Biostatistics Practice Questions

Q: What is the most appropriate statistical test to test the statistical significance of the change in blood cholesterol levels after a month's treatment with atorvastatin?

Paired t-test. ***Paired t-test*** * A **paired t-test** is appropriate when comparing two means from the **same group of subjects** measured at two different time points (before and after treatment). * In this scenario, a single group's blood cholesterol levels are measured *before* and *after* atorvastatin treatment, making the observations dependent. *Unpaired or independent t-test* * An **unpaired t-test** is used to compare the means of two *independent* groups. * It would be used, for instance, if cholesterol levels were being compared between a group receiving atorvastatin and a separate control group. *Analysis of variance* * **Analysis of variance (ANOVA)** is used to compare **three or more means**. * It would be appropriate if there were multiple treatment groups or multiple time points for comparison beyond just two. *Chi-square test* * The **Chi-square test** is used to examine the association between **categorical variables**. * It would not be suitable here, as blood cholesterol level is a continuous numerical variable, not a categorical one.

Q: An investigator wants to know the similarity of the mean peak flow of expiratory rates among non-smokers, light smokers, moderate smokers, and heavy smokers. Which statistical test of significance is appropriate?

One way ANOVA. ***One way ANOVA*** - This test is appropriate for comparing the means of **three or more independent groups** (non-smokers, light, moderate, heavy smokers) on a **single quantitative dependent variable** (peak flow of expiratory rates). - It determines if there's a statistically significant difference between the means of these groups, indicating at least one group mean is different from the others. *Two way ANOVA* - This test is used when there are **two independent categorical variables** (factors) influencing a single continuous dependent variable. - In this scenario, there is only one independent categorical variable (smoking status) with multiple levels. *Student-t test* - The Student-t test is used to compare the means of **only two groups**. - Since this question involves comparing the means of four groups of smokers, a t-test would not be appropriate. *Chi square test* - The Chi-square test is used for analyzing the association between **two categorical variables**. - Here, one variable (peak flow) is continuous, making the Chi-square test unsuitable.

Q: What is the 95% confidence interval for the intraocular pressure (IOP) in the 400 people, given a mean of 25 mm Hg and a standard deviation of 10 mm Hg?

24-26. ***24-26*** - This is the correct 95% confidence interval calculated using the formula: **mean ± (Z-score × standard error of the mean)**. - For a 95% confidence interval, the **Z-score is 1.96**. - The **standard error of the mean (SEM)** = standard deviation / √(sample size) = 10 / √400 = 10 / 20 = **0.5**. - Therefore: 25 ± (1.96 × 0.5) = 25 ± 0.98 = **24.02 to 25.98**, which rounds to **24-26**. *22-28* - This interval is too wide for a 95% confidence interval with the given parameters. - An interval of ±3 would correspond to a Z-score of 3/0.5 = 6, which is far beyond the **1.96 required for 95% confidence**. - This would represent a much higher confidence level (>99.9%). *23-27* - This interval is slightly too wide, implying a larger margin of error than calculated. - A range of ±2 would require a Z-score of 2/0.5 = 4 times the SEM, which **overestimates the 95% confidence interval**. - This would correspond to approximately 99.99% confidence. *21-29* - This interval is significantly too wide for a 95% confidence interval. - An interval of ±4 would require a Z-score of 4/0.5 = 8 times the SEM, which would correspond to an **extremely high confidence level** (virtually 100%). - This dramatically exceeds what is needed for 95% confidence.

Q: For testing the statistical significance of the difference in heights among different groups of school children, which statistical test would be most appropriate?

ANOVA. ***ANOVA (Analysis of Variance)*** - **ANOVA** is used to compare the means of **three or more independent groups** simultaneously. In this scenario, you are comparing heights across "different groups" of school children, implying more than two groups. - It tests whether there are any significant differences between the means of these groups, using the **F-statistic**. *Student's t test* - The **Student's t-test** is designed to compare the means of **only two groups**. It would be inappropriate for comparing more than two groups. - Applying multiple t-tests for several groups would increase the risk of **Type I error** (false positive). *chi-square test* - The **chi-square test** is used for analyzing **categorical data** (frequencies or proportions), not for comparing means of continuous data like height. - It determines if there is a significant association between two categorical variables. *Paired 't' test* - A **paired t-test** is used when comparing the means of two related groups or when measurements are taken from the **same subjects at two different times** (e.g., before and after an intervention). - This scenario involves independent groups of children, not paired or repeated measures.

Q: In which study design is carry-over effect a PRIMARY methodological concern that requires washout periods between treatment phases?

Cross-over study. ***Correct Option: Cross-over study*** - In a **cross-over study**, each participant receives multiple treatments sequentially, meaning the effects of a previous treatment could carry over to the next treatment phase. - A **washout period** is essential in this design to allow the effects of the prior treatment to dissipate, preventing them from influencing the results of the subsequent treatment. - This is the PRIMARY design where carry-over effects are an inherent methodological concern. *Incorrect Option: Case-control study* - This design compares subjects with a condition (cases) to subjects without the condition (controls) to identify past exposures or risk factors; it does not involve sequential treatments that would lead to a carry-over effect. - The primary concern in case-control studies is **recall bias**, as participants must remember past exposures, not carry-over effects. *Incorrect Option: Concurrent parallel design* - Participants are randomly assigned to one of several treatment groups and receive *only one* treatment throughout the study, eliminating the possibility of a treatment from one phase affecting another. - While it avoids carry-over effects, it often requires a larger sample size compared to cross-over studies to achieve similar statistical power. *Incorrect Option: Cohort study* - This design tracks a defined group of individuals (cohort) over time to observe the incidence of disease and identify risk factors; it does not involve the administration of sequential treatments. - Key concerns in cohort studies include **loss to follow-up** and the potential for a long study duration, rather than carry-over effects.

Q: What is the 95% confidence interval in a study with an estimated prevalence of 10% and a sample size of 100, expressed as a percentage range?

4% to 16%. ***4% to 16%*** - To calculate the 95% **confidence interval** for a **proportion**, we use the formula: p ± 1.96 * sqrt((p * (1-p)) / n). - Given a prevalence (**p**) of 0.10 and a **sample size** (**n**) of 100, the standard error is sqrt((0.10 * 0.90) / 100) = sqrt(0.0009) = 0.03. - The 95% confidence interval is 0.10 ± (1.96 * 0.03), which is 0.10 ± 0.0588. This translates to a range of 0.0412 to 0.1588, or approximately **4% to 16%**. *Inadequate information to calculate 95% CI* - The necessary information, including **prevalence** (10%) and **sample size** (100), is provided in the question. - With these two **parameters**, the 95% confidence interval can be calculated using standard statistical formulas. *6% to 16%* - This range is too narrow and suggests a smaller **standard error** or a different **confidence level**. - The correct calculation based on the provided **prevalence** and **sample size** yields a wider interval. *5% to 15%* - This range, while plausible, is slightly narrower than the **calculated interval**. - The use of the standard formula for a **proportion** with the given values results in a lower bound closer to 4% and an upper bound closer to 16%.

Q: For calculation of sample size for a prevalence study, all of the following are necessary except:

Power of the study. ***Power of the study*** - The **power of a study** is primarily relevant when calculating sample sizes for **hypothesis testing** (e.g., comparing two groups) to detect a statistically significant difference if one exists. - In a prevalence study, the goal is to estimate a proportion or prevalence with a certain level of precision, rather than to test a hypothesis. *Prevalence of disease in population* - An **estimated prevalence** is crucial for sample size calculation in prevalence studies, as it directly influences the variability of the proportion being estimated. - A higher or lower estimated prevalence affects the required sample size to achieve a desired level of precision. *Significance level* - The **significance level (alpha)** defines the probability of rejecting the null hypothesis when it is true (Type I error). - While essential for hypothesis testing, it is still used in prevalence studies to define the **confidence level** for the estimated prevalence (e.g., 95% confidence interval corresponds to an alpha of 0.05). *Desired precision* - **Desired precision**, often expressed as the **margin of error**, is a fundamental component of sample size calculation for prevalence studies. - It specifies how close the sample estimate should be to the true population prevalence.

Q: Which among the following is the major practical problem in a cohort study?

Differential loss of follow up. ***Differential loss of follow up*** - **Differential loss to follow-up** occurs when participants lost to follow-up differ systematically concerning exposure and outcome, potentially introducing **selection bias**. - This is a significant practical problem as it can distort the observed association between exposure and outcome, leading to biased results. *Long duration of study* - While **cohort studies** can indeed be **longitudinal** and require a long duration, this is more of an inherent characteristic and resource challenge rather than a "problem" that significantly compromises the validity of the study design itself. - The long duration primarily affects costs and feasibility but doesn't inherently invalidate the findings as much as differential loss to follow-up. *Can be used only for rare conditions* - This statement is incorrect; **cohort studies** are actually **inefficient for rare diseases** because a very large sample size would be needed to observe enough cases of the outcome. - **Case-control studies** are generally preferred for investigating **rare conditions** due to their retrospective outcome-to-exposure design. *No significant problems with cohort studies.* - This statement is incorrect; **cohort studies**, like all observational study designs, have inherent **methodological challenges** and potential sources of bias. - Problems include the **cost** and **time commitment**, **loss to follow-up**, and the potential for **confounding**, all of which require careful consideration in study design and analysis.

Question 1

What is the most appropriate statistical test to test the statistical significance of the change in blood cholesterol levels after a month's treatment with atorvastatin?

Accepted Answer

Paired t-test

Answer

Unpaired or independent t-test

Answer

Analysis of variance

Answer

Chi-square test

Question 2

An investigator wants to know the similarity of the mean peak flow of expiratory rates among non-smokers, light smokers, moderate smokers, and heavy smokers. Which statistical test of significance is appropriate?

Accepted Answer

One way ANOVA

Answer

Two way ANOVA

Answer

Student-t test

Answer

Chi square test

Question 3

What is the 95% confidence interval for the intraocular pressure (IOP) in the 400 people, given a mean of 25 mm Hg and a standard deviation of 10 mm Hg?

Accepted Answer

24-26

Answer

22-28

Answer

23-27

Answer

21-29

Question 4

For testing the statistical significance of the difference in heights among different groups of school children, which statistical test would be most appropriate?

Accepted Answer

ANOVA

Answer

Student's t test

Answer

chi-square test

Answer

Paired 't' test

Question 5

In which study design is carry-over effect a PRIMARY methodological concern that requires washout periods between treatment phases?

Accepted Answer

Cross-over study

Answer

Concurrent parallel design

Answer

Case-control study

Answer

Cohort study

Question 6

This study found a correlation coefficient of +0.7 between self-reported work satisfaction and life expectancy in a random sample of 5,000 corporate workers, with a p-value of 0.01. This means that:

Accepted Answer

Strong statistically significant (+) association between work satisfaction and life expectancy.

Answer

Correlation does not imply that 70% of people who enjoy work shall live longer.

Answer

Correlation coefficient of +0.7 indicates a moderate positive relationship, not a percentage.

Answer

Work satisfaction is moderately associated with life expectancy.

Question 7

What is the 95% confidence interval in a study with an estimated prevalence of 10% and a sample size of 100, expressed as a percentage range?

Accepted Answer

4% to 16%

Answer

Inadequate information to calculate 95% CI

Answer

6% to 16%

Answer

5% to 15%

Question 8

For calculation of sample size for a prevalence study, all of the following are necessary except:

Accepted Answer

Power of the study

Answer

Significance level

Answer

Desired precision

Answer

Prevalence of disease in population

Question 9

Which of the following best describes a cohort study?

Accepted Answer

A study that observes a group of individuals over time to assess the impact of a risk factor.

Answer

A cross-sectional study that collects data at a single point in time.

Answer

A case-control study that compares individuals with a condition to those without.

Answer

A study that randomly assigns participants to intervention and control groups.

Question 10

Which among the following is the major practical problem in a cohort study?

Accepted Answer

Differential loss of follow up

Answer

Long duration of study

Answer

Can be used only for rare conditions

Answer

No significant problems with cohort studies.

Biostatistics — MCQs

Biostatistics — MCQs

On this page

Practice by Chapter

Want unlimited practice?