PBH250 Final study guide

r = 1

Perfect positive correlation

r = -1

Perfect negative correlation

r = 0

No coorelation

Risk Ratio is measured in

Cohort

Odds Ratio is measured in

Case-Control

Error from how participants are chosen (addressed during design)

Selection Bias

Error from mismeasuring exposure/outcome (addressed during design and data collection)

Information Bias

Third variable distorts exposure-outcome link

Confounding Bias

OR or RR = 1

No association between exposure and outcome

OR or RR > 1

Exposure is associated with higher odds/risk of outcome

OR or RR < 1

Exposure is associated with lower odds/risk of outcome

Researchers just watch; no treatment given (e.g., cohort, case-control)

Observational RCT

Researchers assign treatments or exposures

Interventional RCT

Keeping participants, doctors, or researchers unaware of which group a participant is in

Blinding in RCTs

Analyze participants based on the group they were originally randomized to, regardless of whether they followed the treatment

Intent-to-Treat (ITT) Analysis

Analyzes based on what treatment they actually received

Treatment-As-Received

Ability of the instrument to measure consistently

Reliability

Extent to which a method measures what it is supposed to measure (accuracy)

Validity

Can the study measure what it sets out to measure? free of bias?

Internal validity

Is the study applicable to the rest of the world

External validity

% of people with the disease who are correctly identified

Sensitivity

% of people without the disease who are correctly identified

Specificity

% of patients who test positive actually have the disease

PPV

% of patients who tested negative actually have the disease

NPV

Non-medical factors like income, education, housing, and discrimination

Social Determinants of Health (SDOH)

• Temporality (must come first)

• Strength (stronger = more likely causal)

• Consistency (same in many studies)

• Dose-response

• Plausibility

• Coherence

• Experiment

• Analogy

• Specificity (less emphasized today)

Tenants of causality