How To Calculate Attack Rate Ratio

Calculate the attack rate ratio between exposed and unexposed groups to measure relative risk in epidemiological studies.

Comprehensive Guide: How to Calculate Attack Rate Ratio

The attack rate ratio (ARR), also known as the relative risk (RR), is a fundamental measure in epidemiology that compares the risk of developing a disease between an exposed group and an unexposed group. This metric is crucial for understanding the strength of association between an exposure and an outcome, helping public health professionals assess potential causal relationships.

Understanding the Basics

The attack rate ratio is calculated by dividing the attack rate in the exposed group by the attack rate in the unexposed group. The attack rate itself is the proportion of individuals who develop the disease among those at risk during a specific time period.

1. Attack Rate (Exposed): (Number of cases in exposed group) / (Total population in exposed group)
2. Attack Rate (Unexposed): (Number of cases in unexposed group) / (Total population in unexposed group)
3. Attack Rate Ratio: (Attack Rate in Exposed) / (Attack Rate in Unexposed)

When to Use Attack Rate Ratio

The attack rate ratio is particularly useful in:

• Outbreak investigations to identify potential sources of infection
• Cohort studies where researchers follow groups over time
• Evaluating the effectiveness of interventions or exposures
• Comparing disease risk between different population groups

Interpreting the Results

The interpretation of the attack rate ratio depends on its value:

• ARR = 1: No difference in risk between exposed and unexposed groups
• ARR > 1: Increased risk in the exposed group
• ARR < 1: Decreased risk in the exposed group (protective effect)

For example, an ARR of 2.5 indicates that the exposed group has 2.5 times the risk of developing the disease compared to the unexposed group. Conversely, an ARR of 0.4 suggests that the exposed group has 60% lower risk than the unexposed group.

Practical Example: Foodborne Outbreak

Consider a foodborne illness outbreak where investigators suspect contaminated chicken as the source. They collect data from 200 people who attended a picnic:

Ate Chicken	Did Not Eat Chicken
Ill: 45 Well: 55 Total: 100	Ill: 10 Well: 90 Total: 100

Calculations:

• Attack rate (exposed) = 45/100 = 0.45 or 45%
• Attack rate (unexposed) = 10/100 = 0.10 or 10%
• Attack rate ratio = 0.45/0.10 = 4.5

Interpretation: People who ate chicken were 4.5 times more likely to become ill than those who didn’t eat chicken, strongly suggesting chicken as the outbreak source.

Common Mistakes to Avoid

When calculating and interpreting attack rate ratios, be mindful of these potential pitfalls:

1. Small sample sizes: Can lead to unstable estimates and wide confidence intervals
2. Misclassification: Errors in determining who was truly exposed or developed the disease
3. Confounding factors: Other variables that might explain the observed association
4. Zero cells: When one group has zero cases, making the ratio undefined (requires special statistical methods)
5. Overinterpreting significance: Not all statistically significant ratios indicate causation

Advanced Considerations

For more sophisticated analyses, epidemiologists often consider:

• Stratified analysis: Calculating ARRs within subgroups to identify effect measure modification
• Confidence intervals: Providing a range of values within which the true ARR likely falls
• Attributable risk: Calculating the excess risk due to the exposure
• Population attributable risk: Estimating the proportion of cases in the population attributable to the exposure

Comparison with Other Epidemiological Measures

Measure	Calculation	When to Use	Interpretation
Attack Rate Ratio (Relative Risk)	ARexposed / ARunexposed	Cohort studies, outbreak investigations	Compares risk between exposed and unexposed
Odds Ratio	(a/c) / (b/d)	Case-control studies, when disease is rare	Approximates RR when disease is rare
Attributable Risk	ARexposed – ARunexposed	Quantifying excess risk due to exposure	Absolute difference in risk
Risk Difference	Same as Attributable Risk	Public health planning	Number of cases that could be prevented

Note: AR = Attack Rate, a = exposed cases, b = exposed non-cases, c = unexposed cases, d = unexposed non-cases

Real-World Applications

Attack rate ratios have been instrumental in:

• Identifying the source of Legionnaires’ disease outbreaks (contaminated water systems)
• Linking Reye’s syndrome to aspirin use in children with viral infections
• Associating Zika virus with microcephaly in newborns
• Evaluating the effectiveness of vaccines during clinical trials
• Investigating foodborne illness outbreaks from specific restaurants or food products

Calculating Confidence Intervals

To assess the precision of your attack rate ratio estimate, you should calculate confidence intervals. The 95% confidence interval for an attack rate ratio can be calculated using the following formula:

Lower bound = exp[ln(ARR) – 1.96 × √(1/a + 1/c – 1/(a+b) – 1/(c+d))]

Upper bound = exp[ln(ARR) + 1.96 × √(1/a + 1/c – 1/(a+b) – 1/(c+d))]

Where:

• a = number of exposed cases
• b = number of exposed non-cases
• c = number of unexposed cases
• d = number of unexposed non-cases

If the confidence interval includes 1, the result is not statistically significant at the 0.05 level.

Centers for Disease Control and Prevention (CDC) Resources:

For official guidelines on calculating and interpreting attack rate ratios, refer to the CDC’s Principles of Epidemiology course materials.

Source: CDC.gov

World Health Organization (WHO) Outbreak Investigation Guidelines:

The WHO provides comprehensive guidance on using attack rate ratios in outbreak investigations in their Field Epidemiology Manual.

Source: WHO.int

Johns Hopkins Bloomberg School of Public Health:

For academic perspectives on relative risk measures, explore the OpenCourseWare materials from Johns Hopkins, particularly their epidemiological methods courses.

Source: JHSPH.edu

Limitations of Attack Rate Ratio

While the attack rate ratio is a powerful tool, it has several limitations:

• Temporal ambiguity: Doesn’t establish the temporal sequence between exposure and outcome
• Confounding: May be influenced by other variables associated with both exposure and outcome
• Rare outcomes: When outcomes are rare, the odds ratio may be a better measure
• Selection bias: Results may be affected by how study participants were selected
• Information bias: Errors in measuring exposure or outcome can distort results

Best Practices for Reporting

When presenting attack rate ratio findings:

1. Always report the actual numbers (cases and population sizes) used in calculations
2. Include confidence intervals to indicate precision
3. Clearly state the time period and population under study
4. Discuss potential biases and limitations
5. Put findings in context with existing literature
6. Avoid causal language unless the study design supports it

Software Tools for Calculation

While our calculator provides quick results, several professional tools can help with more complex analyses:

• Epi Info: Free CDC software for epidemiological analyses
• R: Open-source statistical software with epidemiological packages
• Stata/SAS: Commercial statistical packages with advanced features
• OpenEpi: Free web-based epidemiological calculators

Case Study: 2011 E. coli Outbreak in Germany

One of the most dramatic demonstrations of attack rate ratio calculations occurred during the 2011 E. coli O104:H4 outbreak in Germany. Investigators used case-control studies to identify the source:

Exposure	Cases (n=26)	Controls (n=77)	ARR
Ate sprouts	25 (96.2%)	29 (37.7%)	25.5
Did not eat sprouts	1 (3.8%)	48 (62.3%)	Reference

The extraordinarily high attack rate ratio of 25.5 provided compelling evidence that sprouts were the outbreak source, leading to rapid public health interventions that helped contain the outbreak.

Future Directions in Epidemiological Measures

As epidemiology evolves, new approaches are being developed to complement traditional measures like the attack rate ratio:

• Machine learning: Identifying complex patterns in exposure-outcome relationships
• Mendelian randomization: Using genetic variants as instrumental variables
• Causal inference methods: More robust approaches to establishing causality
• Real-time surveillance: Using digital data for faster outbreak detection
• Exposome research: Studying the totality of environmental exposures

While these advanced methods are becoming more common, the attack rate ratio remains a fundamental tool in the epidemiologist’s toolkit due to its simplicity and interpretability.