Biostatistics Practice Questions

Q: What is the correlation coefficient if regression coefficient of X on Y is 0.8 and Y on X is 0.9?

0.85. ***Correct: 0.85*** - The correlation coefficient (r) is the **geometric mean** of the two regression coefficients - Formula: r = √(b_xy × b_yx), where b_xy is the regression coefficient of X on Y and b_yx is the regression coefficient of Y on X - Calculation: r = √(0.8 × 0.9) = √0.72 ≈ **0.8485**, which rounds to **0.85** - Since both regression coefficients are positive, the correlation is positive *Incorrect: 0.95* - This would be obtained by taking the **arithmetic mean** [(0.8 + 0.9)/2 = 0.85... wait, that's not 0.95] - Actually, this value is too high and doesn't result from any standard calculation with these regression coefficients - The correct method requires the **geometric mean** (square root of the product), not any simple average *Incorrect: 0.81* - This appears to be the square of one regression coefficient (0.9² = 0.81) - However, the correlation coefficient requires the **square root of the product** of both coefficients, not squaring a single coefficient - This is a common error in calculation *Incorrect: 0.72* - This is the **product** of the two regression coefficients (0.8 × 0.9 = 0.72) - This is an intermediate step in the calculation, but not the final answer - The correlation coefficient requires taking the **square root** of this product: √0.72 ≈ 0.85

Q: Which method is most accurate for estimating the incidence of a disease?

Cohort study. ***Cohort study*** - A **cohort study** tracks a group of individuals over time to observe the development of new cases of a disease, allowing for direct calculation of **incidence rates**. - It starts with a healthy population and identifies who develops the disease, providing the most accurate measure of **risk** and incidence. *Case-control study* - **Case-control studies** are primarily used to investigate **risk factors** for a disease by comparing exposures between individuals with the disease (cases) and those without (controls). - They **cannot directly estimate incidence** because they are retrospective and select participants based on disease status. *Cross-sectional study* - A **cross-sectional study** assesses the prevalence of a disease and/or exposure at a single point in time. - It provides a snapshot of the population's health status but **cannot determine incidence** as it doesn't observe new cases developing over time. *Ecological study* - An **ecological study** examines disease rates and exposures across populations rather than individuals. - While useful for generating hypotheses, it is prone to the **ecological fallacy** and cannot determine individual-level incidence.

Q: In the context of diagnostic testing, how does using a series testing approach affect the specificity and sensitivity of the test?

Increases specificity and decreases sensitivity. ***Increases specificity and decreases sensitivity*** - A **series testing approach** means that for a diagnosis to be made, **all tests in the sequence must be positive**. This approach is designed to reduce false positives because a patient must pass multiple hurdles. - By requiring multiple positive results, the likelihood of a false positive for the overall diagnosis decreases, thereby **increasing the specificity** of the diagnostic process. Conversely, this stricter criterion means some true positives might be missed (if one test in the series is negative, even if the patient has the disease), leading to a **decrease in sensitivity**. *Increases sensitivity and decreases specificity* - This outcome is characteristic of a **parallel testing approach**, where a positive result on **any one of several tests** is sufficient for a diagnosis. - While parallel testing increases the chance of catching true positives (higher sensitivity), it also raises the risk of false positives (lower specificity) because fewer criteria need to be met. *Both increase* - It is generally **not possible** to increase both sensitivity and specificity simultaneously through a simple change in testing strategy without altering the intrinsic properties of the tests themselves. - There is typically an **inverse relationship** between sensitivity and specificity; improving one often comes at the expense of the other. *Both decrease* - A decrease in both sensitivity and specificity would indicate a **poorly designed or executed testing strategy**, or using tests that are individually unreliable. - This outcome would be undesirable as it would lead to both a high rate of missed diagnoses and a high rate of false positive diagnoses.

Q: A study is conducted to assess the relationship between smoking and lung cancer by following a group of smokers and non-smokers over time to calculate incidence rates. What type of study design is being described?

Cohort study. ***Cohort study*** - A **cohort study** follows a group of individuals over time based on their exposure status (smokers vs. non-smokers) to see who develops the outcome (lung cancer). - This design allows for the calculation of **incidence rates** and **relative risk**, providing strong evidence for causality and temporal relationships. - Cohort studies are considered the **gold standard** for observational studies as they establish that exposure precedes disease. *Case-control study* - A **case-control study** starts with individuals who already have the outcome (lung cancer cases) and compares their past exposure (smoking) to a control group without the outcome. - While efficient for rare diseases and diseases with long latency periods, the question specifically describes **following subjects over time**, which is characteristic of a cohort design, not case-control. *Cross-sectional study* - A **cross-sectional study** assesses exposure and outcome simultaneously at a single point in time, providing a snapshot of prevalence. - It cannot establish a **temporal relationship** between smoking and lung cancer, and does not involve following subjects over time. *Randomized controlled trial* - A **randomized controlled trial (RCT)** involves randomly assigning participants to an intervention or control group and is best for evaluating the effectiveness of treatments or preventive interventions. - It would be **unethical** to randomize individuals to smoke to assess lung cancer risk, making this design inappropriate for studying harmful exposures.

Q: Bias in clinical trials can be minimized by all of the following methods, EXCEPT:

Multivariate analysis. ***Multivariate analysis*** - **Multivariate analysis** is a statistical tool used to analyze data with multiple variables, but it cannot prevent bias during the design or execution phases of a clinical trial. - While it can help control for confounding variables during data analysis, it is applied *after* data collection and does not eliminate sources of bias related to participant selection, intervention assignment, or outcome assessment. *Matching* - **Matching** involves selecting control group participants who share similar characteristics (e.g., age, sex, comorbidities) with the intervention group, thereby reducing confounding by known variables. - This method helps to ensure that groups are comparable at baseline, minimizing **selection bias** and making observed differences more attributable to the intervention. *Blinding* - **Blinding** involves concealing the treatment assignment from participants, researchers, or both (single- or double-blinding) to prevent their expectations or preconceptions from influencing outcomes or assessments. - This method effectively minimizes various forms of **performance bias** (e.g., in participant behavior or co-interventions) and **detection bias** (e.g., in outcome assessment). *Randomization* - **Randomization** is the process of assigning participants to intervention or control groups purely by chance, which helps ensure that known and unknown confounding factors are evenly distributed between groups. - This method is crucial for minimizing **selection bias**, making the groups comparable and increasing the likelihood that any observed differences are due to the intervention rather than pre-existing differences.

Q: In an international health study, researchers want to assess the impact of socioeconomic status on health outcomes. Which statistical method is best suited for this analysis?

Multivariate regression. ***Multivariate regression*** - This method is ideal for examining the relationship between multiple independent variables (like **socioeconomic status** components such as income, education, occupation) and one or more dependent variables (various **health outcomes**). - It allows researchers to control for confounding factors and determine the independent contribution of each socioeconomic variable to health outcomes. *Chi-square test* - The Chi-square test is typically used to assess the association between **two categorical variables**, such as different socioeconomic categories and the presence or absence of a specific health condition. - It is not suitable for analyzing the impact of multiple continuous or ordinal socioeconomic factors on complex health outcomes. *ANOVA* - **ANOVA (Analysis of Variance)** is used to compare the means of a continuous dependent variable across **two or more groups** defined by a categorical independent variable. - While it could compare health outcomes between different socioeconomic *groups*, it is less efficient for modeling the continuous or complex relationships among multiple socioeconomic factors and health outcomes. *Survival analysis* - **Survival analysis** focuses on the time until an event occurs (e.g., time until disease onset, death, or recovery) and is often used in longitudinal studies. - While socioeconomic status can be a predictor in survival models, this method is not the primary choice for broadly assessing the overall impact of socioeconomic status on health outcomes in a cross-sectional or general population study.

Q: In a cohort study, which measure is most appropriate for comparing the risk of developing a disease between exposed and non-exposed groups?

Relative risk. ***Relative risk*** - The **relative risk (RR)** directly compares the **incidence of disease** in the exposed group to the incidence in the non-exposed group in a **cohort study**. - It quantifies how many times more likely (or less likely) the exposed group is to develop the disease compared to the unexposed group. *Odds ratio* - The **odds ratio (OR)** is primarily used in **case-control studies** to estimate the association between exposure and outcome. - While it can approximate the relative risk when the disease is rare, it is not the most direct measure of risk comparison in a cohort study. *Absolute risk* - **Absolute risk** refers to the **incidence of the disease** in a specific group and does not inherently involve a comparison between exposed and non-exposed groups. - It represents the probability of developing the disease over a specified period. *Incidence rate* - **Incidence rate** measures the **frequency of new cases** of a disease in a population over a given period. - While essential for calculating relative risk, incidence rate itself is a measure of occurrence, not a direct comparative measure between two groups.

Question 1

What is the correlation coefficient if regression coefficient of X on Y is 0.8 and Y on X is 0.9?

Accepted Answer

0.85

Answer

0.95

Answer

0.81

Answer

0.72

Question 2

Which method is most accurate for estimating the incidence of a disease?

Accepted Answer

Cohort study

Answer

Case-control study

Answer

Cross-sectional study

Answer

Ecological study

Question 3

In the context of diagnostic testing, how does using a series testing approach affect the specificity and sensitivity of the test?

Accepted Answer

Increases specificity and decreases sensitivity

Answer

Increases sensitivity and decreases specificity

Answer

Both increase

Answer

Both decrease

Question 4

A study is conducted to assess the relationship between smoking and lung cancer by following a group of smokers and non-smokers over time to calculate incidence rates. What type of study design is being described?

Accepted Answer

Cohort study

Answer

Case-control study

Answer

Cross-sectional study

Answer

Randomized controlled trial

Question 5

A researcher is conducting a study to compare two treatment modalities for hypertension. The study finds a p-value of 0.02 and a 95% confidence interval for the difference in means that does not include zero. How should these results be interpreted?

Accepted Answer

The results are statistically significant, indicating that the null hypothesis can be rejected.

Answer

There is no significant difference between the two treatments.

Answer

The p-value indicates that the results are not significant.

Answer

The confidence interval suggests that the study has low power.

Question 6

Bias in clinical trials can be minimized by all of the following methods, EXCEPT:

Accepted Answer

Multivariate analysis

Answer

Matching

Answer

Randomization

Answer

Blinding

Question 7

A study comparing two antihypertensive drugs reports a p-value of 0.03 for the primary outcome. What is the correct interpretation of this result?

Accepted Answer

There is a 3% chance that the observed difference occurred by chance if the null hypothesis is true.

Answer

The probability of making a Type I error is 3%.

Answer

The study has a 97% chance of detecting a true difference if it exists.

Answer

There is a 3% chance that the null hypothesis is false.

Question 8

In an international health study, researchers want to assess the impact of socioeconomic status on health outcomes. Which statistical method is best suited for this analysis?

Accepted Answer

Multivariate regression

Answer

Chi-square test

Answer

ANOVA

Answer

Survival analysis

Question 9

In a cohort study, which measure is most appropriate for comparing the risk of developing a disease between exposed and non-exposed groups?

Accepted Answer

Relative risk

Answer

Odds ratio

Answer

Absolute risk

Answer

Incidence rate

Question 10

A study reports a confidence interval (CI) for a relative risk (RR) ranging from 0.8 to 1.2. How should this result be interpreted in relation to the association between exposure and outcome?

Accepted Answer

No significant association exists between the exposure and outcome.

Answer

The exposure reduces the risk of the outcome.

Answer

The study may have low statistical power.

Answer

The exposure increases the risk of the outcome.

Biostatistics — MCQs

Biostatistics — MCQs

On this page

Practice by Chapter

Want unlimited practice?