Biostatistics Practice Questions

Q: Mammography has 90% sensitivity and 98% specificity for breast carcinoma. What is the probability that a woman with breast carcinoma remains undiagnosed for 2 consecutive years?

1%. ### Explanation **1. Understanding the Correct Answer (1%)** To solve this, we must first identify the **False Negative Rate (FNR)**. * **Sensitivity** is the ability of a test to correctly identify those with the disease. Here, it is 90% (0.9). * The **False Negative Rate** is the probability that a person *with* the disease tests negative. It is calculated as: $1 - \text{Sensitivity}$. * $1 - 0.90 = 0.10$ (or 10%). The question asks for the probability of remaining undiagnosed for **two consecutive years**. Since each year's screening is an independent event, we use the **Multiplication Rule of Probability**: * Year 1 False Negative (0.10) × Year 2 False Negative (0.10) = **0.01 or 1%**. **2. Why Other Options are Incorrect** * **10% (Implicitly considered):** This represents the probability of a false negative in a single year. It fails to account for the second consecutive screening. * **2%:** This is a common distractor often confused with the complement of specificity (False Positive Rate). Specificity (98%) is irrelevant here because the question specifies the woman *has* breast carcinoma. * **0.1%:** This would be the result if the sensitivity were 99% ($0.01 \times 0.01$). **3. Clinical Pearls & High-Yield Facts for NEET-PG** * **Sensitivity (True Positive Rate):** Best for **screening** tests (SNOUT: Sensitivity rules OUT). * **Specificity (True Negative Rate):** Best for **confirmatory** tests (SPIN: Specificity rules IN). * **False Negative Rate (Type II Error/β):** The probability of missing a diagnosis. * **False Positive Rate (Type I Error/α):** Calculated as $1 - \text{Specificity}$. * **Sequential Testing:** When two tests are used in series (like yearly mammograms), the overall sensitivity decreases, but the overall specificity increases. However, the probability of a "miss" (False Negative) across multiple tests is the product of individual false negative rates.

Q: Twenty-five children are chosen randomly from a school playground. Their heights are measured in 2 consecutive months. What statistical test could be used to test the hypothesis that their heights are not different in consecutive months?

Paired t-test. ### Explanation **Why Paired t-test is the Correct Answer:** The core concept here is the **comparison of means in a "before-and-after" scenario** involving the same group of individuals. In this question, the same 25 children are measured twice (Month 1 and Month 2). Since the two sets of data are dependent (linked to the same person), we use the **Paired t-test**. This test evaluates whether the mean difference between these two related observations is significantly different from zero. **Analysis of Incorrect Options:** * **A. Unpaired t-test (Independent t-test):** This is used to compare the means of two **independent** groups (e.g., comparing heights of 25 children from School A with 25 different children from School B). * **C. Z-test:** While also used to compare means, a Z-test requires a **large sample size (n > 30)** and a known population variance. Here, the sample size is small (n = 25). * **D. Regression:** This is used to determine the **strength and direction of a relationship** between a dependent and independent variable (e.g., how much height increases for every year of age), rather than testing the difference between two sets of measurements. **High-Yield Clinical Pearls for NEET-PG:** * **Paired data** occurs in: Before-after studies, cross-over trials, and case-control studies with 1:1 matching. * **Parametric vs. Non-parametric:** If the data were not normally distributed, the non-parametric alternative to the Paired t-test would be the **Wilcoxon Signed-Rank Test**. * **Sample Size Rule:** Use **t-test** for n 30. * **ANOVA:** Use this when comparing means of **more than two** independent groups.

Q: For a group of 250 subjects, what value represents the 40th percentile?

100. **Explanation:** **1. Why the Correct Answer is Right:** In biostatistics, a **percentile** is a measure used to indicate the value below which a given percentage of observations in a group of observations falls. To calculate the position of a specific percentile in a dataset, the formula used is: * **Rank (n) = (P / 100) × N** * *P* = Percentile (40) * *N* = Total number of subjects (250) Applying the formula: $(40 / 100) \times 250 = 0.4 \times 250 = \mathbf{100}$. Therefore, the 100th value in the ordered dataset represents the 40th percentile. **2. Why Incorrect Options are Wrong:** * **Option A (7):** This is a random low number and does not correspond to any standard statistical calculation for this dataset. * **Option B (40):** This is a common distractor where students confuse the *percentile rank* (40th) with the *absolute value* (100). 40 is the percentage, not the count of subjects. * **Option D (140):** This value is mathematically incorrect. It might be reached if a student incorrectly adds the percentile to the total or miscalculates the percentage. **3. NEET-PG High-Yield Clinical Pearls:** * **Median:** The 50th percentile is always the Median. In this dataset, the median would be the 125th value. * **Quartiles:** The 25th percentile is the 1st Quartile (Q1), and the 75th percentile is the 3rd Quartile (Q3). * **Interquartile Range (IQR):** Calculated as Q3 – Q1; it contains the middle 50% of the data and is used to describe skewed distributions. * **Growth Charts:** In clinical practice, percentiles are most commonly used in pediatric growth charts (e.g., a child at the 95th percentile for weight is heavier than 95% of children their age).

Q: A study surveyed 3500 patients with thyroid cancer regarding their past exposure to radiation. The study design most appropriately illustrates which of the following?

Case series report. **Explanation:** The correct answer is **Case series report (Option A)**. In this scenario, the researchers are looking at a group of 3500 patients who already have the disease (thyroid cancer) and are documenting their history of exposure. A **Case Series** is a descriptive study design that describes the characteristics of a group of individuals with a common disease or exposure. It does not utilize a control group, making it purely descriptive rather than analytical. **Why other options are incorrect:** * **Case-control study (B):** This requires a **comparison group** (controls) of people without thyroid cancer to compare the frequency of radiation exposure. Since no control group is mentioned, it cannot be a case-control study. * **Case report (C):** A case report describes a single patient or a very small number of patients (usually <10) with a unique or unusual presentation. A sample size of 3500 is far too large for a case report. * **Clinical trial (D):** This is an interventional study where the researcher assigns an exposure (like a drug). In this scenario, the exposure (radiation) happened in the past, and the study is observational. **Clinical Pearls for NEET-PG:** * **Hierarchy of Evidence:** Case reports and case series are at the bottom of the evidence pyramid because they lack a control group. * **Key Identifier:** If a question describes a group of diseased individuals without a comparison group, it is a **Case Series**. * **Utility:** Case series are excellent for identifying the emergence of new diseases or rare side effects but cannot prove **causality** or calculate **Relative Risk/Odds Ratio**. * **Directionality:** Case series are retrospective in nature (looking back at history/records).

Q: While analyzing data, allocation into similar groups is done to ensure what?

Comparability. ### Explanation **1. Why Comparability is Correct:** In epidemiological studies, particularly in **Randomized Controlled Trials (RCTs)** and **Case-Control studies**, the primary goal of allocation (randomization or matching) is to ensure that the study and control groups are as similar as possible in all respects except for the intervention or exposure being studied. This process ensures **Comparability**. When groups are comparable, any difference in the outcome can be confidently attributed to the intervention rather than to baseline differences or "confounding variables." **2. Analysis of Incorrect Options:** * **Accuracy:** Refers to how close a measurement is to the true value. While proper allocation reduces bias, accuracy is more closely related to the precision of instruments and data collection methods. * **Validity:** This is a broader term. While comparability contributes to *Internal Validity* (the degree to which the results are true for the study population), allocation itself is the specific mechanism used to achieve the prerequisite state of comparability. * **Sensitivity:** This is a measure of a screening or diagnostic test's ability to correctly identify those with the disease (True Positives). It is unrelated to the grouping of subjects in data analysis. **3. NEET-PG High-Yield Pearls:** * **Randomization** is the "heart" of an RCT because it eliminates **selection bias** and ensures comparability of both known and unknown confounders. * **Matching** is the technique used in Case-Control studies to ensure comparability between cases and controls. * **Confounding** occurs when the groups are *not* comparable, leading to a distortion of the true association between exposure and outcome. * **Blinding** is done to maintain the integrity of the study *after* allocation, whereas **Allocation** is done to ensure baseline similarity.

Q: What is the population attributable risk given the incidence rate of lung cancer among smokers is 8 per 1000 and among non-smokers is 1 per 1000?

89%. ### Explanation **1. Understanding the Concept** The question asks for the **Attributable Risk (AR)**, also known as Risk Difference. This measures the amount of disease incidence that can be attributed directly to a specific exposure (smoking). It represents the potential reduction in disease if the exposure were eliminated. **Formula:** $$AR = \frac{\text{Incidence in Exposed} (I_e) - \text{Incidence in Non-exposed} (I_o)}{\text{Incidence in Exposed} (I_e)} \times 100$$ **Calculation:** * $I_e$ (Smokers) = 8 per 1000 * $I_o$ (Non-smokers) = 1 per 1000 * $AR = \frac{8 - 1}{8} \times 100 = \frac{7}{8} \times 100 = 87.5\%$ Rounding to the nearest option, **89% (Option A)** is the correct choice. This means 87.5% of lung cancer cases among smokers are specifically due to smoking. **2. Analysis of Incorrect Options** * **Option B (95%):** This would require a much higher ratio between exposed and non-exposed (e.g., 20 per 1000 vs 1 per 1000). * **Option C (10%):** This would occur if the incidence rates were very close (e.g., 1.1 vs 1.0), suggesting the exposure has a weak association with the disease. * **Option D (100%):** This is only possible if the incidence in non-exposed is zero, which is clinically impossible for lung cancer. **3. High-Yield Clinical Pearls for NEET-PG** * **Relative Risk (RR):** Measures the *strength* of association (Formula: $I_e / I_o$). Here, $RR = 8/1 = 8$. It is best for identifying etiological roles. * **Attributable Risk (AR):** Measures the *public health impact*. It tells us how much of the disease can be prevented by removing the risk factor. * **Population Attributable Risk (PAR):** Differs from AR as it considers the prevalence of the exposure in the total population, not just the exposed group. * **Key Distinction:** RR is used in **Cohort studies**, while Odds Ratio (OR) is used in **Case-control studies**.

Q: Which is the ideal and most scientific setting for an epidemiological study?

Experimental study. ### Explanation **Why the Correct Answer is Right:** An **Experimental study (Randomized Controlled Trial)** is considered the "gold standard" and the most scientific setting in epidemiology. Its superiority lies in the **principle of manipulation and randomization**. Unlike observational studies, the investigator actively controls the exposure (intervention). Randomization ensures that both known and unknown confounding factors are equally distributed between the study and control groups. This allows for the most accurate determination of **causality** (the direct link between an intervention and an outcome) with the highest degree of internal validity. **Why the Other Options are Wrong:** * **Cohort Study (B):** While it is the best *observational* study for determining incidence and relative risk, it is prone to selection bias and confounding because the investigator does not manipulate the exposure. * **Case-Control Study (A):** This is a retrospective study used to generate hypotheses. It is prone to recall bias and cannot determine incidence, making it less "scientific" than experimental designs. * **Cross-sectional Study (D):** This provides a "snapshot" of a population (prevalence). Because it measures exposure and outcome simultaneously, it cannot establish a temporal relationship, which is a key requirement for scientific causality. **NEET-PG High-Yield Pearls:** * **Hierarchy of Evidence:** Meta-analysis > Systematic Reviews > RCT (Experimental) > Cohort > Case-Control > Cross-sectional. * **Randomization** is the "heart" of a clinical trial; it eliminates **selection bias**. * **Blinding** is used in experimental studies to eliminate **investigator and participant bias**. * Experimental studies are the only way to definitively prove a **cause-and-effect relationship**.

Question 1

Mammography has 90% sensitivity and 98% specificity for breast carcinoma. What is the probability that a woman with breast carcinoma remains undiagnosed for 2 consecutive years?

Accepted Answer

1%

Answer

2%

Answer

2%

Question 2

Twenty-five children are chosen randomly from a school playground. Their heights are measured in 2 consecutive months. What statistical test could be used to test the hypothesis that their heights are not different in consecutive months?

Accepted Answer

Paired t-test

Answer

Unpaired t-test

Answer

Z-test

Answer

Regression

Question 3

For a group of 250 subjects, what value represents the 40th percentile?

Accepted Answer

100

Answer

7

Answer

40

Answer

140

Question 4

A screening test for colon cancer was applied to 1000 individuals. The test was positive in 320 and negative in 680. Of those with a positive test, 60 actually had the disease. Of those with a negative test, 20 actually had the disease. Calculate the positive predictive value of the test.

Accepted Answer

81%

Answer

12%

Answer

75%

Answer

28%

Question 5

If the American Heart Association lowered the cut-off threshold for hypertension, what would be the effect on the properties of blood pressure assessment as a screening test?

Accepted Answer

The negative predictive value will rise

Answer

The false negative rate will rise

Answer

The positive predictive value will rise

Answer

The sensitivity will fall

Question 6

A study surveyed 3500 patients with thyroid cancer regarding their past exposure to radiation. The study design most appropriately illustrates which of the following?

Accepted Answer

Case series report

Answer

Case control study

Answer

Case report

Answer

Clinical trial

Question 7

In a prospective study comprising 10,000 subjects, 6000 subjects were assigned to receive beta carotene and 4000 were not. Three out of the 6000 subjects receiving beta carotene developed lung cancer, and two out of the 4000 subjects not receiving beta carotene developed lung cancer. What is the interpretation of these results?

Accepted Answer

Beta carotene and lung cancer have no statistically significant relation.

Answer

Beta carotene is protective against lung cancer.

Answer

The study design is not sufficient to draw any meaningful conclusions.

Answer

Beta carotene is carcinogenic.

Question 8

While analyzing data, allocation into similar groups is done to ensure what?

Accepted Answer

Comparability

Answer

Accuracy

Answer

Validity

Answer

Sensitivity

Question 9

What is the population attributable risk given the incidence rate of lung cancer among smokers is 8 per 1000 and among non-smokers is 1 per 1000?

Accepted Answer

89%

Answer

95%

Answer

10%

Answer

100%

Question 10

Which is the ideal and most scientific setting for an epidemiological study?

Accepted Answer

Experimental study

Answer

Case-Control study

Answer

Cohort study

Answer

Cross-sectional study

Biostatistics — MCQs

Biostatistics — MCQs

On this page

Practice by Chapter

Want unlimited practice?