Lung Cancer Risk Models for Screening (R package: lcrisks)

In both the absence and presence of screening, the R package calculates individual risks of lung cancer and lung cancer death based on ovariates: age, education, sex, race, smoking intensity/duration/quit-years, Body Mass Index, family history of lung-cancer, and self-reported emphysema. In the presence of CT screening akin to the NLST (3 yearly screens, 5 years of follow-up), it uses the covariates to estimate risk of false-positive CT screen as well as the reduction in risk of lung cancer death and increase in risk of lung cancer screening. This package also estimates the Life Years Gained From Screening-CT (LYFS-CT) as per Cheung et al., 2019. It requires the same variables as LCDRAT plus 12 additional comorbidities and the year of patient assessment.

• R Documentation
• License Agreement

Software Download (Version 4.1.1):

• Windows O/S - lcrisks.zip
• Linux O/S - lcrisks.tar.gz
• Excel Tool - lcrisk_tool.xlsx

 

References:

• Katki HA, Kovalchik SA, Berg CD, Cheung LC, Chaturvedi AK. Development and validation of risk models to select ever-smokers for CT lung cancer screening. JAMA 2016; 315(21):2300-2311. doi: 10.1001/jama.2016.6255. [PubMed]
• Cheung LC, Berg CD, Castle PE, Katki HA, Chaturvedi AK. Life-gained versus risk-based selection of smokers for lung cancer screening. Ann Intern Med 2019; doi:107326/M19.1263

Support:

Email: Li Cheung and Hormuzd Katki