Biostatistics Practice Questions

Q: In a case-control study, which of the following cannot be directly measured?

Relative risk. In a **Case-Control Study**, researchers start with the outcome (cases) and look backward to determine exposure. Because the study begins with people who already have the disease, the **Incidence** (number of new cases over time) cannot be calculated. **Why Relative Risk is the correct answer:** Relative Risk (RR) is the ratio of the incidence of disease in the exposed group to the incidence in the non-exposed group ($RR = \frac{I_e}{I_u}$). Since a case-control study does not provide the "population at risk" to calculate incidence, **Relative Risk cannot be directly measured.** It can only be estimated using the Odds Ratio, provided the disease is rare. **Analysis of Incorrect Options:** * **B. Incidence:** While incidence also cannot be measured, the question specifically targets the primary measure of association. In most NEET-PG contexts, RR is the classic answer for what cannot be derived from case-control data. * **C. Odds Ratio (OR):** This is the **primary measure of association** in case-control studies. It calculates the odds of exposure among cases versus controls. * **D. Cause and effect relationship:** While case-control studies are weaker than Cohort studies or RCTs for establishing causality (due to recall bias and lack of temporality), they can still suggest a relationship. However, the mathematical impossibility of calculating RR is the definitive epidemiological rule. **High-Yield Clinical Pearls for NEET-PG:** * **Case-Control:** Proceeds from **Effect to Cause**. Best for **rare diseases**. * **Cohort Study:** Proceeds from **Cause to Effect**. Can directly measure **Incidence, Relative Risk, and Attributable Risk**. * **Matching:** Used in case-control studies to eliminate **confounding factors**. * **Recall Bias:** The most common bias encountered in case-control studies.

Q: In a village of 1 lakh population, among 20,000 individuals exposed to smoking, 200 developed cancer. Among 40,000 individuals not exposed to smoking, 40 developed cancer. What is the relative risk of smoking in the development of cancer?

10. ### Explanation **1. Understanding the Correct Answer (B: 10)** Relative Risk (RR), also known as Risk Ratio, is the ratio of the incidence of a disease among the exposed group to the incidence of the disease among the non-exposed group. It measures the strength of association between a risk factor and a disease. **Formula:** $$RR = \frac{\text{Incidence among exposed (Ie)}}{\text{Incidence among non-exposed (Io)}}$$ * **Step 1: Calculate Incidence in Exposed ($I_e$):** $I_e = \frac{200 \text{ (cases)}}{20,000 \text{ (total exposed)}} = 0.01$ (or 10 per 1,000) * **Step 2: Calculate Incidence in Non-exposed ($I_o$):** $I_o = \frac{40 \text{ (cases)}}{40,000 \text{ (total non-exposed)}} = 0.001$ (or 1 per 1,000) * **Step 3: Calculate RR:** $RR = \frac{0.01}{0.001} = 10$ An RR of 10 signifies that smokers are 10 times more likely to develop cancer compared to non-smokers. **2. Why Other Options are Incorrect** * **Options A (20), C (5), and D (15):** These are incorrect mathematical derivations. They often result from common student errors such as dividing the total population by cases, comparing only the number of cases (200/40 = 5), or failing to account for the different denominators in the two groups. **3. NEET-PG High-Yield Pearls** * **Study Design:** Relative Risk is calculated from **Cohort Studies**. * **Interpretation:** * $RR > 1$: Positive association (Risk factor). * $RR = 1$: No association. * $RR < 1$: Negative association (Protective factor). * **Attributable Risk (AR):** Measures the impact of a risk factor. Formula: $\frac{I_e - I_o}{I_e} \times 100$. In this case, AR is 90%, meaning 90% of cancer in smokers is due to smoking. * **Odds Ratio (OR):** Used in Case-Control studies as an estimate of RR.

Q: What is the probability to achieve a live birth within a single cycle?

Fecundity. ### Explanation The correct answer is **Fecundity**. In demography and biostatistics, it is crucial to distinguish between the biological potential and the actual realization of birth. **1. Why Fecundity is Correct:** **Fecundity** is defined as the physiological capacity of a woman (or a couple) to participate in reproduction and achieve a **live birth** within a single reproductive cycle. It represents the biological potential for successful reproduction ending in a live neonate. **2. Analysis of Incorrect Options:** * **Fertility (Option A):** In medical statistics, fertility refers to the **actual reproductive performance** (the number of live births achieved) rather than the capacity. While fecundity is the *potential*, fertility is the *realization*. * **Fecundability (Option C):** This is a high-yield distractor. Fecundability is the probability of **conception** (becoming pregnant) within a single menstrual cycle. It does not guarantee a live birth, as the pregnancy may end in miscarriage or stillbirth. * **Sterility (Option D):** This is the absolute physiological inability to conceive or produce a live birth. It is the opposite of fecundity. **3. High-Yield Clinical Pearls for NEET-PG:** * **Fecundability vs. Fecundity:** Think of Fecundability as "Probability of Pregnancy" and Fecundity as "Probability of Live Birth." * **Total Fertility Rate (TFR):** The average number of children a woman would have if she were to pass through her reproductive years (15–49 years) bearing children according to age-specific fertility rates. * **Replacement Level Fertility:** The TFR at which a population exactly replaces itself from one generation to the next (Value = **2.1**). * **Net Reproduction Rate (NRR):** The number of daughters a newborn girl will bear during her lifetime. An NRR of **1** is the demographic goal for population stabilization.

Q: What is true regarding Pictogram?

Method to represent data to the 'man in the street'.. ### Explanation **Correct Answer: D. Method to represent data to the 'man in the street'.** A **Pictogram** is a visual representation of data using small pictures or symbols. Each symbol represents a specific number of units (e.g., a "person" icon representing 1,000 people). It is considered the most popular and simplest method of data presentation because it is easily understood by the general public, including those without a background in statistics—often referred to as the **"man in the street."** #### Analysis of Incorrect Options: * **Option A (Areas shaded with different colors):** This describes a **Choropleth Map**. These maps use different shades or colors to represent the density or prevalence of a disease in different geographical areas. * **Option B (Trends of events over time):** This describes a **Line Diagram** (or Line Graph). Line diagrams are the gold standard for showing chronological changes or trends (e.g., birth rates over decades). * **Option C (Pictorial diagram of frequency distribution):** This is a distractor. While a pictogram shows frequency, the standard pictorial representation of a frequency distribution is a **Histogram**. #### High-Yield NEET-PG Pearls: * **Simplest method for laymen:** Pictogram. * **Best for trends over time:** Line Diagram. * **Best for comparing two or more discrete categories:** Bar Chart. * **Best for showing parts of a whole (proportions):** Pie Chart. * **Correlation between two continuous variables:** Scatter Diagram (Dot diagram). * **Frequency distribution of continuous data:** Histogram. * **Frequency distribution of discrete data:** Bar Chart.

Q: A study reports patient satisfaction levels as 'Satisfied', 'Very satisfied', and 'Dissatisfied'. Which type of scale best represents these categories?

Ordinal. ### Explanation **Why Ordinal is Correct:** The data presented (Satisfied, Very satisfied, Dissatisfied) represents **Ordinal data**. In biostatistics, an ordinal scale is used when data can be categorized into distinct groups that have a **natural rank or inherent order**, but the mathematical distance between the categories is not uniform or measurable. In this study, "Very satisfied" is clearly higher than "Satisfied," which is higher than "Dissatisfied," but you cannot quantify exactly *how much* more satisfied one patient is compared to another. **Why Other Options are Incorrect:** * **Nominal:** This scale is for naming or labeling categories without any inherent order (e.g., Blood groups A, B, O; Gender; or Color of eyes). Since satisfaction levels have a logical hierarchy, they are not merely nominal. * **Interval:** This scale has a defined order and equal intervals between values, but **no true zero point** (e.g., Temperature in Celsius). Satisfaction levels do not have measurable, equal mathematical intervals. * **Ratio:** This is the highest level of measurement. It has all the properties of an interval scale plus a **true zero point** (e.g., Height, Weight, Blood Pressure). Satisfaction levels cannot be zero in a mathematical sense. **Clinical Pearls & High-Yield Facts for NEET-PG:** * **Mnemonic for Scales (Lowest to Highest Complexity):** **NOIR** (**N**ominal → **O**rdinal → **I**nterval → **R**atio). * **Likert Scales:** Most surveys using "Strongly Agree" to "Strongly Disagree" are classic examples of **Ordinal** data. * **Cancer Staging:** TNM staging or WHO functional grades are **Ordinal** scales. * **Qualitative vs. Quantitative:** Nominal and Ordinal are **Qualitative (Categorical)**, while Interval and Ratio are **Quantitative (Numerical)**. * **Central Tendency:** For Ordinal data, the **Median** is the most appropriate measure of central tendency.

Q: Amount of alcohol consumption amongst a group of alcoholics, before and after intervention, was recorded. What is the most appropriate statistical test to assess 'significant change' resulting from the intervention program?

Paired t-test. ### Explanation The core of this question lies in identifying the type of data and the relationship between the study groups. **1. Why Paired t-test is correct:** * **Type of Data:** "Amount of alcohol consumption" is a **quantitative (numerical/continuous)** variable (e.g., ml/day). * **Study Design:** The measurements are taken from the **same group** of individuals at two different points in time (**Before and After** intervention). These are "dependent" or "paired" observations. * **Purpose:** To compare the means of two related groups to determine if the intervention caused a statistically significant change, the **Paired t-test** is the standard parametric test used. **2. Why other options are incorrect:** * **Unpaired (Independent) t-test:** This is used to compare the means of two **independent** groups (e.g., comparing alcohol intake between Group A and Group B). * **Chi-square test:** This is used for **qualitative (categorical)** data to compare proportions (e.g., comparing the number of "drinkers" vs. "non-drinkers" in two groups). It cannot analyze the "amount" of consumption directly. * **McNemar test:** This is used for **paired qualitative** data. It would be appropriate if we were looking at a "Yes/No" change in addiction status before and after treatment, rather than the specific amount consumed. ### High-Yield Clinical Pearls for NEET-PG: * **Quantitative Data + 2 Groups:** * Paired (Before/After) $\rightarrow$ **Paired t-test** * Unpaired (Group A vs B) $\rightarrow$ **Unpaired t-test** * **Quantitative Data + >2 Groups:** Use **ANOVA**. * **Qualitative Data + 2 Groups:** * Unpaired $\rightarrow$ **Chi-square test** * Paired $\rightarrow$ **McNemar test** * **Non-parametric alternative:** If the data is not normally distributed, the non-parametric alternative to the Paired t-test is the **Wilcoxon Signed-Rank test**.

Q: The Hb level in healthy women has a mean of 13.5 g/dl and a standard deviation of 1.5 g/dl. What is the Z-score for a woman with an Hb level of 15.0 g/dl?

1.0. ### Explanation **Concept:** The **Z-score** (Standard Score) is a fundamental biostatistical tool used to determine how many standard deviations a specific observation is from the mean. It allows us to compare individual data points within a normal distribution. The formula for calculating the Z-score is: $$Z = \frac{X - \mu}{\sigma}$$ *Where:* * **X** = Individual value (15.0 g/dl) * **μ (Mean)** = Average value (13.5 g/dl) * **σ (Standard Deviation)** = 1.5 g/dl **Calculation:** $Z = \frac{15.0 - 13.5}{1.5} = \frac{1.5}{1.5} = \mathbf{1.0}$ A Z-score of +1.0 indicates that the woman's Hb level is exactly **one standard deviation above the mean**. --- ### Analysis of Options: * **Option D (1.0) is Correct:** As calculated above, the difference between the value and the mean equals exactly one unit of standard deviation. * **Option C (2.0) is Incorrect:** This would require an Hb level of 16.5 g/dl ($13.5 + [2 \times 1.5]$). * **Options A (9.0) and B (10.0) are Incorrect:** These represent extreme outliers. In a normal distribution, 99.7% of all values fall within a Z-score of ±3. A Z-score of 10 would be physiologically improbable in this context. --- ### High-Yield Clinical Pearls for NEET-PG: 1. **Normal Distribution (Gaussian Curve):** * Mean ± 1 SD covers **68.3%** of values. * Mean ± 2 SD covers **95.4%** of values. * Mean ± 3 SD covers **99.7%** of values. 2. **Z-score of 0:** This means the individual's value is exactly equal to the mean. 3. **Standard Normal Distribution:** A specific normal distribution where the **Mean is 0** and the **Standard Deviation is 1**. 4. **Application:** Z-scores are clinically used in **WHO Growth Charts** (e.g., Weight-for-height Z-scores) to diagnose malnutrition (Wasting/Stunting).

Q: In a study, a new treatment resulted in 36 deaths or treatment failures out of a sample of 120. With an alternative treatment, 26 treatment failures were reported from a sample size of 130. How many patients should be treated with the first treatment to avert one death?

10. ### Explanation The question asks for the number of patients needed to be treated with the first treatment to avert one death, which is the definition of **Number Needed to Treat (NNT)**. **1. Calculation of the Correct Answer (B):** * **Event Rate in Group 1 (EER):** 36/120 = 0.3 (30%) * **Event Rate in Group 2 (CER):** 26/130 = 0.2 (20%) * **Absolute Risk Reduction (ARR):** This is the arithmetic difference between the two event rates. * ARR = |EER – CER| = |0.3 – 0.2| = **0.1** * **Number Needed to Treat (NNT):** This is the inverse of ARR. * NNT = 1 / ARR = 1 / 0.1 = **10** * Therefore, 10 patients must be treated with the first treatment to prevent one additional death compared to the second treatment. **2. Why other options are incorrect:** * **Option A (100):** This is a common calculation error where the decimal point is misplaced (1/0.01 instead of 1/0.1). * **Option C (250):** This value does not correlate with the mathematical relationship between the provided sample sizes and outcomes. * **Option D (160):** This might result from incorrectly adding the total number of failures (36+26) or misapplying the formula. **3. Clinical Pearls & High-Yield Facts for NEET-PG:** * **NNT Formula:** $1 / \text{Absolute Risk Reduction (ARR)}$. * **Interpretation:** The lower the NNT, the more effective the treatment. An NNT of 1 means every single patient treated benefits. * **Number Needed to Harm (NNH):** Calculated similarly ($1 / \text{Absolute Risk Increase}$), it indicates how many patients must be exposed to a risk factor to cause one adverse event. * **Relative Risk Reduction (RRR):** Unlike ARR, RRR does not account for the baseline risk and can often make a treatment's effect look more impressive than it is.

Question 1

What does the General Fertility Rate indicate?

Accepted Answer

The number of live births per 1,000 women of reproductive age (usually 15-49 years).

Answer

The average number of children a woman is expected to have in her lifetime.

Answer

A measure of the number of births in a population relative to the total population.

Answer

All of the above.

Question 2

In a case-control study, which of the following cannot be directly measured?

Accepted Answer

Relative risk

Answer

Incidence

Answer

Odds ratio

Answer

Cause and effect relationship

Question 3

In a village of 1 lakh population, among 20,000 individuals exposed to smoking, 200 developed cancer. Among 40,000 individuals not exposed to smoking, 40 developed cancer. What is the relative risk of smoking in the development of cancer?

Accepted Answer

10

Answer

20

Answer

5

Answer

15

Question 4

What is the probability to achieve a live birth within a single cycle?

Accepted Answer

Fecundity

Answer

Fertility

Answer

Fecundability

Answer

Sterility

Question 5

What is true regarding Pictogram?

Accepted Answer

Method to represent data to the 'man in the street'.

Answer

Areas are shaded with different colours.

Answer

Used to show the trends of events with the passage of time.

Answer

A pictorial diagram of frequency distribution.

Question 6

What is a frequency curve?

Accepted Answer

When the number of observations is large and the group interval is reduced, a frequency polygon loses its angulations and becomes a curve.

Answer

It is a frequency polygon presenting variation by a line, showing the trend of an event over a period of time.

Answer

It is a graph of cumulative frequency distribution.

Answer

It is an area diagram of frequency distribution developed over a histogram.

Question 7

A study reports patient satisfaction levels as 'Satisfied', 'Very satisfied', and 'Dissatisfied'. Which type of scale best represents these categories?

Accepted Answer

Ordinal

Answer

Nominal

Answer

Interval

Answer

Ratio

Question 8

Amount of alcohol consumption amongst a group of alcoholics, before and after intervention, was recorded. What is the most appropriate statistical test to assess 'significant change' resulting from the intervention program?

Accepted Answer

Paired t-test

Answer

Unpaired t-test

Answer

Chi-square test

Answer

McNemar test

Question 9

The Hb level in healthy women has a mean of 13.5 g/dl and a standard deviation of 1.5 g/dl. What is the Z-score for a woman with an Hb level of 15.0 g/dl?

Accepted Answer

1.0

Answer

9.0

Answer

10.0

Answer

2.0

Question 10

In a study, a new treatment resulted in 36 deaths or treatment failures out of a sample of 120. With an alternative treatment, 26 treatment failures were reported from a sample size of 130. How many patients should be treated with the first treatment to avert one death?

Accepted Answer

10

Answer

100

Answer

250

Answer

160

Biostatistics — MCQs

Biostatistics — MCQs

On this page

Practice by Chapter

Want unlimited practice?