EG

Measures of Association in Epidemiology

Measures of Frequency

  • Types:

    • Cumulative Incidence

      • Synonyms: Risk, Attack Rate

      • Formula: New cases during specified time period / Persons in the specified population at risk.

      • Interpretation: Risk of new disease over the study period, expressed as a proportion or percent.

    • Incidence Rate

      • Also known as Incidence Density or Person-Time Rate.

      • Formula: New cases during specified time period / Person-time at risk during study period.

      • Interpretation: Rate or speed of new cases development; assumes constant risk across time.

    • Prevalence

      • Formula: Number of current cases (new and preexisting) at a specified point in time / Persons in the population at specified time.

      • Interpretation: Burden of disease as a proportion or percentage; does not consider at-risk status.

Measures of Association

Absolute/Relative Measures of Association

  • Types:

    • Attributable Risk / Rate Difference (RD) among exposed.

    • Attributable Proportion (Percent) among exposed.

    • Population Attributable Risk / Rate Difference.

    • Risk or Rate Ratio (Relative Risk; RR).

    • Odds Ratio (OR).

    • Number Needed to Treat (NNT).

  • Formulas and Interpretations:

    • Risk Difference: Rexp - Ruexp (exposed vs unexposed).

    • Attributable Proportion (%): [(Rexp - Ruexp) / Rexp] * 100.

    • Population Attributable Risk: Rtotal - Ruexp (total population).

    • Risk Ratio (RR): Rexp / Runexp.

    • Odds Ratio: (a * d) / (b * c) in case-control studies.

    • Efficacy and NNT calculated accordingly.

Relative Risk vs Risk/Rate Differences

  • Relative Risk (RR): Measures strength of association.

  • Attributable Risk: Measures public health impact (preventable cases).

    • Risk Difference (additive scale).

    • Attributable Proportion and Population Risk Differences.

Measure of Association by Study Design

  • Cohort Studies: Risk/Rate Ratio (Relative Risk); Attributable Risk calculations.

  • Case Control Studies: Odds Ratio as an estimate of relative risk.

  • Cross-Sectional Studies: Prevalence Ratio.

  • Ecologic Studies: Rate Ratios based on population-level data.

Understanding Relative Risk

  • **Definitions: **

    • RR < 1 indicates a protective exposure; used for risk assessment.

    • How to interpret: When RR is calculated, it serves as a multiplier for risk comparison.

Attributable Risk Analysis

Among the Exposed

  • Example: Incidence rates for Chronic Heart Disease (CHD) among smokers and non-smokers.

  • Calculation of AR shows how many cases could be prevented via interventions (e.g., smoking cessation).

Total Population Attributable Risk

  • Captures excess incidence based on prevalence of exposure across the population.

  • Helps visualize public health impact.

Odds Ratio (OR) in Studies

  • Definition: Odds ratio compares likelihood of exposure between cases and controls.

  • OR Interpretation:

    • OR > 1 indicates a positive association; OR < 1 indicates a protective factor.

  • Relies on accurate representation of population for robust conclusions.

Odds Ratio Examples

  • Case-Control Studies: Calculating and interpreting odds ratios importance in determining risk exposure relationships.

  • Odds ratios provide insights into the strength and direction of association, particularly when the disease is rare.

Conclusion

  • Conclusion drawn from various risk measures helps evaluate causal relationships and public health implications regarding disease exposure.