Biostatistics Practice Questions

Q: Which of the following is NOT a characteristic of a health index?

Affordability. ***Affordability*** - **Affordability** is not considered a defining characteristic or psychometric property of a health index itself. - The core characteristics of a health index relate to its **measurement properties** (validity, reliability, sensitivity) and **practical applicability** (feasibility), not economic considerations. - While cost may influence whether a health index is adopted in practice, it does not determine the index's ability to accurately measure health status. - Affordability is an external factor related to resource availability rather than an intrinsic quality of the measurement tool. *Validity* - **Validity** is a fundamental characteristic referring to whether a health index measures what it intends to measure. - Types include content validity, criterion validity, and construct validity. - Essential for ensuring the index accurately reflects the health concept being assessed. *Reliability* - **Reliability** is a core characteristic indicating consistency and reproducibility of measurements. - Includes test-retest reliability, inter-rater reliability, and internal consistency. - A reliable health index produces stable results when measuring the same phenomenon under similar conditions. *Feasibility* - **Feasibility** is a recognized characteristic referring to the practical usability of a health index in real-world settings. - Includes ease of administration, time requirements, scoring simplicity, and acceptability to users. - A feasible index can be implemented effectively within available resources and constraints, though this differs from the economic concept of affordability.

Q: In a positively skewed distribution, which of the following is true?

Mean > Median > Mode. ***Mean > Median > Mode*** - In a **positively skewed distribution**, the tail of the distribution is longer on the right side, meaning there are more extreme large values. - These large values pull the **mean** towards the right, making it greater than the **median**, which in turn is greater than the **mode**. *Mean = Median = Mode* - This equality holds true for a **symmetrical distribution**, such as a normal distribution, where data points are evenly distributed around the center. - A **positively skewed distribution** is asymmetrical, with a distinct longer tail on one side due to outliers. *Mode > Median > Mean* - This relationship is characteristic of a **negatively skewed distribution**, where the tail extends to the left, indicating a presence of more extreme small values. - In such a case, the **mode** is the largest and the **mean** is the smallest, pulled by the left-skewed tail. *None of the options* - This option is incorrect because the statement **Mean > Median > Mode** accurately describes the relationship between these measures of central tendency in a **positively skewed distribution**. - The other options describe different types of distributions.

Q: A screening test has sensitivity of 90% and specificity of 99%. The prevalence of disease under investigation is 5 per 1000 population. What is the PPV of the given screening test?

33. ***33*** * **Positive Predictive Value (PPV)** = \[Sensitivity \times Prevalence] / \[(Sensitivity \times Prevalence) + (1 – Specificity) \times (1 – Prevalence)] * Here, 0.9 x 0.005 / \[(0.9 x 0.005) + (1-0.99) x (1-0.005)] = 0.0045 / \[0.0045 + (0.01 x 0.995)] = 0.0045 / \[0.0045 + 0.00995] = 0.0045 / 0.01445 ≈ 0.3114 or 31.14%. Converting to a percentage closest to the answer choices, it would be 33%. *10* * This value would be obtained if there was an error in calculating either the **prevalence** of the disease or the contribution of false positives to the total positive tests. * Inaccuracies in the formula or arithmetic would lead to results far from the correct PPV. *70* * This value suggests a much higher **prevalence** or a significantly lower number of **false positives** than indicated by the given sensitivity, specificity, and prevalence. * Such a high PPV is inconsistent with a low disease prevalence (0.5%) and a relatively high false positive rate (1-specificity = 1%). *99* * This value is close to the given **specificity**, which is the probability of a true negative test among individuals without the disease, not the PPV. * A PPV of 99% would be extremely high for a disease with a prevalence of 0.5% and would typically require a much higher specificity and/or sensitivity.

Q: Which of the following statements is true about cohort studies?

A study that measures the incidence of a disease.. ***A study that measures the incidence of a disease.*** - Cohort studies are **prospective studies** that follow a defined group of individuals over time to observe the development of diseases or health outcomes [1]. - By following an at-risk population and documenting new cases, they can directly calculate the **incidence rate** of a disease [1]. - This is the **primary strength** of cohort studies - they provide the best epidemiological evidence for disease incidence. *A study that describes characteristics of a population.* - This describes **descriptive studies** or **cross-sectional surveys**, which characterize a population at a single point in time. - While cohort studies may initially describe baseline characteristics, their primary purpose is to observe disease occurrence over time, not just description. *A study that follows participants over time to observe outcomes but does not measure incidence.* - This is **contradictory** - the act of following participants over time and observing new disease cases IS the measurement of incidence [1]. - Incidence (new cases per unit of person-time) is precisely what cohort studies are designed to measure. *A study that can determine cause and effect.* - While cohort studies establish **temporal relationships** (exposure precedes outcome) and provide strong evidence for causality, the word "determine" is too absolute. - Establishing definitive causation requires **multiple lines of evidence**, including criteria like biological plausibility, dose-response relationships, and consistency across studies. - **Randomized controlled trials** provide stronger causal evidence due to randomization eliminating confounding.

Q: Which of the following is a parametric test?

Student t-test. ***Student t-test*** - The **Student t-test** is a **parametric test** used to compare the means of two groups. - It assumes the data is **normally distributed** and has **equal variances**. *Sign test* - The **Sign test** is a **non-parametric test** used for paired data. - It examines the direction of differences between pairs, not their magnitude. *Fisher exact test* - The **Fisher exact test** is a **non-parametric test** used for analyzing **categorical data** in a 2x2 contingency table. - It is particularly useful when **sample sizes are small** and the assumptions for a chi-square test are not met. *Chi-square test* - The **Chi-square test** is a **non-parametric test** used to assess independence between **categorical variables** or to compare observed frequencies with expected frequencies. - It does not assume any specific distribution for the data, making it suitable for nominal data.

Q: In a study a patient does not know the nature of drug [whether a placebo or curative drug] he is taking. The researcher knows the drug type to be given to the individuals in study. Type of blinding in this study is?

Single. ***Single*** - In **single-blinded studies**, the participant (patient) is unaware of whether they are receiving the experimental treatment or a placebo. - The investigator (researcher) and/or staff administering the treatment are aware of the treatment assignment. *Double* - In a **double-blinded study**, neither the participants nor the investigators/staff administering the treatment know who is receiving the experimental treatment and who is receiving the placebo. - This method helps to minimize bias from both participant expectations and researcher influence. *Triple* - A **triple-blinded study** extends double-blinding by also ensuring that those analyzing the data are unaware of the treatment assignments. - This adds an extra layer of protection against bias in the interpretation of results. *Combined double/triple* - This option refers to scenarios where a study design might aim to incorporate aspects of both double and triple blinding. - However, in this specific case, the researcher's knowledge of the drug type prevents it from being a fully double or triple-blinded study.

Q: What is the denominator used in calculating the under-5 proportional mortality rate?

Total deaths. ***Total deaths*** - The **under-5 proportional mortality rate** measures the proportion of all deaths that occur in children under five years of age. - The denominator for this rate is the **total number of deaths** in a given population across all age groups. *Number of deaths under 5 years of age* - This value represents the **numerator** of the under-5 proportional mortality rate. - It indicates the specific number of deaths within the under-5 age group, rather than the total deaths across all ages. *Mid-year under-5 population* - This is the denominator used in calculating the **under-5 mortality rate** (a true rate), not the proportional mortality rate. - It measures the risk of death in children under five relative to the population of children in that age group. *Mid-year population* - The mid-year population is often used as the denominator for **crude mortality rates** or **incidence rates** for specific diseases within a general population. - It represents the total population at risk, not solely the total number of deaths for proportional mortality.

Q: The Chi-square test is most commonly used in medical statistics to test for:

Test for independence of categorical variables. ***Test for independence of categorical variables*** - The **Chi-square test of independence** is the most commonly used application in medical research to determine if there is a statistically significant **association between two categorical variables**. - It assesses whether the observed frequencies in a contingency table differ significantly from the frequencies expected if the variables were independent. - **Common medical applications**: relationship between exposure and disease, association between risk factors and outcomes, comparing proportions across groups. *Test for goodness of fit* - The Chi-square **goodness of fit test** is another important application that tests whether observed frequencies of a **single categorical variable** match an expected theoretical distribution. - While this is a valid primary use of Chi-square, it is **less commonly employed** in medical research compared to the test of independence. - **Example use**: testing if observed genetic ratios match Mendelian expectations, or if disease distribution matches a theoretical model. *Estimating population mean* - Estimating a **population mean** requires methods for **continuous data** such as confidence intervals or t-tests. - The Chi-square test operates on **frequency counts of categorical data**, not on numerical measurements, making it inappropriate for mean estimation. *Comparing two population means* - **Comparing means** requires tests designed for continuous data such as the **t-test** (for two groups) or **ANOVA** (for multiple groups). - The Chi-square test analyzes **associations between categories**, not differences in central tendency of continuous variables.

Q: The figure below represents a statistical distribution. What type of skewness does it exhibit?

Right skewed. ***Right skewed*** - A distribution is **right-skewed** (positively skewed) when its tail extends longer to the right side, indicating that the majority of the data points are concentrated on the left side of the distribution. - In a right-skewed distribution, the **mean** is typically greater than the **median**, which is also greater than the **mode** (Mean > Median > Mode). *Normal distribution* - A **normal distribution** is **symmetrical**, meaning its left and right sides are mirror images, with no skewness. - In a normal distribution, the **mean, median, and mode** are approximately equal and located at the center of the distribution. *Left skewed* - A **left-skewed** (or negatively skewed) distribution has a tail that extends longer to the left side, with a concentration of data points on the right. - For a left-skewed distribution, the **mean** is usually less than the **median**, which is less than the **mode** (Mean < Median < Mode). *Bimodal distribution* - A **bimodal distribution** has two distinct peaks or modes, indicating two different groups or subpopulations in the data. - This is different from skewness, which describes the asymmetry of a single-peaked distribution.

Question 1

Which of the following is NOT a characteristic of a health index?

Accepted Answer

Affordability

Answer

Validity

Answer

Reliability

Answer

Feasibility

Question 2

Sensitivity of a screening test tells about

Accepted Answer

Percentage of individuals with the disease who test positive.

Answer

Percentage of healthy individuals among those with a negative test result.

Answer

Percentage of individuals with the disease among those with a positive test result.

Answer

Percentage of healthy individuals among those with a positive test result.

Question 3

In a positively skewed distribution, which of the following is true?

Accepted Answer

Mean > Median > Mode

Answer

Mean = Median = Mode

Answer

Mode > Median > Mean

Answer

None of the options

Question 4

A screening test has sensitivity of 90% and specificity of 99%. The prevalence of disease under investigation is 5 per 1000 population. What is the PPV of the given screening test?

Accepted Answer

33

Answer

10

Answer

70

Answer

99

Question 5

Which of the following statements is true about cohort studies?

Accepted Answer

A study that measures the incidence of a disease.

Answer

A study that follows participants over time to observe outcomes but does not measure incidence.

Answer

A study that describes characteristics of a population.

Answer

A study that can determine cause and effect.

Question 6

Which of the following is a parametric test?

Accepted Answer

Student t-test

Answer

Sign test

Answer

Fisher exact test

Answer

Chi-square test

Question 7

In a study a patient does not know the nature of drug [whether a placebo or curative drug] he is taking. The researcher knows the drug type to be given to the individuals in study. Type of blinding in this study is?

Accepted Answer

Single

Answer

Double

Answer

Triple

Answer

Combined double/triple

Question 8

What is the denominator used in calculating the under-5 proportional mortality rate?

Accepted Answer

Total deaths

Answer

Number of deaths under 5 years of age

Answer

Mid-year under-5 population

Answer

Mid-year population

Question 9

The Chi-square test is most commonly used in medical statistics to test for:

Accepted Answer

Test for independence of categorical variables

Answer

Test for goodness of fit

Answer

Estimating population mean

Answer

Comparing two population means

Question 10

The figure below represents a statistical distribution. What type of skewness does it exhibit?

Accepted Answer

Right skewed

Answer

Normal distribution

Answer

Left skewed

Answer

Bimodal distribution

Biostatistics — MCQs

Biostatistics — MCQs

On this page

Practice by Chapter

Want unlimited practice?