Predicting lung cancer deaths from smoking prevalence data

Area of Science:

• Epidemiology
• Public Health
• Cancer Research

Background:

• Accurate lung cancer burden data is scarce in developing countries lacking cancer registries.
• Previous models estimated Ethiopia's lung cancer deaths significantly higher than WHO and GLOBOCAN figures.
• Existing models require refinement to improve the accuracy of lung cancer mortality estimations.

Purpose of the Study:

• To enhance and validate a model for estimating lung cancer deaths in countries with limited cancer registration.
• To incorporate data on ex-smoker mortality risks and smoking duration into the estimation model.
• To re-evaluate non-smoker mortality rates in lung cancer estimations.

Main Methods:

• Developed two improved models by including ex-smoker risk data and smoking duration.
• Re-analyzed non-smoker mortality data within the refined models.
• Validated the models by estimating lung cancer deaths in Germany, UK, Canada, and Utah, USA, comparing results to observed data.

Main Results:

• The refined model, incorporating non-smoker mortality and detailed smoking dose, accurately estimated observed lung cancer deaths in developed countries.
• The updated estimate for yearly lung cancer deaths in Ethiopia using the refined model is 2946.
• The study suggests that World Health Organization (WHO) lung cancer death estimates may be underestimated in Ethiopia.

Conclusions:

• The refined model provides a more accurate estimation of lung cancer deaths, particularly in regions with limited data.
• Ethiopia's lung cancer mortality may be higher than previously reported by international health organizations.
• Further research is needed to confirm these findings and improve global cancer burden assessments.