Study Designs
The main goal of an observational study is to:
A. Identify sources of measurement error and bias.
B. Establish direct causality.
C. Describe associations between exposures and outcomes.
D. Maximize external validity.
Describe associations between exposures and outcomes? (C)
Which type of study design is most susceptible to bias?
A. Case control study
B. Randomized controlled trial
C. Case report
D. Cohort study
Case report? (C)
Controlling for important confounders by regression or stratification is a method often used in observational studies to reduce which type of bias?
A. Performance
B. Attrition
C. Reporting
D. Selection
Selection? (D)
Which measure is affected by the prevalence of the condition in the population being tested?
A. Sensitivity and specificity
B. Positive and negative predictive values
C. Positive and negative likelihood ratios
D. Receiver operating characteristic curves
Positive and negative predictive values? (B)
A test that best represents the true state of a condition is a(n):
A. Screening test
B. Reference test (also referred to as the gold standard)
C. Index test
D. Follow-up test
Reference test (also referred to as the gold standard)? (B)
Studies may be conducted to generate or test hypotheses. The best design for testing a hypothesis is a:
A. Case-control study
B. Cross-sectional study
C. Prospective cohort study
D. Randomized controlled trial
Randomized controlled trial? (D)
Studies without comparison groups are unable to account for other factors that may explain outcomes. Which term describes the phenomenon that participating in a research study may result in positive outcomes regardless of receiving treatment?
A. Hawthorne effect
B. Placebo effect
C. Regression to the mean
D. Natural history
Hawthorne effect? (A)
Which concept describes a limitation of retrospective studies that is of less concern in prospective studies?
A. Measurement bias
B. Recall bias
C. Confounder
D. Effect modifier
Recall bias? (B)
Which formula is incorrect?
A. Sensitivity = true positive / (true positive + false negative)
B. Specificity = true negative / (false positive + true negative)
C. Positive predictive value = true positive / (true positive + false positive)
D. Negative predictive value = (true negative + false negative) / true negative
Negative predictive value = (true negative + false negative) / true negative? (D)
Which of the following can reduce spectrum bias in a diagnostic accuracy study?
A. Including large numbers of diverse patients
B. Including patients with uncertain diagnoses
C. Including patients in whom we would consider using the test in clinical practice
D. All of the above
All of the above? (D)
What type of study can quickly and efficiently estimate the level of immunity to measles in a population offered measles vaccination?
A. Case-control study of measles infection
B. Cross-sectional survey of vaccination status
C. Randomized trial of measles vaccination
D. Retrospective cohort study of measles vaccination
Cross-sectional survey of vaccination status? (B)
Randomization is a strategy to reduce which type of bias?
A. Performance
B. Attrition
C. Reporting
D. Selection
Selection (D)?
How does a hazard ratio (HR) differ from a relative risk or odds ratio?
A. HR is a ratio of outcomes in the treatment and control groups
B. HR is a cumulative risk at a defined endpoint
C. HR is a relative measure of effect
D. HR cannot be estimated from a 2 x 2 table
HR cannot be estimated from a 2 x 2 table? (D)
Which statement about the receiver operating characteristic curve (ROC) is incorrect?
A. It is a plot of specificity (y-axis) against 1 minus sensitivity (x-axis) for different cutoff thresholds
B. An area under the ROC of 1.0 is a perfectly accurate test
C. An area under the ROC of 0.5 is a test that does not differentiate between the existence or absence of the condition
D. ROC is used for nonbinary outcomes
It is a plot of specificity (y-axis) against 1 minus sensitivity (x-axis) for different cutoff thresholds ? (A)
Which statement about studies of clinical prediction tools is incorrect?
A. An impact analysis provides evidence of reproducible accuracy
B. Applications may be narrow or broad
C. Studies quantify the contributions of multiple patient factors in predicting a clinical outcome
D. Validation studies are required before applying clinically
An impact analysis provides evidence of reproducible accuracy? (A)
A case-control study may have an advantage over a cohort study when the disease in question is:
A. Rare
B. Fatal
C. Infectious
D. Indolent
Rare? (A)
Performance bias can be reduced by:
A. Including all prespecified outcomes in the analysis.
B. Intention-to-treat analysis.
C. Blinding participants and investigators.
D. Use of prespecified enrollment criteria.
Blinding participants and investigators? (C)
Which statement about Kaplan-Meier curves is incorrect?
A. It is a graphical representation of the probability or likelihood of an outcome or event occurring during the study period
B. Shows time-to-event analysis or survival analysis
C. The occurrence of an outcome is always represented by a downward step in the curve
D. Can compare KP curves for different treatments or exposures
The occurrence of an outcome is always represented by a downward step in the curve ? (C)
In which situation is diagnostic testing most important?
A. When the pretest probability is very low
B. When the pretest probability is very high
C. To reduce risk of malpractice lawsuits
D. To provide prognostic information or change clinical management
Provide prognostic information or change clinical management? (D)
Which condition below helps rule out a disease:
A. High pretest probability
B. Small likelihood ratio test (-)
C. Large likelihood ratio test (+)
D. High specificity
Small likelihood ratio test (-) ? (B)
In a case-control study that is being planned to study possible causes of myocardial infarction (MI), patients with MI serve as the cases. Which of the following would be a good choice to serve as controls?
A. Patients whose cardiac risk factors are similar to those of the cases and who have never had an MI in the past.
B. Patients whose cardiac risk factors are similar to those of the cases and who have had an MI in the past.
C. Patients who have never had an MI in the past.
D. Patients whose cardiac risk factors are unknown and who have had an MI in the past.
Patients who have never had an MI in the past? (C)
Effects of attrition bias are greater for all the following situations except:
A. Groups are not randomized.
B. There is high loss to follow-up.
C. The differences between those lost to follow-up and those who complete the study are large.
D. There are large differences in attrition rates between comparison groups.
Groups are not randomized? (A)
Censored observations in a study include all information on a participant up to the point of follow-up under all but one of the following circumstances:
A. Participant died during the study
B. Participant was lost to follow-up
C. Participant was routinely late to follow-up appointments
D. Participant had a prespecified disqualifying event
Participant was routinely late to follow-up appointments? (C)
For diagnostic accuracy studies, which type of study design better reflects clinical practice and is less biased?
A. Cross-sectional study
B. Before-after study
C. Case-control study
D. Case-series study
Cross-sectional study? (A)
Which condition below helps rule in a disease:
A. Low pretest probability
B. Small likelihood ratio test (-)
C. High sensitivity
D. High specificity
High specificity? (D)