Biostatistics Practice Questions

Q: What is the preferred measure of central tendency for ordinal data?

Median. ### Explanation In biostatistics, the choice of central tendency depends entirely on the **scale of measurement** of the data. **Why Median is the Correct Answer:** Ordinal data consists of categories that have a natural rank or order (e.g., stages of cancer, socio-economic status, or Likert scales), but the mathematical distance between these ranks is not uniform or quantifiable. The **Median** is the preferred measure because it identifies the middle value in a ranked distribution. It respects the relative positioning of the data points without requiring the numerical intervals necessary for calculating a mean. **Analysis of Incorrect Options:** * **A. Mean:** Requires **Interval or Ratio (Quantitative)** data. Since ordinal data lacks equal intervals (e.g., the "gap" between Stage I and II cancer isn't necessarily the same as between Stage III and IV), calculating an average is mathematically invalid. * **B. Mode:** While the mode can be used for ordinal data (representing the most frequent category), it is less descriptive than the median because it ignores the inherent ranking of the other data points. It is the preferred measure for **Nominal** data (e.g., blood groups). * **D. Range:** This is a measure of **dispersion (variability)**, not central tendency. It describes the spread between the maximum and minimum values. **High-Yield Clinical Pearls for NEET-PG:** * **Nominal Data:** Use **Mode** (e.g., Gender, Religion). * **Ordinal Data:** Use **Median** (e.g., Pain scales, APGAR score). * **Symmetrical/Normal Distribution:** Mean = Median = Mode. * **Skewed Distribution:** The **Median** is the best measure of central tendency because the Mean is sensitive to outliers (extreme values). * **Most common measure** used in medical statistics is the **Mean**, but the **most robust** is the **Median**.

Q: Which epidemiological study design is represented by the following illustration?

Case control study. ***Case control study*** - This study design works **retrospectively**, starting with individuals who already have the disease (cases) and those without (controls), then looking back to assess **past exposure**. - The illustration shows two groups being compared based on their **outcome status** first, then examining their **exposure history**. *Cohort study* - This design follows participants **prospectively** from exposure to outcome, moving forward in time. - Participants are selected based on their **exposure status** (exposed vs. unexposed), not their disease status. *Ecological study* - This design examines **populations or groups** rather than individuals, comparing disease rates across different communities or regions. - The unit of analysis is the **population level**, not individual cases and controls. *Cross-sectional study* - This design examines both **exposure and outcome simultaneously** at a single point in time. - It provides a **snapshot** of the population without following participants over time or looking backwards.

Q: What is a disadvantage of calculating the mean?

May be influenced by abnormal high or low values. **Explanation:** The **Arithmetic Mean** is the most commonly used measure of central tendency in biostatistics. It is calculated by summing all observations and dividing by the total number of items ($Mean = \Sigma x / n$). **Why Option C is Correct:** The primary disadvantage of the mean is its **sensitivity to extreme values (outliers)**. Because every single value in a dataset is used in the calculation, a single abnormally high or low value will "pull" the mean toward it, making it an unrepresentative measure of the "average." For example, in a study of five patients with recovery times of 3, 4, 5, 6, and 50 days, the mean is 13.6 days—a value that does not accurately reflect the typical recovery time of the group. **Why Incorrect Options are Wrong:** * **Option A & B:** The mean is actually the **easiest** measure of central tendency to calculate mathematically and the most **intuitive** to understand for clinicians and researchers. * **Option D:** This is incorrect as Option C is a well-documented statistical limitation. **High-Yield Clinical Pearls for NEET-PG:** * **Skewed Data:** In a skewed distribution (non-normal), the mean is the most affected measure. In **positively skewed** data, Mean > Median > Mode. In **negatively skewed** data, Mean < Median < Mode. * **Best Measure:** The **Median** is the preferred measure of central tendency for skewed data or data containing outliers (e.g., incubation periods, survival times). * **Normal Distribution:** In a perfectly symmetrical (Gaussian) distribution, the Mean, Median, and Mode are all equal.

Q: Randomization is useful to eliminate which of the following?

Confounding factors. **Explanation:** **Why Confounding Factors is Correct:** Randomization is the "heart" of a Randomized Controlled Trial (RCT). Its primary purpose is to ensure that both known and unknown **confounding factors** are distributed equally between the study and control groups. By allocating participants purely by chance, randomization eliminates **selection bias** and balances prognostic factors, ensuring that any observed difference in outcome is due to the intervention alone and not an extraneous variable. It is the only method that can control for **unknown confounders**. **Why Other Options are Incorrect:** * **A & C (Observer and Patient Bias):** These are types of information/measurement bias. Randomization does not prevent these; they are eliminated through **Blinding** (Masking). Blinding ensures that neither the patient nor the researcher knows who is receiving the treatment, thus preventing subjective prejudice. * **D (Sampling Bias):** This occurs when the study population is not representative of the target population. It is eliminated through **Random Sampling** (e.g., simple random sampling), not Randomization (Random Allocation). **High-Yield Clinical Pearls for NEET-PG:** * **Randomization** = Eliminates Confounding + Selection Bias. * **Blinding** = Eliminates Observer + Patient (Information) Bias. * **Matching** = A technique used in Case-Control studies to eliminate known confounders (but cannot control for unknown ones). * **The "Gold Standard"** for testing a new drug is the Double-Blind Randomized Controlled Trial. * **Confounding by Indication:** A specific type of bias where the factor that dictates the treatment choice is also related to the outcome.

Q: A study was conducted on a group of children to assess the seasonal variation of Sudden Infant Death Syndrome (SIDS) during the summer months (June-July) and a subsequent period (August-September) in a group with similar characteristics from the same area. Which statistical test is used to compare the data from these two groups?

Chi-Square test. ### Explanation **1. Why Chi-Square Test is Correct:** The study aims to compare the occurrence of Sudden Infant Death Syndrome (SIDS) between two distinct time periods (June-July vs. August-September). In biostatistics, SIDS is a **categorical (qualitative)** variable—an infant either experiences SIDS or does not. When comparing the frequencies or proportions of a categorical outcome between two independent groups, the **Chi-Square test** is the standard test of significance. It assesses whether the observed variation in SIDS cases across seasons is due to chance or a statistically significant association. **2. Why Other Options are Incorrect:** * **Paired T-test:** This is used for **quantitative (numerical)** data when comparing means of the same group before and after an intervention (e.g., blood pressure before and after a drug). It is not for categorical outcomes. * **Wilcoxon Rank-Sum/Signed-Rank Test:** These are **non-parametric** tests used for ordinal data or non-normally distributed quantitative data. They are not used for simple frequency comparisons of nominal data like SIDS. * **ANOVA (Analysis of Variance):** This is used to compare the **means of three or more** independent groups for quantitative data. It is not applicable to categorical data or a comparison of only two groups. **3. Clinical Pearls & High-Yield Facts:** * **Rule of Thumb:** If the data is in **proportions, percentages, or 2x2 tables**, think **Chi-Square**. If the data is **means/averages**, think **T-test**. * **Fisher’s Exact Test:** Use this instead of Chi-Square if the sample size is very small (any cell value in the 2x2 table is <5). * **SIDS Risk Factors:** High-yield associations include prone sleeping position (Back to Sleep campaign), maternal smoking, and overheating. Peak incidence typically occurs between 2–4 months of age.

Q: Which of the following is NOT evaluated by a cohort study?

Odds ratio. **Explanation** In epidemiology, the choice of study design dictates which measures of association can be calculated. A **Cohort Study** is a longitudinal, prospective study that starts with a group of exposed and non-exposed individuals and follows them over time to observe the development of an outcome. **Why Odds Ratio is the Correct Answer:** The **Odds Ratio (OR)** is the primary measure of association used in **Case-Control studies**. It estimates the odds of exposure among cases compared to controls. While OR can be calculated in a cohort study, it is not the standard or intended evaluation metric because cohort studies allow for the direct calculation of risk. **Analysis of Incorrect Options:** * **A. Incidence:** Cohort studies are the only observational design that can directly calculate incidence (the number of new cases in a population at risk) because they follow subjects from a healthy state to the onset of disease. * **B. Attributable Risk (AR):** AR measures the impact of an exposure (Incidence in exposed – Incidence in non-exposed). Since cohort studies provide incidence, AR can be easily derived. * **C. Relative Risk (RR):** Also known as Risk Ratio, this is the hallmark measure of a cohort study. It compares the incidence of disease in the exposed group to the incidence in the non-exposed group. **High-Yield Clinical Pearls for NEET-PG:** * **Cohort Study:** Proceeds from **Cause to Effect**. Best for rare exposures. * **Case-Control Study:** Proceeds from **Effect to Cause**. Best for rare diseases. * **Incidence** can *only* be calculated in Cohort studies (and Randomized Controlled Trials). * If a disease is rare, the Odds Ratio (from a case-control study) numerically approximates the Relative Risk.

Q: For the diagnosis of Deep Vein Thrombosis, two tests are done together: Impedance Plethysmography and leg scanning after injecting 125I fibrinogen. What is the primary purpose of combining these diagnostic procedures?

Increasing the negative predictive value. ### Explanation In clinical practice, diagnostic tests can be performed in two ways: **Parallel Testing** (tests done together) or **Serial Testing** (tests done one after another). **1. Why the correct answer is right (Option B):** The scenario describes **Parallel Testing**. When two tests are performed simultaneously, a patient is considered "positive" if *either* test is positive and "negative" only if *both* tests are negative. * **Mechanism:** This approach maximizes **Sensitivity** because it is less likely to miss a case of the disease. * **Impact on Predictive Value:** As sensitivity increases, the **Negative Predictive Value (NPV)** also increases. A negative result in parallel testing provides high confidence that the patient truly does not have Deep Vein Thrombosis (DVT), effectively "ruling out" the disease. **2. Why the incorrect options are wrong:** * **Option A & D:** Increasing the **Positive Predictive Value (PPV)** and **Specificity** is the goal of **Serial Testing** (e.g., an ELISA followed by a Western Blot for HIV). Serial testing "rules in" a disease by ensuring that only those who pass multiple diagnostic hurdles are labeled positive. * **Option C:** **Pretest odds** (or pretest probability) are determined by the prevalence of the disease in the population and the clinical presentation of the patient *before* any tests are conducted. Performing diagnostic tests does not change the pretest odds. **3. High-Yield Clinical Pearls for NEET-PG:** * **Parallel Testing:** ↑ Sensitivity, ↑ NPV, ↓ Specificity. (Used for **Screening** or emergency "rule-out"). * **Serial Testing:** ↑ Specificity, ↑ PPV, ↓ Sensitivity. (Used for **Confirmation** of a diagnosis). * **Net Sensitivity in Parallel:** $1 - [(1 - \text{Sens}_1) \times (1 - \text{Sens}_2)]$. * **Net Specificity in Parallel:** $\text{Spec}_1 \times \text{Spec}_2$.

Q: A screening test was carried out among 120 people. 20 had the disease, and 40 showed a positive test result, out of which 15 had the disease. What is the specificity of the test?

75%. ### Explanation To solve this problem, we must first organize the data into a standard **2x2 Contingency Table**. | | Disease Present (D+) | Disease Absent (D-) | Total | | :--- | :---: | :---: | :---: | | **Test Positive (T+)** | 15 (True Positive) | 25 (False Positive) | 40 | | **Test Negative (T-)** | 5 (False Negative) | 75 (True Negative) | 80 | | **Total** | **20** | **100** | **120** | **Step-by-Step Calculation:** 1. **Total Population:** 120. 2. **Disease Present (D+):** 20. Therefore, **Disease Absent (D-):** $120 - 20 = 100$. 3. **Test Positive (T+):** 40. Out of these, 15 have the disease (True Positives). 4. **False Positives (FP):** $40 - 15 = 25$. 5. **True Negatives (TN):** Total Disease Absent (100) minus False Positives (25) = **75**. **Specificity Formula:** $$\text{Specificity} = \frac{\text{True Negatives (TN)}}{\text{Total Disease Absent (D-)}} \times 100$$ $$\text{Specificity} = \frac{75}{100} \times 100 = \mathbf{75\%}$$ --- #### Analysis of Incorrect Options: * **A (50%):** This is an incorrect calculation, likely confusing the ratio of test positives to the total diseased. * **B (65%):** No direct statistical correlation to the data provided. * **D (25%):** This represents the **False Positive Rate** ($25/100$). Remember: $\text{Specificity} = 1 - \text{False Positive Rate}$. --- #### NEET-PG Clinical Pearls: * **Specificity (SIn):** Highly specific tests, when positive, help **Rule In** the disease (SpPIn). It measures the ability of a test to identify true health. * **Sensitivity (SnOut):** Highly sensitive tests, when negative, help **Rule Out** the disease (SnNOut). * **Screening vs. Diagnosis:** Screening tests require high sensitivity (to catch all cases), while confirmatory/diagnostic tests require high specificity (to avoid false labeling). * **Prevalence:** Note that Sensitivity and Specificity are **independent** of disease prevalence, whereas Predictive Values (PPV/NPV) are prevalence-dependent.

Question 1

What is the preferred measure of central tendency for ordinal data?

Accepted Answer

Median

Answer

Mean

Answer

Mode

Answer

Range

Question 2

Which epidemiological study design is represented by the following illustration?

Accepted Answer

Case control study

Answer

Cohort study

Answer

Ecological study

Answer

Cross-sectional study

Question 3

What is a disadvantage of calculating the mean?

Accepted Answer

May be influenced by abnormal high or low values

Answer

Difficult to calculate

Answer

Difficult to understand

Answer

None of the above

Question 4

Randomization is useful to eliminate which of the following?

Accepted Answer

Confounding factors

Answer

Observer bias

Answer

Patient bias

Answer

Sampling bias

Question 5

A study was conducted on a group of children to assess the seasonal variation of Sudden Infant Death Syndrome (SIDS) during the summer months (June-July) and a subsequent period (August-September) in a group with similar characteristics from the same area. Which statistical test is used to compare the data from these two groups?

Accepted Answer

Chi-Square test

Answer

Paired T-test

Answer

Wilcoxon Rank test

Answer

ANOVA

Question 6

If a Hemoccult test is negative for screening of colonic cancer, no further test is done. If the Hemoccult test is positive, the individual will have a second stool sample tested with Hemoccult II test. If this second sample is positive for blood, the individual will be referred for more extensive evaluation. What is the effect of this screening method on net sensitivity and net specificity?

Accepted Answer

Net sensitivity is decreased and net specificity is increased.

Answer

Net sensitivity and net specificity are both increased.

Answer

Net sensitivity is increased and net specificity is decreased.

Answer

Net sensitivity remains the same and net specificity is increased.

Question 7

Which of the following is NOT evaluated by a cohort study?

Accepted Answer

Odds ratio

Answer

Incidence

Answer

Attributable risk

Answer

Relative risk

Question 8

A cessation experiment is a type of:

Accepted Answer

Randomized controlled trial (RCT)

Answer

Cohort study

Answer

Case control study

Answer

Cross-sectional study

Question 9

For the diagnosis of Deep Vein Thrombosis, two tests are done together: Impedance Plethysmography and leg scanning after injecting 125I fibrinogen. What is the primary purpose of combining these diagnostic procedures?

Accepted Answer

Increasing the negative predictive value

Answer

Increasing the positive predictive value

Answer

Increasing the pretest odds

Answer

Increasing the specificity

Question 10

A screening test was carried out among 120 people. 20 had the disease, and 40 showed a positive test result, out of which 15 had the disease. What is the specificity of the test?

Accepted Answer

75%

Answer

50%

Answer

65%

Answer

25%

Biostatistics — MCQs

Biostatistics — MCQs

On this page

Practice by Chapter

Want unlimited practice?