Name: Score:
221 Multiple choice questions
Term 1 of 221
what makes it a case-control vs. cohort study?
As a first step when searching for a cause of an
adverse health outcome
After finding an association in a case -control study, next step is to carry out a cohort study
to further elucidate the relationship.
1.The difference in response rates to be detected.
2.An estimate of the response rate in one of the groups
3. Level of statistical significance (alpha)
4.The value of the power desired (1-beta)
5.Whether the test should be one-sided or two-sided
The two study designs are distinguished by whether the study
begins with diseased & non- diseased people (case -control study) orwith exposed & non-
exposed people (cohort study),
not calendar time.
Hybrid of the case-control & cohort study designs
Nested case-control studies
Case-cohort studies
Both offer substantial improvements over traditional case-
control
,Term 2 of 221
us preventative services task force grades: recommendations...
chapter 14 bottom of slide
appendix a of the report
summary section of chapter 5
introduction of chapter 3
Term 3 of 221
how is the number of cases determined?
The number of cases is determined by the ratio of controls per
case, & this proportion is determined by the investigator; it does not reflect the true
prevalence of the disease in the population
because randomization doesnt guarantee variability in age..
and it is possible that all old peeps are put into the same group = all high risk..
Poses a serious challenge in analyzing data with no
perfect solution
Current practice is to perform the primary analysis
by intention to treat-according to the original randomized assignment
The odds ratio is not a good estimate of the relative risk when a
disease is not infrequent
Definition 4 of 221
Type 2 error when there really is a difference but on the basis of our samples, we decide that
there is no difference (designated by beta -β)
statistical power (1-β)
type 2 error (-B)
confidence interval (ci)
type 1 error (α)
,Term 5 of 221
what would you expect to result from a dichotomous example?
This factor alone can produce the disease, but so can other factors that are acting alone
factor a or b or c = disease
Radiation/benzene exposure alone-> leukemia
Radiation/benzene exposure ≠> leukemia
Other cofactors (immune status, translocations)-> leukemia
for evaluating new drugs and other treatments of
disease, including tests of new health and medical
care technology
to assess new programs for screening and early
detection
to assess new ways of organizing and delivering health
services.
Tells us how many times more likely it is that an event will occur in the treatment group
relative to the control group.
RR of 1-- no difference between the 2 groups thus, treatment
had no effect.
RR< 1 -- treatment decreases risk of outcome.
RR> 1 -- treatment increased risk of outcome
In a dichotomous (exposed or not exposed) example, if exposure is associated with
disease, we would expect the proportion of the cases who were exposed to be greater than
the proportion of the controls who were exposed
, Definition 6 of 221
in this design, the controls are a sample of individuals who are at risk for the disease at the time
each case of the disease develops
The cases & controls are, in effect, matched on calendar time & length of follow -up
Because a control is selected each time a case develops, a control who is selected early in the
study could later develop the disease & become a case in the same study.
Typical sequence of approach: study etiology
what are Nested case-control studies?
Cohort Studies Advantages
how do you interpreting the risk?
Term 7 of 221
recall bias
First, select cases with the disease & controls without the disease
Second, measure past exposure; interviews, review medical/employee records, or results of
chemical/ biologic assays of blood, urine, or tissues.
Is randomization ethical (withholding new treatment from the control group)?
Yes-only when we don't know whether drug A is better than drug B.
Is it ethical to withhold effective treatment (e.g., when using a
placebo)?
Trade -off between randomizing & not randomizing
.
arises in any situation in which individuals select themselves into a group (by purposeful
intent) causing a biased sample with abnormal or undesirable conditions in the group.
recall bias is a systematic error caused by differences in the accuracy or completeness of
the recollections retrieved ("recalled") by study participants regarding events or
experiences from the past
221 Multiple choice questions
Term 1 of 221
what makes it a case-control vs. cohort study?
As a first step when searching for a cause of an
adverse health outcome
After finding an association in a case -control study, next step is to carry out a cohort study
to further elucidate the relationship.
1.The difference in response rates to be detected.
2.An estimate of the response rate in one of the groups
3. Level of statistical significance (alpha)
4.The value of the power desired (1-beta)
5.Whether the test should be one-sided or two-sided
The two study designs are distinguished by whether the study
begins with diseased & non- diseased people (case -control study) orwith exposed & non-
exposed people (cohort study),
not calendar time.
Hybrid of the case-control & cohort study designs
Nested case-control studies
Case-cohort studies
Both offer substantial improvements over traditional case-
control
,Term 2 of 221
us preventative services task force grades: recommendations...
chapter 14 bottom of slide
appendix a of the report
summary section of chapter 5
introduction of chapter 3
Term 3 of 221
how is the number of cases determined?
The number of cases is determined by the ratio of controls per
case, & this proportion is determined by the investigator; it does not reflect the true
prevalence of the disease in the population
because randomization doesnt guarantee variability in age..
and it is possible that all old peeps are put into the same group = all high risk..
Poses a serious challenge in analyzing data with no
perfect solution
Current practice is to perform the primary analysis
by intention to treat-according to the original randomized assignment
The odds ratio is not a good estimate of the relative risk when a
disease is not infrequent
Definition 4 of 221
Type 2 error when there really is a difference but on the basis of our samples, we decide that
there is no difference (designated by beta -β)
statistical power (1-β)
type 2 error (-B)
confidence interval (ci)
type 1 error (α)
,Term 5 of 221
what would you expect to result from a dichotomous example?
This factor alone can produce the disease, but so can other factors that are acting alone
factor a or b or c = disease
Radiation/benzene exposure alone-> leukemia
Radiation/benzene exposure ≠> leukemia
Other cofactors (immune status, translocations)-> leukemia
for evaluating new drugs and other treatments of
disease, including tests of new health and medical
care technology
to assess new programs for screening and early
detection
to assess new ways of organizing and delivering health
services.
Tells us how many times more likely it is that an event will occur in the treatment group
relative to the control group.
RR of 1-- no difference between the 2 groups thus, treatment
had no effect.
RR< 1 -- treatment decreases risk of outcome.
RR> 1 -- treatment increased risk of outcome
In a dichotomous (exposed or not exposed) example, if exposure is associated with
disease, we would expect the proportion of the cases who were exposed to be greater than
the proportion of the controls who were exposed
, Definition 6 of 221
in this design, the controls are a sample of individuals who are at risk for the disease at the time
each case of the disease develops
The cases & controls are, in effect, matched on calendar time & length of follow -up
Because a control is selected each time a case develops, a control who is selected early in the
study could later develop the disease & become a case in the same study.
Typical sequence of approach: study etiology
what are Nested case-control studies?
Cohort Studies Advantages
how do you interpreting the risk?
Term 7 of 221
recall bias
First, select cases with the disease & controls without the disease
Second, measure past exposure; interviews, review medical/employee records, or results of
chemical/ biologic assays of blood, urine, or tissues.
Is randomization ethical (withholding new treatment from the control group)?
Yes-only when we don't know whether drug A is better than drug B.
Is it ethical to withhold effective treatment (e.g., when using a
placebo)?
Trade -off between randomizing & not randomizing
.
arises in any situation in which individuals select themselves into a group (by purposeful
intent) causing a biased sample with abnormal or undesirable conditions in the group.
recall bias is a systematic error caused by differences in the accuracy or completeness of
the recollections retrieved ("recalled") by study participants regarding events or
experiences from the past
