Stats Flashcards
In an analysis to compare the time to event outcomes for two different groups, where the survival probability lines of two groups cross, what assumption is violated if a Cox regression was used to analyze these data?
A) Proportional hazards
B) Data are normally distributed
C) Noninformative censoring
D) Survival times are independent between subjects
A) Proportional hazards
Key assumption of Cox regression analysis is that the hazard ratio is constant over time. If survival curves cross, this indicates that the ratio of the hazards between the two groups varies over time.
In clinical trials, what randomization procedure ensures that treatment groups are balanced among prognostic factors?
A) Stratified randomization
B) Simple randomization
C) Block randomization
D) Unequal randomization
A) Stratified randomization
Where by treatments are randomly assigned within prognostic factors, e.g., biological sex.
Which research study design is used to determine the effectiveness of an intervention?
A) Phase I clinical trial
B) Cohort study
C) Phase III clinical trial
D) Case control study
C) Phase III clinical trial
Randomized to control or intervention group, aimed at determining the effectiveness of the intervention.
Which type of trial compare the benefits of a new intervention to control data from an earlier study
A) Concurrent control trial
B) Historical control trial
C) Randomized control trial
D) Phase I clinical trial
B) Historical control trial
New intervention is compared to previously collected data from observation or clinical trial research studies.
A study used a paired design in which samples from each study participant were tested for the presence of a disease using the two different diagnostic tests. What is the appropriate statistical test to compare the sensitivity between the two tests?
A) McNemar test
B) F-test
C) t-test
D) Wilcoxon rank-sum test
A) McNemar test
Two diagnostic tests were performed on the same person = paired nominal outcomes are clustered. The analysis needs to adjust correlation of the observation within clusters. The McNemar test is the only test listed that accounts for clustering for nominal outcomes
What is the mode of this number set? 3, 5, 5, 4, 8
A) 3
B) 4
C) 5
D) 8
C) 5
Mode = value that occurs most frequently
What is the mean of this number set? 3, 5, 5, 4, 8
A) 3
B) 4
C) 5
D) 8
C) 5
Average = mean, sum of all values divided by number of values
What is the median of this number set? 3, 5, 5, 4, 8
A) 3
B) 4
C) 5
D) 8
C) 5
Median is the 50th percentile of a set of values
What is a phase I trial used to determine?
A) Sample size of a phase III trial
B) Efficacy of an experimental drug
C) The maximum tolerated dose
D) The duration of a phase III trial
C) The maximum tolerated dose
Particularly for oncology phase I trials
When using historical controls in a Phase III trial, what is the top priority to maintain trial integrity?
A) Reduce the sample size of the phase III trial
B) Reduce the duration of the phase III trial
C) Remove the placebo arm
D) Control the type I error
D) Control the type I error
Historical data may behave differently with the placebo data from a phase III trial; thus any possibility of inflated type I error (rejecting the null hypothesis when it is true) may be controversial and jeopardize the trial outcome.
What are the criteria for an ideal screening test that balances sensitivity and specificity?
A) Low sensitivity and low specificity
B) High sensitivity and low specificity
C) Low sensitivity and high specificity
D) High sensitivity and high specificity
D) High sensitivity and high specificity
Screening refers to application of a test to people who have no symptoms with disease likely or unlikely as the outcome. The sensitivity is the percentage of subjects with disease correctly identified and specificity is the percentage of subjects without disease correctly identified.
What is the advantage of a retrospective study?
A) Controls are representative of the general population
B) They determine both causation and association
C) The are less expensive than prospective studies
D) Abstractors are not blinded
C) They are less expensive than prospective studies
What are the FDA’s commonly recommend endpoints for approval for new cancer drugs?
A) Time to progression and PFS
B) DFS and event free survival
C) Time to treatment failure
D) A & B
D) A and B
Which of the following is TRUE about a Chi Square test for statistical significance?
A) Indicates the strength of an association between two variables
B) Indicates the direction of an association between two variables
C) Provides absolutes, conclusive proof of a relationship
D) Needs to consider degrees of freedom.
D) Needs to consider degrees of freedom
Chi Square tests require that one put the obtained values into context. The degrees of freedom refer to the size of contingency table upon which the Chi Square value has been calculated.
Which of the following is true about clinical study designs?
A) Phase I studies aim to identify MTD in animal studies
B) Phase I trial usually include more people (more than a few dozen)
C) Case series include a group of case reports of patients with a similar exposure, treatment, or disease
D) Cohort studies can be prospective only
C) Case series include a group of case reports of patients with a similar exposure, treatment or disease.
