Abstract
Background
Measures of attributable risk are critical in epidemiology, particularly for planning and evaluating public health interventions. Current definitions, however, often overlook temporal relationships between exposure and risk. This study proposes extended definitions of attributable risk within the framework of distributed lag non-linear models (DLNMs), which model delayed and non-linear exposure-response associations.
Methods
The study classifies attributable risk measures using forward and backward perspectives. Forward measures estimate future burdens from current exposures, while backward measures summarize current burdens from past exposures. These methods are demonstrated using time-series data on temperature-related mortality in London (1993–2006) and Rome (1992–2010).
Results
The analysis estimates the overall mortality burden attributable to temperature and further separates components due to cold, heat, mild, and extreme temperatures. The backward method indicates that cold temperatures contribute significantly more to mortality than heat in both cities.
Conclusions
Extended definitions of attributable risk, accounting for temporal dimensions, provide more appropriate measures in contexts with complex exposure-response associations. These methods have broad applications beyond temperature-related mortality, enabling more precise public health planning.
Key Questions
1. What are distributed lag non-linear models (DLNMs)?
DLNMs are statistical models that account for delayed and non-linear associations between exposure and risk, offering a nuanced understanding of cumulative effects over time.
2. How do forward and backward perspectives differ in attributable risk analysis?
Forward measures estimate future risks due to current exposures, while backward measures evaluate current risks from past exposures, providing complementary insights.
3. What role does temperature play in mortality risk?
The study highlights that cold temperatures account for a larger share of temperature-attributable mortality than heat, emphasizing the need for targeted interventions in colder climates.
4. What is the significance of separating mild and extreme temperature effects?
By distinguishing between mild and extreme temperatures, the analysis reveals that mild cold contributes more to mortality than extreme cold, challenging common assumptions about temperature-related health risks.
