HSTPH SEEM 1.3

What is the primary purpose of Phase I clinical trials?

To assess safety, tolerability, and pharmacokinetics in a small group of healthy volunteers.

What is evaluated in Phase II clinical trials?

Efficacy, optimal dosing, and short-term side effects in patients.

What is the focus of Phase III clinical trials?

Confirming efficacy, monitoring side effects, and comparing with standard treatment in a large population.

What is the goal of Phase IV clinical trials?

Post-marketing surveillance to detect rare or long-term side effects.

What does the null hypothesis (H₀) state?

That there is no difference or effect between the groups being studied.

What does the alternative hypothesis (H₁) state?

That there is a true difference or effect between the groups.

What is a Type I error?

Incorrectly rejecting the null hypothesis when it is true (false positive).

What is a Type II error?

Failing to reject the null hypothesis when the alternative is true (false negative).

What does a p-value represent?

The probability of obtaining results as extreme as observed, assuming the null hypothesis is true.

What p-value is commonly considered statistically significant?

Less than 0.05.

What does a 95% confidence interval imply?

That in 95% of samples, the true population parameter lies within the interval.

What does it mean if a confidence interval includes 0?

The result is not statistically significant.

What is a parametric test?

A statistical test that assumes the data follows a normal distribution.

What is a non-parametric test?

A test that does not assume a specific distribution; used for skewed or ordinal data.

What is a paired test used for?

Comparing two related groups or repeated measures on the same subjects.

What is the purpose of a two-tailed test?

To test for an effect in either direction (increase or decrease).

What test is used to compare means between two independent groups with normal distribution?

Independent t-test.

What test is appropriate for paired samples with normal distribution?

Paired t-test.

What non-parametric test compares two independent groups?

Mann-Whitney U test.

What non-parametric test compares paired samples?

Wilcoxon signed-rank test.

What test compares means across more than two groups?

ANOVA (Analysis of Variance).

What non-parametric test compares more than two independent groups?

Kruskal-Wallis test.

What is correlation?

A measure of the linear relationship between two variables.

What is the range of correlation coefficients (r)?

From -1 to +1.

What does r = 0 mean?

No linear correlation between the variables.

What does r = 1 mean?

Perfect positive linear correlation.

What does r = -1 mean?

Perfect negative linear correlation.

What is regression analysis used for?

Predicting an outcome based on one or more predictor variables.

What does the slope (β) in linear regression represent?

The change in the outcome for each unit change in the predictor.

What is logistic regression used for?

Modeling binary outcomes (e.g., disease yes/no).

What distinguishes causation from correlation?

Causation implies a direct relationship; correlation does not.

Can correlation imply causation?

No, correlation does not imply causation.

What is confounding?

A third variable that affects both the independent and dependent variables, distorting their association.

What is selection bias?

Systematic differences in characteristics between those selected and not selected for the study.

What is information bias?

Error due to inaccurate measurement or classification of study variables.

What is recall bias?

Differential accuracy of memory between groups.

In what type of study is recall bias most common?

Case-control studies.

What is observer bias?

Systematic differences in measurement due to the observer’s expectations or knowledge.

How can observer bias be minimized?

Through blinding of outcome assessors.

What is attrition bias?

Systematic differences due to participants dropping out of the study.

What is the effect of increasing sample size on confidence intervals?

It narrows the confidence interval, increasing precision.

How does increasing sample size affect power?

It increases the statistical power of the study.

What is statistical power?

The probability of correctly rejecting a false null hypothesis.

What increases statistical power?

Larger sample size, higher effect size, and less variability.

What is the significance of a wide confidence interval?

It indicates low precision of the estimate.

What type of variable is required for Pearson correlation?

Two continuous, normally distributed variables.

When should Spearman’s rank correlation be used?

When data are ordinal or not normally distributed.

What test is used to assess association between two categorical variables?

Chi-square test.

What is the primary purpose of randomization?

To minimize confounding and selection bias.

What is blinding in clinical trials?

Keeping study participants or researchers unaware of group allocation to reduce bias.

What is double blinding?

Both participants and investigators are unaware of the treatment assignments.

What is lead time bias?

Overestimation of survival due to earlier detection without real improvement in outcome.

What is performance bias?

Systematic differences in care between groups aside from the intervention being studied.

What is allocation concealment?

Hiding the sequence of treatment assignments to prevent selection bias.

How does a one-tailed test differ from a two-tailed test?

It tests for an effect in only one direction.

What is effect size?

The magnitude of difference or association being measured.

What is an R² value in regression?

The proportion of variance in the dependent variable explained by the model.

What does an R² of 0.75 mean?

75% of the variability in the outcome is explained by the predictors.

What is multicollinearity in regression?

High correlation among independent variables in a regression model.

What does a p-value of 0.50 indicate?

No statistically significant difference.

What is the implication of a CI that includes the null value?

The result is not statistically significant.

What is internal validity?

The degree to which the results of a study are free from bias.

What is external validity?

The degree to which study results can be generalized to other populations.

What does stratification mean in study design?

Separating participants into subgroups before analysis to control for confounding.

What is the difference between prospective and retrospective studies?

Prospective follow subjects forward in time; retrospective look back at existing data.

What is a case-control study?

A study comparing individuals with a condition (cases) to those without (controls) to find prior exposure.

What is a cohort study?

A study following exposed and unexposed groups over time to assess outcomes.

What does an odds ratio (OR) estimate?

The odds of exposure in cases versus controls.

What does a relative risk (RR) estimate?

The risk of outcome in the exposed group relative to the unexposed group.

What is intention-to-treat analysis?

Including all randomized participants in the groups to which they were assigned, regardless of protocol adherence.

What is per-protocol analysis?

Includes only those participants who completed the study as per the protocol.

What is a surrogate endpoint?

A substitute marker intended to predict clinical outcomes.

What is a composite endpoint?

A single outcome that combines multiple individual outcomes.

What is data censoring?

When participants do not experience the event before study ends or are lost to follow-up.

What is the purpose of interim analysis in trials?

To evaluate data before study completion for safety or efficacy.

What is a pilot study?

A small-scale preliminary study to assess feasibility before a full-scale study.

What is publication bias?

Tendency to publish positive findings more than negative or null results.

What is heterogeneity in meta-analysis?

Variability in study outcomes due to differences in study design, populations, or interventions.

What is a forest plot?

A graphical representation of results from multiple studies in a meta-analysis.

What does a funnel plot assess?

Potential publication bias in meta-analysis.

What is a systematic review?

A comprehensive summary of all relevant studies on a specific topic using a structured method.

What is a meta-analysis?

A statistical technique that combines results from multiple studies to produce a pooled estimate.

What is standard deviation (SD)?

A measure of the spread of data around the mean.

What is standard error (SE)?

An estimate of the variability of the sample mean.

What is a normal distribution?

A symmetric bell-shaped distribution with most values near the mean.

What is skewness?

A measure of the asymmetry of a distribution.

What does a positive skew indicate?

A longer right tail; mean > median.

What does a negative skew indicate?

A longer left tail; mean < median.

What is kurtosis?

The "tailedness" or peakedness of a distribution.

What is the central limit theorem?

With a large enough sample size, the sampling distribution of the mean will be approximately normal.

What is data imputation?

Filling in missing values using statistical methods.

What is a null value for difference in means?

Zero.

What is a null value for odds ratio or relative risk?

One.

What does it mean when an OR = 1?

No association between exposure and outcome.

What does it mean when RR < 1?

Exposure is associated with reduced risk of outcome.

What does it mean when RR > 1?

Exposure is associated with increased risk of outcome.

What is a clinical endpoint?

A direct measure of how a patient feels, functions, or survives.

What is a surrogate endpoint?

An indirect measure that substitutes for a clinical endpoint.

Why are surrogate endpoints used in trials?

To allow earlier assessment of treatment effects.

What is the main limitation of using surrogate endpoints?

They may not accurately predict true clinical outcomes.