Use LEFT and RIGHT arrow keys to navigate between flashcards;
Use UP and DOWN arrow keys to flip the card;
H to show hint;
A reads text to speech;
61 Cards in this Set
- Front
- Back
independent variable
|
Variables that define study groups
|
|
dependent variable
|
Outcome measures of interest
|
|
Descriptive vs. Analytic studies
|
Descriptive: describe something
you observe and describe what you see Analytic: manipulate something to study it |
|
Observational study
|
simply observe the occurrence of the outcome among the determined study groups
No manipulation of group characteristics (no intervention) --I'm not telling you to do anything, just look at what naturally occurs |
|
Interventional
|
manipulate exposures/group characteristics and follow study population for outcome of interest
must be PROSPECTIVE! you have influenced the population |
|
for something to be causal, what is a key point?
|
the exposure or characteristic has to occur BEFORE the outcome
|
|
what are the 2 observational study types used for generating hypotheses?
|
Ecological Studies
Cross-sectional studies |
|
please list the 4 observational studies
|
Ecological studies
Cross-sectional studies Case-control studies Cohort studies |
|
please list interventional study
|
Randomized controlled clinical trials (RCCTs)
|
|
what are ecological studies
|
tells us about the population
examine characteristics (risks/exposures, the independent variables) of a sample of an entire population as well as the presence of an outcome of interest (cases/disease, the dependent variable(s)) in the same sample of the entire population |
|
What are the advantages of ecological studies?
|
Generate hypotheses
Quick; easy; inexpensive |
|
what do ecological studies measure?
|
outcome frequency (prevalence; incidence rates)
|
|
What are disadvantages of ecological studies
|
No temporal relationships no causality
Cannot test hypotheses Ecological fallacy Late-look bias |
|
What is the ecological fallacy?
|
tendency to over-interpret the findings of an ecological study and ascribe characteristics of the sample population to specific individuals in the population. For example, as an extension of what we looked at in slide #8, we might be tempted to say that women who consume more dietary fat are more likely to develop breast cancer
Seen in ecological studies |
|
What is the late look bias?
|
Seen in ecological studies
|
|
What is a cross-sectional study?
|
Prevalence of disease and association of disease with risk factor(s) at one point in time
|
|
Can a cross-sectional study address causality?
|
NO
|
|
can a cross-sectional study determine incidence?
|
NO
|
|
do you know the temporal relationship between exposure and outcome in a a cross-sectional study?
|
No!
|
|
so what does a cross-sectional study measure?
|
prevalence
|
|
what are the advantages of a cross-sectional study
|
Can examine possible association between risk and outcome
Relatively quick & easy; inexpensive Can generate hypotheses |
|
what are the DISadvantages of a cross-sectional study
|
Cannot assess causation
Late-look bias Not good for hypothesis testing |
|
what do you use to test hypotheses about associations and possible causation btw exposure and outcome?
|
control groups
|
|
describe case-control studies
|
The study design begins with people with disease (cases, or outcome) and compares them to people without disease (controls, or outcome)
|
|
Can you measure incidence with case-control studies? what about prevalence?
Why? |
NEITHER!
Investigator determines the number of cases and the number of controls that will be compared Disease is not just “naturally occurring in the study population |
|
case-control studies occur in what time-frame?
|
Retrospective
|
|
can case-control studies asses many risk factors for a single disease?
|
YES!
|
|
can case-control studies examine temporal relationships btw exposure and outcome?
|
YES!
|
|
can case-control studies show causality?
|
possibly,
because you see the exposure come before the outcome |
|
What are the challenges of case-control studies
|
Selection of controls (matched controls, hospitalized vs. non-hospitalized, problem of "overmatching"-dilute the ability to detect changes)
Recall limitations Recall |
|
measurement of case-control studies?
|
proportions, OR (odds of exposure ratio)
Can use OR to estimate risk in certain circumstances … what are these circumstances? (see later slide) |
|
disadvantages of case-control studies
|
Recall bias
Challenge of selecting appropriate controls Cannot measure incidence or prevalence |
|
advantages of case-control studies
|
Can test hypotheses and examine causality
Can look at many risks for one disease Relatively short; less expensive Can more easily study conditions of low incidence or prevalence (good for studying rare diseases) |
|
******
Which test is good for studying rare diseases? ON TEST |
case-control studies
don't have to wait for people to develop the disease, you just find those that have it |
|
please describe cohort studies
|
Study groups are arranged based upon exposure status and then followed for development of the outcome
Outcome: disease incidence rates, mortality rates, etc. Can evaluate incidence (unlike case-control studies) Incidence rate is similar to risk (proportion of those exposed who develop disease) |
|
can you measure incidence with cohort studies?
|
yessir!
|
|
Are cohort studies prospective or retrospective?
|
Can be either!
|
|
is there randomization in cohort studies?
|
no!
|
|
must know how to calculate odds and relative risk still
|
check out slide 29
|
|
when is odds ratio a good sign of relative risk?
|
when the disease does not occur frequently
this was on the last test |
|
what do Cohort studies measure?
|
incidence rates and risks; relative risk
|
|
cohort study: retrospective or prospective?
|
either!
|
|
Disadvantages of cohort prospective
|
time-consuming and costly; losses to follow-up; can study only the risk measured at the beginning of the study
|
|
Disadvantages of cohort Retrospective
|
recall limitations and bias
|
|
Disadvantages of cohort in general (both retro and pro)
|
Can only be used for common diseases
|
|
what are the advantages of cohort studies? main advantage?
|
Good at studying RARE RISK
Flexible: retrospective or prospective Can use to determine actual measure of risk Can study many disease outcomes Good for studying rare risk factors |
|
what is a Nested Case-Control Study?
|
Case-control study within a cohort study
(start with propective cohort study and take baseline measurements of characteristics. as the study moves forward through time cases develop as you are following them. as these cases develop, you match them with controls cases and controls can be compared wtih respect to a variety of exposures A way to enhance data collection and maximize cost-effectiveness in a prospective study |
|
what do nested case control studies start with?
|
a prospective cohort study
|
|
what do interventional studies test?
what is the one example we have? |
hypotheses
Clinical trials |
|
What is the gold standard for studying interventions?
|
the RCCT (randomized control ed clinical trial)
|
|
What does the randomization do in the RCCT?
|
lowers the bias
|
|
can RCCT be retrospective?
|
shit naw
|
|
What is done to the exposure in RCCT? how is this different from other studies
|
In RCCT the exposure is manipulated
in other studies, things are just observed |
|
RCFT (randomized controlled field trials) main goal is what? what bout RCCT?
|
RCFT focuses on preventive interventions
RCCT focuses on therapeutic interventions |
|
what is a single blind study?
|
subjects do not know to which group they have been randomized
Group identity (experimental or non-experimental) is unknown half get drug A, half get placebo subjects don't know what they are getting |
|
what is a double blind study?
|
both subjects and investigators are blinded to group identity
half get drug A, half get placebo the experimenter and subjects dont know what group ppl are in |
|
Stratification of group independent variables are ways of enhancing effects of randomization, how does it do this?
|
split up groups that could be potential confounders (take 1000 pts, can stratify by gender, age, etc)
|
|
what are the advantages of RCCT cross-over design?
|
increases the amount of data collected
allows for more comparisons between groups |
|
measurement of RCCT?
|
incidence, prevalence, RR
|
|
disadvantages of RCCT? probably on the test
|
**External validity – can the results be effectively generalized to a broader population?
Most expensive and time consuming |
|
Advantages of RCCT
|
Most scientifically rigorous
**Can be least subject to bias (double-blind studies) |