How To Calculation Population Attributable Risk Example

Calculate the proportion of disease cases in a population that can be attributed to a specific risk factor

Comprehensive Guide to Calculating Population Attributable Risk (PAR)

Population Attributable Risk (PAR) is a fundamental concept in epidemiology that quantifies the proportion of disease cases in a population that can be attributed to a specific risk factor. This metric helps public health professionals prioritize interventions by identifying which risk factors contribute most significantly to disease burden.

Understanding the Key Components

1. Relative Risk (RR): The ratio of disease probability in exposed vs. unexposed groups
2. Odds Ratio (OR): Approximates RR for rare diseases (≤5% prevalence)
3. Exposure Prevalence (P_e): Proportion of population exposed to the risk factor
4. Attributable Fraction (AF): Proportion of disease in exposed individuals due to the exposure

The Population Attributable Risk Formula

The standard formula for calculating PAR is:

PAR Formula

PAR = P_e × (RR – 1) / [1 + P_e × (RR – 1)]

Where:

• P_e = Exposure prevalence in the population
• RR = Relative Risk (or OR for case-control studies)

For case-control studies where you can’t directly calculate RR, use this alternative formula:

Case-Control PAR Formula

PAR = (P_e × (OR – 1)) / (P_e × (OR – 1) + 1)

Step-by-Step Calculation Process

1. Determine Disease Prevalence:
   • Exposed group (I_e): 15% (0.15)
   • Unexposed group (I_u): 5% (0.05)
2. Calculate Relative Risk:

   RR = I_e / I_u = 0.15 / 0.05 = 3.0

3. Determine Exposure Prevalence:

   P_e = 30% (0.30)

4. Apply PAR Formula:

   PAR = 0.30 × (3.0 – 1) / [1 + 0.30 × (3.0 – 1)] = 0.2308 or 23.08%

Interpreting PAR Results

PAR Range	Interpretation	Public Health Action
<10%	Low attributable risk	Monitor but no immediate intervention needed
10-25%	Moderate attributable risk	Targeted interventions for high-risk groups
25-50%	High attributable risk	Population-wide prevention programs
>50%	Very high attributable risk	Urgent public health priority with comprehensive strategies

Real-World Examples of PAR Applications

Risk Factor	Disease	Reported PAR	Source
Smoking	Lung Cancer	80-90%	CDC (2023)
Physical Inactivity	Coronary Heart Disease	12-15%	WHO (2022)
Alcohol Consumption	Liver Cirrhosis	45-50%	NIAAA (2023)
Obesity	Type 2 Diabetes	30-35%	CDC Diabetes (2023)

Common Mistakes in PAR Calculation

• Confusing RR with OR: While OR approximates RR for rare diseases, using them interchangeably for common diseases (>10% prevalence) leads to overestimation
• Ignoring confounding factors: Failing to adjust for confounders like age, sex, or socioeconomic status can bias results
• Misinterpreting exposure prevalence: Using exposure prevalence from the study sample rather than the target population
• Assuming causality: PAR quantifies association, not causation – the exposure must be causally related to the disease
• Overlooking effect measure modification: Not considering that the RR might vary across population subgroups

Advanced Considerations

For more sophisticated analyses, epidemiologists often use:

1. Adjusted PAR:

   Accounts for multiple risk factors simultaneously using regression models

2. Sequential PAR:

   Calculates the PAR after removing the effect of previous risk factors in a specified order

3. Potential Impact Fraction (PIF):

   Extends PAR to estimate the effect of changing exposure prevalence to a counterfactual level

4. Bayesian Approaches:

   Incorporates prior information to improve PAR estimates when data is limited

Limitations of Population Attributable Risk

• Dependence on exposure prevalence: PAR varies with exposure prevalence even if the RR remains constant
• Not applicable to individual risk: PAR is a population-level measure and cannot predict individual risk
• Sensitive to measurement error: Small errors in RR or exposure prevalence can lead to substantial bias
• Limited to modifiable risk factors: Only meaningful for exposures that can be modified through intervention
• Assumes causal relationship: Requires established causality between exposure and disease

Practical Applications in Public Health

PAR calculations directly inform public health policy and resource allocation:

• Tobacco Control:

  The high PAR for smoking-related diseases (80-90% for lung cancer) justified comprehensive tobacco control policies including taxation, advertising bans, and smoking cessation programs

• Vaccination Programs:

  PAR estimates for vaccine-preventable diseases help determine optimal vaccination coverage levels to achieve herd immunity

• Obesity Prevention:

  With obesity contributing to 30-35% of type 2 diabetes cases, PAR data supports policies like sugar taxes and nutrition labeling requirements

• Workplace Safety:

  Occupational PAR calculations identify high-risk industries and justify regulatory interventions to reduce work-related injuries and illnesses

• Environmental Regulations:

  PAR estimates for air pollution-related diseases inform clean air standards and emission control policies

Emerging Trends in PAR Research

Recent advancements are expanding the application of PAR:

• Genomic Epidemiology:

  Integrating genetic risk scores with traditional risk factors to calculate “polygenic PAR”

• Machine Learning:

  Using algorithmic approaches to handle complex interactions between multiple risk factors

• Life Course Epidemiology:

  Calculating cumulative PAR across the lifespan to account for exposure timing

• Spatial Epidemiology:

  Geographic information systems (GIS) to calculate localized PAR for targeted interventions

• Implementation Science:

  Combining PAR with cost-effectiveness analysis to optimize intervention strategies

Learning Resources

For those seeking to deepen their understanding of population attributable risk:

• Books:
  • “Modern Epidemiology” by Kenneth J. Rothman (Chapter 8 on Attributable Fractions)
  • “Epidemiology: Beyond the Basics” by Moyses Szklo and F. Javier Nieto (Chapter 19 on Measures of Potential Impact)
• Online Courses:
  • Coursera: “Epidemiology: The Basic Science of Public Health” (University of North Carolina)
  • edX: “Epidemiology for Public Health” (University of Michigan)
• Software Tools:
  • R packages: epiR, attribRisk
  • Stata commands: punaf, punafcc
  • SAS macros: %PAR, %PAF

Key Takeaway

Population Attributable Risk bridges epidemiological research and public health practice by quantifying the potential impact of risk factor modification at the population level. When properly calculated and interpreted, PAR becomes a powerful tool for evidence-based decision making in disease prevention and health promotion.