Epidemiology Concepts Quiz
28 Questions
0 Views

Choose a study mode

Play Quiz
Study Flashcards
Spaced Repetition
Chat to Lesson

Podcast

Play an AI-generated podcast conversation about this lesson

Questions and Answers

What factor is necessary for establishing dietary causation of diabetes mellitus?

  • Popularity of the diet
  • Cost-effectiveness of the diet
  • Theoretical plausibility (correct)
  • Availability of resources

The concept of specificity asserts that an exposure can lead to multiple outcomes.

False (B)

What is an example of a relationship that should not contradict established medical science?

Smoking and hypertension.

The ______ relationship indicates that increasing the dose of an exposure will affect the outcome.

<p>dose-response</p> Signup and view all the answers

Match the following criteria for establishing causation with their descriptions:

<p>Theoretical plausibility = Aligns with existing scientific knowledge Coherence = Consistency with established facts Specificity = Single exposure leads to a single effect Dose-response relationship = Increased exposure leads to increased effect</p> Signup and view all the answers

What defines internal validity in a study?

<p>The accuracy of the estimate of effect measure (A)</p> Signup and view all the answers

Causal inference refers to the process of identifying noncausal relationships between variables.

<p>False (B)</p> Signup and view all the answers

Name one source of error that can occur in epidemiologic studies.

<p>random errors or sampling errors</p> Signup and view all the answers

The goal of epidemiologic studies is to estimate the value of the parameter with little __________.

<p>error</p> Signup and view all the answers

Match the following types of validity with their descriptions:

<p>Internal Validity = Validity within the study External Validity = Validity beyond the study</p> Signup and view all the answers

What is the primary aim of randomization in the design stage of a study?

<p>To randomize the distribution of confounders (A)</p> Signup and view all the answers

Stratified analysis ensures that confounders are unevenly distributed within each stratum.

<p>False (B)</p> Signup and view all the answers

List one of Bradford Hill's criteria for causal inference.

<p>Strength of association, Temporality, Consistency (any one is acceptable)</p> Signup and view all the answers

A common method to restrict confounding factors in a study is called __________.

<p>Restriction</p> Signup and view all the answers

Match the following terms with their definitions:

<p>Randomization = Aim to evenly distribute confounders Restriction = Limiting study participation based on confounders Matching = Creating equal distribution of confounders Multivariate analysis = Applying multiple statistical tests to analyze data</p> Signup and view all the answers

What does the term 'temporality' refer to in causal inference?

<p>A temporal relationship where exposure precedes disease (B)</p> Signup and view all the answers

In causation studies, consistency of findings across different populations and designs supports a causal inference.

<p>True (A)</p> Signup and view all the answers

What does the presence of a higher risk ratio suggest regarding the association being studied?

<p>It is more likely to be causal.</p> Signup and view all the answers

What is one of the main results of confounding in research?

<p>It distorts the estimation of the magnitude of association between exposure and disease. (A)</p> Signup and view all the answers

A causal association means that changes in one event are not followed by changes in another event.

<p>False (B)</p> Signup and view all the answers

What is selection bias?

<p>Selection bias occurs when a non-representative sample is used in a study.</p> Signup and view all the answers

A ______ association occurs when the relationship of both factor and disease involves a third variable.

<p>non-causal</p> Signup and view all the answers

Match the type of bias with its description:

<p>Selection Bias = Non-representative sample Information Bias = Inaccurate information collected Misclassification = Incorrect categorization of subjects Differential Bias = Non-random errors in data collection</p> Signup and view all the answers

Which of the following best describes confounding?

<p>The mixing of effects between two variables caused by a third variable. (B)</p> Signup and view all the answers

Information bias can occur due to inaccurate information collected from subjects.

<p>True (A)</p> Signup and view all the answers

What are the two forms of misclassification?

<p>Differential and Non-differential misclassification.</p> Signup and view all the answers

The presence of a ______ associated with exposure and disease can lead to over or underestimation of associations.

<p>confounder</p> Signup and view all the answers

Which of the following is NOT a source of information bias?

<p>Random sampling (B)</p> Signup and view all the answers

Flashcards

Causal Inference

The process of determining if a relationship between two variables is causal, not just an association.

Association

A statistical relationship between two variables, where they tend to change together, either positively or negatively.

Internal Validity

The accuracy of an estimate of an effect in a study.

External Validity

The extent to which the results can be applied to a larger population.

Signup and view all the flashcards

Causal Association

A relationship where one variable directly or indirectly causes a change in another.

Signup and view all the flashcards

Causal relationship (diet and diabetes)

A diet is the cause of diabetes if the diet is actually the cause and is associated and causal.

Signup and view all the flashcards

Theoretical Plausibility

A plausible causal link must not contradict scientific knowledge of the disease's natural history and biology.

Signup and view all the flashcards

Coherence (causation)

The causal link must make sense and not contradict existing knowledge about the disease.

Signup and view all the flashcards

Specificity of causes

The cause should lead to only one effect.

Signup and view all the flashcards

Dose-response relationship

Increasing the dose of a potential cause should affect the outcome.

Signup and view all the flashcards

Non-Causal Association

An association where the relationship is not a direct cause-and-effect. It results from a common/third factor affecting both exposure and disease.

Signup and view all the flashcards

Bias

Systematic error that distorts the association between exposure and disease. Can be selection or information bias.

Signup and view all the flashcards

Selection Bias

Non-representative sample selection that leads to biased results when studying disease or exposure.

Signup and view all the flashcards

Information Bias

Inaccurate information collected about either exposure or disease in a study, distorting the association.

Signup and view all the flashcards

Confounding

A link between exposure and disease where a third factor influences both is present.

Signup and view all the flashcards

Confounder

A third variable associated with the exposure and also a risk factor for the disease.

Signup and view all the flashcards

Direct Causation

A direct cause-and-effect relationship between the exposure and the disease.

Signup and view all the flashcards

Indirect Causation

A causal relationship where the exposure affects the disease through an intermediate factor.

Signup and view all the flashcards

Misclassification

An error in classifying exposure or disease status in a study, leading to inaccurate results.

Signup and view all the flashcards

Randomization in Causal Inference

A method to randomly assign participants to study groups, aiming to equalize potential confounding factors between groups.

Signup and view all the flashcards

Restriction in Causal Inference

Limiting study participants to those without certain confounding factors.

Signup and view all the flashcards

Matching in Causal Inference

Equating study groups based on similar confounding characteristics to improve comparability.

Signup and view all the flashcards

Stratified Analysis

Analysis that divides the data into subgroups (strata) based on confounders and then examines the effect within each group independently

Signup and view all the flashcards

Multivariate Analysis

Statistical methods that consider the effects of multiple variables simultaneously when investigating associations.

Signup and view all the flashcards

Hill's Criteria for Causality (Strength)

A strong association between an exposure and outcome (higher risk ratio) provides stronger evidence for causality.

Signup and view all the flashcards

Hill's Criteria for Causality (Temporality)

Exposure must precede the disease, meaning the exposure happens before the disease develops.

Signup and view all the flashcards

Hill's Criteria for Causality (Consistency)

Consistent finding across different study designs, populations, and investigators increase confidence in causality.

Signup and view all the flashcards

Study Notes

Statistical Association

  • Characterizes the relationship between variables (positive or negative)
  • Uses statistical methods to describe the association

Causal Inference

  • Assigns causal relationships to observed associations
  • Aims to understand if one variable causes a change in another

Internal Validity

  • Accuracy of the study's effect estimate
  • Absence of systematic errors

External Validity

  • Generalizability to a larger population
  • Absence of random errors

Goal of Epidemiologic Studies

  • Accurate estimation of population parameters
  • Minimizing errors (random and systematic)

Types of Errors

  • Random errors: Sampling errors (differences between population and sample values)
  • Systematic errors (biases and confounding): Distort association estimates (over or under-estimate the association)

Association vs. Cause

  • Association: Identifiable relationship between exposure and outcome
  • Cause: Mechanisms connecting exposure to outcome

Causal vs. Non-causal

  • Causal: Alteration in one event leads to a change in another
  • Non-causal: Association resulting from a relationship with a third variable

Bias

  • Selection Bias: Non-representative sample (e.g., some people more likely to be chosen/be subjects)
  • Information Bias: Inaccurate information collected (e.g., misclassification)
    • Differential (non-random): Non-equal misclassification between groups
    • Non-differential (random): Similar misclassification in all groups

Confounding

  • Mixture of exposure effect with that of a third factor (confounder)
  • Confounder is associated with the exposure and a risk factor for the outcome.

Controlling Confounding

  • Design stage: Randomization, Restriction, Matching
  • Analysis stage: Stratified analysis, Multivariate analysis

Bradford Hill's Criteria for Causal Inference

  • Strength of Association: Higher risk ratio, higher likelihood of causation
  • Temporality: Exposure precedes the outcome
  • Consistency: Consistent findings across different studies, designs, and populations
  • Coherence: Consistency with existing knowledge about the natural history
  • Specificity: Exposure leads to a single outcome
  • Dose-Response Relation: Increase in exposure strength corresponds to increase in outcome
  • Experimental Evidence: Experimental data supporting the relationship (if applicable)
  • Analogy: Similar relationships observed in other situations

Studying That Suits You

Use AI to generate personalized quizzes and flashcards to suit your learning preferences.

Quiz Team

Related Documents

Week 15 Causal Inference BIOE211 PDF

Description

Test your knowledge on key concepts in epidemiology, including statistical association, causal inference, and validity. This quiz will assess your understanding of how relationships between variables are characterized and the importance of minimizing errors in studies.

More Like This

Types of Epidemiological Studies Quiz
23 questions

Epidemiology Quiz: Types of Epidemiological Studies

FriendlyUnakite avatar
FriendlyUnakite
Causal Inference in Epidemiology
11 questions

Causal Inference in Epidemiology

ValuableHeliotrope5203 avatar
ValuableHeliotrope5203
Associació i Causalitat en Epidemiologia
24 questions

Associació i Causalitat en Epidemiologia

LovingAntigorite6493 avatar
LovingAntigorite6493
Use Quizgecko on...
Browser
Open
Browser
Browser