease has to be an
important public health in seriousness and magnitude, must have a detectable preclinical phase, disease must be treatable/ prevention must be key, early treatment is better than late treatment
Effect Modification (Interaction) - CORRECT ANSWER: impact of the exposure is
noticeably different based on the presence of a third variable
need to report data as stratified by presence/absence of third variable
Examples of screening tests - CORRECT ANSWER: PPD for tuberculosis, Beck
Depression Inventory for Depression, Pap smear for precancerous lesions, mammography for breast cancer
Hawthorne effect - CORRECT ANSWER: A change in a subject's behavior caused
simply by the awareness of being studied
healthy worker effect (a type of *selection* bias) - CORRECT ANSWER: observation that employed populations tend to have a lower mortality experience than the general population
If there is confounding, what risk estimate do we present? - CORRECT ANSWER:
since ______ is a confounder, we should not present the crude estimates as it is not correct. 2 / 4
we should adjust for ____ (sex, race, etc.) in the analysis and present an adjusted estimate. If you don't have the adjusted estimate, you could present the strata specific estimates, but the adjusted is preferred.
If there is effect modification, what risk estimate do we present? - CORRECT ANSWER: Present the strata specific estimates because we would not want to adjust for sex in the analysis. Since this is a real effect we want to describe it.
If there is no confounding, what risk estimate can we present? - CORRECT ANSWER: Crude estimate. Or adjusted if it is not exactly the same.
indirect/ distal - CORRECT ANSWER: cause alone is not sufficient, but is in a causal chain that results in a causal sequence that is sufficient
necessary but not sufficient
information bias - CORRECT ANSWER: an error that occurs when the methods for
obtaining data about study subjects are inadequate and some participants are incorrectly classified on their exposure or outcome status
interpretations: predictive values - CORRECT ANSWER: these interpretations are important for the patient. this will help the patient interpret the likelihood of disease given the test result.
these values can change with the prevalence of disease in the population with higher prevalence populations resulting in higher PV+ but slight decreases in PV-
interpretations: sensitivity and specificity - CORRECT ANSWER: these interpretations are important for public health.
only changing the TEST (cut-offs, procedures, etc.) can alter these
interviewer bias - CORRECT ANSWER: intentional or unintentional influence exerted by an interviewer in such a way that the actual or interpreted behavior of respondents is consistent with the interviewer's expectations
systematic difference in soliciting, recording, interpreting information
lead time bias - CORRECT ANSWER: Bias introduced when screening detects a
disease earlier and thus lengthens the time from diagnosis to death.
example: "if we compare survival time from the point of diagnosis, the subject whose disease was identified through screening appears to survive longer, but only because their disease was identified earlier"
- / 4
Length bias sampling - CORRECT ANSWER: screening tests preferentially identifies those with longer preclinical phase of disease
comparing outcomes identified by screening vs. usual means (better outcomes in screened groups)
Example: overestimation of survival duration among screening-detected cases caused by the relative excess of slowly progressing cases. These cases are disproportionately identified by screening because the probability of detection is directly proportional to the length of time during which they are detectable.
Mass screening - CORRECT ANSWER: screening on a large scale of total population
groups regardless of risk status
measures of association for bias - CORRECT ANSWER: risk difference
Attributable risk among exposed
population attributable risk
Minimizing information bias - CORRECT ANSWER: use objective, calibrated, validated, standardized assessment tools
maximize follow-up in all comparison groups to avoid losing data
minimize missing data in all comparison groups
blind study staff to exposure and/or disease status
**non-differential misclassification is preferable to differential misclassification**
multiplicative interactions - CORRECT ANSWER: assumes two exposures increase
risk synergistically
much more common in literature and facilitated by our statistical tools
necessary cause - CORRECT ANSWER: a condition that must be present for the effect to occur
without Factor A, disease will never develop
neyman bias - CORRECT ANSWER: incidence-prevalence bias
association is underestimated
- / 4