NTP 677 Exam 2 Study Guide
Epidemiology - Answer The study of thee distribution and determinants of health and
disease in human populations
What subset of epidemiology studies the distribution of health and disease? - Answer
Descriptive epidemiology
What subset of epidemiology studies thee determinants of health and disease? - Answer
Analytic or etiologic epidemiology
What are clues that you might be looking at an epidemiology study? - Answer The
phrase "observational study"
"in a national sample"
Descriptive epidemiology - Answer How does a disease vary over person, place, and
time?
Useful for investigating the prevalence and incidence of diseases, exposures, and
health behaviors. Important for policy-making, resource allocation, planning, etc.
Measuring prevalence and incidence can give us clues into causes of disease
Analytic epidemiology - Answer Examines associations between "exposures" and
"diseases" in hope of gaining insight into causal relationships
Describe one example of an analytic study design to answer the question: doe mild to
moderate sleep apnea cause elevated blood pressure in humans? If so, to what degree?
- Answer If there were no ethical concerns, we could
Experimentally manipulate how much sleep apnea people were experiencing and put
them into different groups
Name the three basic epidemiological study designs - Answer Cross-sectional
Case-control
Cohort
Describe a cross-sectional study and list the advantages and disadvantages - Answer
Exposure and disease are assessed simultaneously
Advantage: does not take much time
Disadvantage: can not identify which variable came first
Describe a case-control study and list advantages and disadvantages - Answer D
assessed, then E (concurrently or retrospectively)
, Advantage: efficient for rare diseases
Disadvantage: knowledge of disease status might reduce validity
Describe a cohort study and list advantages and disadvantages - Answer Tracks people
over time. All participants begin with no D
Advantages: able to get rates out of this, E precedes D
Disadvantages: large, expensive, time consuming
Describe a valid measurement - Answer "on average" correctly "measures what it
purports to measure"
What is the opposite of a valid measurement - Answer Biased measurement
Prevalence - Answer Proportion of a population found to have a condition
Incidence rate - Answer Frequency of events in a defined population per unit of
person-time at risk
Risk - Answer probability of developing disease over a defined time interval
Odds - Answer risk/(1-risk) or prevalence/(1-prevalence)
Information bias - Answer Bias due to measurement error
Non-differential measurement error - Answer The degree of error in measuring E (or D)
does not depend on D (or E)
Differential measurement error - Answer The degree of error in measuring E depends on
D (or vice versa). More problematic
Selection bias - Answer A systematic distortion of the association between exposure
and disease
What percent of genes were shown to change based on behavioral state (spontaneous
wake vs sleep) - Answer Anywhere from 1-10%
Sleep genes - Answer Genes that are expressed at higher levels in sleep groups relative
to both wake conditions
Wake genes - Answer Genes that are upregulated in sleep deprivation and waking
groups relative to sleep groups
Which functional categories of genes are correlated to sleep states? - Answer Protein
synthesis
Synaptic depression
Lipid metabolism
Epidemiology - Answer The study of thee distribution and determinants of health and
disease in human populations
What subset of epidemiology studies the distribution of health and disease? - Answer
Descriptive epidemiology
What subset of epidemiology studies thee determinants of health and disease? - Answer
Analytic or etiologic epidemiology
What are clues that you might be looking at an epidemiology study? - Answer The
phrase "observational study"
"in a national sample"
Descriptive epidemiology - Answer How does a disease vary over person, place, and
time?
Useful for investigating the prevalence and incidence of diseases, exposures, and
health behaviors. Important for policy-making, resource allocation, planning, etc.
Measuring prevalence and incidence can give us clues into causes of disease
Analytic epidemiology - Answer Examines associations between "exposures" and
"diseases" in hope of gaining insight into causal relationships
Describe one example of an analytic study design to answer the question: doe mild to
moderate sleep apnea cause elevated blood pressure in humans? If so, to what degree?
- Answer If there were no ethical concerns, we could
Experimentally manipulate how much sleep apnea people were experiencing and put
them into different groups
Name the three basic epidemiological study designs - Answer Cross-sectional
Case-control
Cohort
Describe a cross-sectional study and list the advantages and disadvantages - Answer
Exposure and disease are assessed simultaneously
Advantage: does not take much time
Disadvantage: can not identify which variable came first
Describe a case-control study and list advantages and disadvantages - Answer D
assessed, then E (concurrently or retrospectively)
, Advantage: efficient for rare diseases
Disadvantage: knowledge of disease status might reduce validity
Describe a cohort study and list advantages and disadvantages - Answer Tracks people
over time. All participants begin with no D
Advantages: able to get rates out of this, E precedes D
Disadvantages: large, expensive, time consuming
Describe a valid measurement - Answer "on average" correctly "measures what it
purports to measure"
What is the opposite of a valid measurement - Answer Biased measurement
Prevalence - Answer Proportion of a population found to have a condition
Incidence rate - Answer Frequency of events in a defined population per unit of
person-time at risk
Risk - Answer probability of developing disease over a defined time interval
Odds - Answer risk/(1-risk) or prevalence/(1-prevalence)
Information bias - Answer Bias due to measurement error
Non-differential measurement error - Answer The degree of error in measuring E (or D)
does not depend on D (or E)
Differential measurement error - Answer The degree of error in measuring E depends on
D (or vice versa). More problematic
Selection bias - Answer A systematic distortion of the association between exposure
and disease
What percent of genes were shown to change based on behavioral state (spontaneous
wake vs sleep) - Answer Anywhere from 1-10%
Sleep genes - Answer Genes that are expressed at higher levels in sleep groups relative
to both wake conditions
Wake genes - Answer Genes that are upregulated in sleep deprivation and waking
groups relative to sleep groups
Which functional categories of genes are correlated to sleep states? - Answer Protein
synthesis
Synaptic depression
Lipid metabolism