Biostatistics Practice Questions

Q: Which of the following statements regarding Receiver Operating Characteristic (ROC) curves is FALSE?

None of the above statements are false.. ### Explanation The **Receiver Operating Characteristic (ROC) curve** is a graphical plot used to evaluate the performance of a diagnostic or screening test. It is created by plotting the **Sensitivity (True Positive Rate)** on the y-axis against **1-Specificity (False Positive Rate)** on the x-axis for various cut-off points. **Why the correct answer is "None of the above":** All three statements (A, B, and C) are fundamentally true in the context of epidemiology: * **Statement A is True:** As you change the cut-off value to increase sensitivity, specificity invariably decreases, and vice versa. The curve visually maps this trade-off. * **Statement B is True:** A curve that shifts toward the upper-left corner (closer to the y-axis) represents a better test. The **Area Under the Curve (AUC)** quantifies this; an AUC of 1.0 is a perfect test, while 0.5 represents a test no better than chance. * **Statement C is True:** The slope of the tangent at any point on the ROC curve represents the **Likelihood Ratio (LR)** for that specific cut-off value. **Analysis of Options:** Since all statements are scientifically accurate, Option D is the only logical choice. There are no false statements provided in the list. **High-Yield Clinical Pearls for NEET-PG:** * **X-axis:** 1-Specificity (False Positive Rate). * **Y-axis:** Sensitivity (True Positive Rate). * **Ideal Test:** The closer the curve is to the top-left corner, the higher the accuracy. * **Diagonal Line (45°):** Represents a "worthless test" (AUC = 0.5). * **Use Case:** ROC curves are the gold standard for determining the **optimal cut-off point** for a continuous variable (e.g., blood glucose levels for diagnosing Diabetes).

Q: What is the relationship between mosquito population density and increasing altitude?

Negative correlation. ### Explanation **1. Why Negative Correlation is Correct:** In biostatistics, a **negative correlation** (inverse relationship) occurs when one variable increases while the other decreases. As **altitude increases**, the **mosquito population density decreases**. This is primarily due to the drop in ambient temperature and atmospheric pressure at higher elevations. Mosquitoes are ectothermic (cold-blooded); lower temperatures inhibit their metabolic rates, slow down larval development, and reduce the frequency of blood-feeding. Furthermore, extreme cold at very high altitudes is lethal to most mosquito species, leading to a natural decline in their numbers as one moves upward. **2. Analysis of Incorrect Options:** * **Positive Correlation:** This would imply that mosquito density increases as you go higher (e.g., more mosquitoes on a mountain peak than in a valley), which is biologically incorrect. * **Bidirectional Relationship:** This suggests that both variables influence each other in a complex loop. While altitude affects mosquitoes, mosquitoes do not affect the altitude of a geographical location. * **Zero Correlation:** This would mean there is no linear relationship between height and mosquito density, which contradicts established entomological data. **3. High-Yield Facts for NEET-PG:** * **Correlation Coefficient (r):** Ranges from -1 to +1. A negative correlation has an 'r' value between 0 and -1. * **Malaria Transmission:** The "malaria limit" is typically around 2,000–2,500 meters. Above this, the *Anopheles* mosquito struggle to survive or complete the sporogonic cycle of *Plasmodium*. * **Climate Change Impact:** Global warming is currently shifting this "negative correlation" boundary, allowing mosquitoes to survive at higher altitudes than previously recorded (altitudinal range expansion). * **Scatter Diagram:** On a graph, a negative correlation is represented by a line sloping downwards from left to right.

Q: Which of the following equations correctly relates incidence, prevalence, and duration of a disease in a stable situation?

Prevalence = Incidence x Duration. ### Explanation **1. Understanding the Relationship (The "Bathtub" Analogy)** In biostatistics, the relationship between incidence and prevalence is best understood through the **Steady State Model**. * **Incidence (I):** Represents the rate of *new* cases entering the population. * **Duration (D):** Represents how long a person stays in the "diseased state" before recovery or death. * **Prevalence (P):** Represents the total *existing* cases at a given time. In a stable population (where the number of people entering the diseased state equals the number leaving it), **Prevalence = Incidence × Mean Duration (P = I × D)**. This is because the total pool of disease depends on how fast new cases occur and how long they persist. **2. Analysis of Options** * **Option B (Correct):** Correctly reflects that prevalence is a product of the frequency of new cases and their longevity. * **Option A (Incorrect):** This suggests that incidence increases with duration, which is logically flawed. Incidence is determined by risk factors, not how long a disease lasts. * **Options C & D (Incorrect):** These suggest an additive relationship. In epidemiology, these variables are multiplicative; if the duration of a disease doubles (e.g., due to better life-prolonging treatment), the prevalence will also double, even if the incidence remains the same. **3. NEET-PG Clinical Pearls & High-Yield Facts** * **The Rule of Thumb:** * If a treatment **cures** a disease quickly, **Duration ↓** and **Prevalence ↓**. * If a treatment **prevents death** but doesn't cure (e.g., Insulin for Diabetes), **Duration ↑** and **Prevalence ↑**. * **Prevalence** is a measure of **burden** of disease (useful for healthcare planning). * **Incidence** is a measure of **risk** (useful for determining etiology). * **Note:** This formula (P = I × D) is only valid when prevalence is low (less than 10%).

Q: The Child Pugh score is a type of what kind of scale?

Ordinal. **Explanation:** The **Child-Pugh Score** is used to assess the prognosis of chronic liver disease (cirrhosis). It is an **Ordinal Scale** because it categorizes patients into groups (Class A, B, or C) based on a numerical score derived from five parameters (Bilirubin, Albumin, INR, Ascites, and Encephalopathy). 1. **Why Ordinal?** An ordinal scale represents data that has a **natural order or rank**, but the mathematical distance between the ranks is not necessarily equal. In the Child-Pugh system, Class A (5–6 points) is "better" than Class B (7–9 points), which is "better" than Class C (10–15 points). Since there is a clear progression of severity, it is ordinal. 2. **Why other options are incorrect:** * **Nominal:** This scale is for naming or labeling categories without any quantitative value or order (e.g., Blood groups A, B, O; Gender). Since Child-Pugh implies a "rank" of severity, it is not nominal. * **Quantitative (Interval/Ratio):** These scales involve precise measurements where the difference between numbers is consistent (e.g., Height in cm, Weight in kg). While the score uses numbers, the "gap" between Class A and B isn't a physical measurement. * **Qualitative:** While ordinal data is a type of categorical (qualitative) data, "Ordinal" is the more specific and correct biostatistical term for ranked data. **High-Yield Clinical Pearls for NEET-PG:** * **Parameters of Child-Pugh (Mnemonic: ABCDE):** **A**lbumin, **B**ilirubin, **C**oagulation (INR), **D**istension (Ascites), **E**ncephalopathy. * **APGAR Score** and **Glasgow Coma Scale (GCS)** are also classic examples of **Ordinal Scales** frequently asked in exams. * **Visual Analogue Scale (VAS)** for pain is another high-yield Ordinal scale.

Q: While testing a hypolipidemic drug, serum lipid levels were tested both before and after its use. Which statistical test is best suited for the analysis of the results?

Paired T test. ### Explanation **Why Paired T-test is the Correct Answer:** The Paired T-test is used to compare the **means of two related groups** (dependent samples). In this scenario, the serum lipid levels are measured in the **same set of individuals** at two different time points: "before" and "after" the intervention. Since each subject acts as their own control, the data points are paired. This test determines if the mean difference between these two observations is statistically significant. **Analysis of Incorrect Options:** * **Student’s T-test (Unpaired/Independent T-test):** This is used to compare the means of two **independent** groups (e.g., comparing lipid levels between Group A taking the drug and Group B taking a placebo). It is incorrect here because the measurements are taken from the same group. * **Chi-square Test:** This is a non-parametric test used for **categorical (qualitative) data** to compare proportions (e.g., "improved" vs. "not improved"). Since serum lipid levels are **quantitative (numerical)** data, the Chi-square test is inappropriate. **Clinical Pearls & High-Yield Facts for NEET-PG:** 1. **Quantitative Data (Means):** * 2 groups (Related/Before-After) → **Paired T-test** * 2 groups (Independent/Different people) → **Unpaired T-test** * >2 groups (Independent) → **ANOVA** (Analysis of Variance) 2. **Qualitative Data (Proportions):** * Comparing two or more proportions → **Chi-square Test** * Small samples (any cell value <5) → **Fisher’s Exact Test** 3. **Key Concept:** Always identify the **type of data** (Numerical vs. Categorical) and the **relationship** (Dependent vs. Independent) before choosing a statistical test.

Q: A diagnostic test was performed. Out of 80 individuals who tested positive, 40 did not have the disease. Out of 9920 individuals who tested negative, 9840 did not have the disease. What is the sensitivity of the test?

33%. ### Explanation To solve this problem, we must first organize the data into a standard **2x2 contingency table**. | | Disease Present (+) | Disease Absent (-) | Total | | :--- | :---: | :---: | :---: | | **Test Positive (+)** | **40 (TP)** | 40 (FP) | 80 | | **Test Negative (-)** | 80 (FN) | **9840 (TN)** | 9920 | | **Total** | 120 | 9880 | 10000 | **Step-by-Step Calculation:** 1. **True Positives (TP):** Total test positives (80) minus False Positives (40) = **40**. 2. **False Negatives (FN):** Total test negatives (9920) minus True Negatives (9840) = **80**. 3. **Sensitivity Formula:** $\frac{TP}{TP + FN} \times 100$ 4. **Calculation:** $\frac{40}{40 + 80} = \frac{40}{120} = \frac{1}{3} \approx \mathbf{33.3\%}$. #### Why the Correct Answer is Right: Sensitivity (True Positive Rate) measures the ability of a test to correctly identify those with the disease. In this cohort, there are 120 diseased individuals, but the test only caught 40 of them, resulting in 33%. #### Why Other Options are Wrong: * **Option A (13%):** Incorrect calculation; likely a result of misplacing values in the 2x2 table. * **Option C (50%):** This is the **Positive Predictive Value (PPV)**: $\frac{TP}{TP+FP} = \frac{40}{80} = 50\%$. * **Option D (99%):** This is the **Specificity**: $\frac{TN}{TN+FP} = \frac{9840}{9840+40} \approx 99.6\%$. #### High-Yield Clinical Pearls for NEET-PG: * **SNOUT:** **S**ensitivity rules **OUT** disease (when a highly sensitive test is negative). * **SPIN:** **S**pecificity rules **IN** disease (when a highly specific test is positive). * **Prevalence Impact:** Sensitivity and Specificity are **independent** of disease prevalence, whereas PPV and NPV are directly affected by it. * **Screening vs. Diagnosis:** Screening tests require high sensitivity; confirmatory tests require high specificity.

Question 1

Which of the following statements regarding Receiver Operating Characteristic (ROC) curves is FALSE?

Accepted Answer

None of the above statements are false.

Answer

The ROC curve demonstrates the trade-off between sensitivity and specificity.

Answer

A higher ROC curve indicates a better screening test or cut-off value.

Answer

The ROC curve represents the likelihood ratio of a positive test.

Question 2

A community has a population of 10,000 and a birth rate of 36 per 1000. If 5 maternal deaths were reported in the current year, calculate the Maternal Mortality Rate (MMR).

Accepted Answer

13.8

Answer

14.5

Answer

20

Answer

5

Question 3

What is the relationship between mosquito population density and increasing altitude?

Accepted Answer

Negative correlation

Answer

Positive correlation

Answer

Bidirectional relationship

Answer

Zero correlation

Question 4

Mrs. Rekha has a 0.50 probability of having a baby with a birth weight (BW) less than 2500 grams and a 0.20 probability of having a baby with a BW between 2500 and 2999 grams. What is the probability that Mrs. Rekha will have a baby with a BW less than 3 kg?

Accepted Answer

0.7

Answer

0.3

Answer

1

Answer

0.1

Question 5

Which of the following equations correctly relates incidence, prevalence, and duration of a disease in a stable situation?

Accepted Answer

Prevalence = Incidence x Duration

Answer

Incidence = Prevalence x Duration

Answer

Incidence = Prevalence + Duration

Answer

Prevalence = Incidence + Duration

Question 6

In a study to establish smoking as a risk factor for disease, 30 out of 50 smokers developed disease while 10 out of 50 non-smokers developed disease. Calculate the Odds Ratio.

Accepted Answer

5

Answer

3

Answer

6

Answer

10

Question 7

The Child Pugh score is a type of what kind of scale?

Accepted Answer

Ordinal

Answer

Nominal

Answer

Quantitative

Answer

Qualitative

Question 8

While testing a hypolipidemic drug, serum lipid levels were tested both before and after its use. Which statistical test is best suited for the analysis of the results?

Accepted Answer

Paired T test

Answer

Student's T test

Answer

Chi-square test

Answer

None of the above

Question 9

In a case-control study of a suspected association between breast cancer and the contraceptive pill, all of the following are true statements except?

Accepted Answer

The attributable risk of breast cancer resulting from the pill may be directly measured.

Answer

The control should come from a population that has the same potential for breast cancer as the cases.

Answer

The control should exclude women known to be taking the pill at the time of the survey.

Answer

All the controls need to be healthy.

Question 10

A diagnostic test was performed. Out of 80 individuals who tested positive, 40 did not have the disease. Out of 9920 individuals who tested negative, 9840 did not have the disease. What is the sensitivity of the test?

Accepted Answer

33%

Answer

13%

Answer

50%

Answer

99%

Biostatistics — MCQs

Biostatistics — MCQs

On this page

Practice by Chapter

Want unlimited practice?