PBH 452: Chapter 15 - Causal Inference: Bias, Confounding, and Interaction

0.0(0)
Studied by 0 people
call kaiCall Kai
learnLearn
examPractice Test
spaced repetitionSpaced Repetition
heart puzzleMatch
flashcardsFlashcards
GameKnowt Play
Card Sorting

1/13

flashcard set

Earn XP

Description and Tags

Vocabulary practice flashcards covering systematic and random errors, bias types, confounding criteria, and measures of association from Chapter 15 of Epidemiology.

14 Terms

1
New cards

Systematic Error

A predictable type of error caused by a flaw in the measurement system that affects all measurements in the same manner, which can be minimized through sound study design but not by increasing sample size.

2
New cards

Random Error

An unpredictable type of error caused by chance, imprecise measurement, or an excessively small comparison group, which can be minimized by increasing sample size but cannot be assessed for bias.

3
New cards

Repeatability

The practice of taking repeated measurements with the same instrument to reduce random error.

4
New cards

Bias

Any method in data collection, analysis, or interpretation that systematically distorts findings away from the truth, producing misleading estimates of association and inaccurate health outcome conclusions.

5
New cards

Information Bias

A form of systematic error arising from inaccurate measurement or classification of an exposure, outcome, or other study variable.

6
New cards

Selection Bias

A form of systematic error that occurs when the study sample selected is not representative of the target population.

7
New cards

Recall Bias

A type of systematic error occurring when participants cannot accurately remember past exposures, events, or experiences.

8
New cards

Researcher Bias

A systematic error introduced when the personal opinions, prejudices, or preconceived beliefs of the investigator influence study execution or findings.

9
New cards

Confounding Bias

The distortion of an apparent association between an exposure and an outcome caused by an extraneous underlying variable that is related to both.

10
New cards

Confounder Determination Criterion

An epidemiologic rule indicating confounding when the adjustment for an extraneous variable produces a change in the estimate of the odds ratio or relative risk of at least 10%10\%, calculated as ∣RRcrude−RRadjusted∣RRadjusted×100%\frac{|\text{RR}_{\text{crude}} - \text{RR}_{\text{adjusted}}|}{\text{RR}_{\text{adjusted}}} \times 100\%.

11
New cards
<p>Positive Confounding</p>

Positive Confounding

A situation demonstrated when the crude association (e.g., OR=2.60\text{OR} = 2.60) is significantly exaggerated away from the null compared to the stratum-specific estimates (e.g., OR=1.05\text{OR} = 1.05 and OR=0.98\text{OR} = 0.98) and the adjusted Mantel-Haenszel estimate (e.g., OR=1.02\text{OR} = 1.02).

12
New cards

Retrospective Cohort Study

An analytical study design in which past exposure and non-exposure to a factor are identified first to assess their relationship with a currently observed outcome or disease.

13
New cards

Protective Effect (Odds Ratio)

An association indicated by an odds ratio of less than 1.001.00 (such as OR=0.87\text{OR} = 0.87 or OR=0.72\text{OR} = 0.72), showing lower odds of the outcome among those exposed.

14
New cards

No Association (Odds Ratio)

An association outcome indicated by an odds ratio equal to 1.001.00, demonstrating that the odds of disease do not differ between exposed and unexposed groups.