lcmodels: Risk Predictions from Lung Cancer Models
In marskar/lcmodels: Risk Predictions from Lung Cancer Models

Description Usage Arguments Value Warning Author(s) References Examples

View source: R/lcmodels.R

The R package provides individual risks of lung cancer and lung cancer death based on various published papers: Bach et al., 2003; Spitz et al., 2007; Cassidy et al., 2008 (LLP); Hoggart et al., 2012; Tammemagi et al., 2013 (PLCOm2012); Marcus et al., 2015 (LLPi); Wilson and Weissfeld, 2015 (Pittsburgh); Katki et al., 2016 (LCRAT and LCDRAT).

1	lcmodels(x)

x

A data frame or matrix containing individuals' covariate values. Covariates should be in the following columns and numerical formats:

column 1 - current age (numeric);
column 2 - gender (1=Female, 0=Male);
column 3 - years smoked (numeric);
column 4 - years quit (numeric with 0 to indicate current smoker);
column 5 - cigarettes per day (numeric);
column 6 - race (0=Non-hispanic white, 1=Non-hispanic Black/African American, 2=Hispanic, 3=Other Ethnicity);
column 7 - lung disease (1=COPD or Emphysema, 0=No COPD or Emphysema);
column 8 - number of first degree relatives with lung cancer (0,1,2);
column 9 - bmi;
column 10 - highest education level (1=<12 grade, 2=HS graduate, 3=post hs, no college, 4=associate degree/some college, 5=bachelors degree, 6=graduate school);
column 11 - asbestos exposure binary indicator;
column 12 - prior history of pneumonia indicator;
column 13 - prior history of cancer indicator;
column 14 - family history of lung cancer (0=none, 1=early onset, 2=late onset);
column 15 - Dust exposure (binary indicator);
column 16 - 2 or more first degree relatives with cancer (binary indicator);
column 17 - 1 or more first degree relatives with smoking cancer (binary indicator);
column 18 - no hay fever (binary indicator);
column 19 - asian ethnicity (binary indicator);
column 20 - islander ethnicity (binary indicator);
column 21 - American indian ethnicity (binary indicator);

Risk factors used by each model

Covariate	Bach	Spitz	LLP	Hoggart	PLCOm2012	Pittsburgh	LLPi	LCRAT	LCDRAT
Age	Yes	Yes	Yes	Yes	Yes	Yes	Yes	Yes	Yes
Gender	Yes	Yes	Yes	No	No	No	Yes	Yes	Yes
Race/ethnicity	No	No	No	No	Yes	No	No	Yes	Yes
Asian	No	No	No	No	Yes	No	No	No	No
Pacific islander	No	No	No	No	Yes	No	No	No	No
America Indian	No	No	No	No	Yes	No	No	No	No
Education	No	No	No	No	Yes	No	No	Yes	Yes
BMI	No	No	No	No	Yes	No	No	Yes	Yes
Smoking status	No	Yes	No	Yes	Yes	Yes	No	No	No
Years/Age quit	Yes	Yes	No	Yes	Yes	No	No	Yes	Yes
Years smoked	Yes	No	Yes	Yes	Yes	Yes	Yes	Yes	Yes
Cigs per day	Yes	No	No	Yes	Yes	Yes	No	Yes	Yes
Pack-years	No	Yes	No	No	No	No	No	Yes	Yes
Prior cancer	No	No	Yes	No	Yes	No	Yes	No	No
Lung disease	No	Yes	No	No	Yes	No	Yes	Yes	Yes
Pneumonia	No	No	Yes	No	No	No	No	No	No
Hayfever	No	Yes	No	No	No	No	No	No	No
Asbestos exposure	Yes	Yes	Yes	No	No	No	No	No	No
Dust exposure	No	Yes	No	No	No	No	No	No	No
Any FDR w/ cancer	No	Yes	No	No	No	No	No	No	No
Any FDR w/ smoking-related cancer	No	Yes	No	No	No	No	No	No	No
Any FDR w/ LC	No	No	Yes	No	Yes	No	Yes	Yes	Yes
Num. FDR w/ LC	No	No	No	No	No	No	No	Yes	Yes
Early onset FDR w/ LC	No	No	Yes	No	No	No	Yes	No	No

A numeric matrix containing individuals' predictions:

column 1 - An indicator variable for whether the individual is eligible for CT lung screening according to US Preventive Services Task Force (USPSTF) recommendations.
column 2 - This is the probability of dying from lung cancer within 5 years if not undergoing screening (Katki, 2016).
column 3 - This is the reduction in the probability of dying from lung cancer in 5 years
column 4 - This is the probability of being diagnosed with lung cancer within 5 years if not undergoing screening (Katki, 2016).
column 5 - This is the extra probability of lung cancer diagnosis in 5 years if undergoing 3 yearly CT lung screens as in the NLST (Katki, 2016).
column 6 - This is the probability of having at least one false-positive CT screen out of 3 screens (Katki, 2016).
column 7 - This is the expected number of false-positive CT screens after 3 screens (Katki, 2016).
column This is the probability of being diagnosed with lung cancer within 10 years if not undergoing screening (Bach, 2003).
column 9 - This is the probability of being diagnosed with lung cancer within 1 years if not undergoing screening (Hoggart, 2012).
column 10 - This is the probability of being diagnosed with lung cancer within 5 years if not undergoing screening (LLP, 2008).
column 11 - This is the probability of being diagnosed with lung cancer within 8.7 years if not undergoing screening (LLPi, 2015).
column 12 - This is the probability of being diagnosed with lung cancer within 1 years if not undergoing screening (Spitz, 2007).
column 13 - This is the probability of being diagnosed with lung cancer within 6 years if not undergoing screening (Tammemagi, 2013).
column 14 - This is the probability of being diagnosed with lung cancer within 6 years if not undergoing screening (Pittsburgh, 2015).

VGAM is a required dependency of this package. VGAM may automatically be installed the first time this package is used. Inputs must be in numerical format to ensure correct output. For data frame x, this can be checked using sapply(x,class)

Li C. Cheung, li.cheung@nih.gov, Stephanie A. Kovalchik, Hormuzd A. Katki

Bach PB, Kattan MW, Thornquist MD, et al. Variations in lung cancer risk among smokers. J Natl Cancer Inst 2003;95:470-8.
Spitz MR, Hong WK, Amos CI, et al. A risk model for prediction of lung cancer. J Natl Cancer Inst 2007;99:715-26.
Cassidy A, Myles JP, van Tongeren M, et al. The LLP risk model: an individual risk prediction model for lung cancer. Br J Cancer 2008;98:270-6.
Hoggart C, Brennan P, Tjonneland A, et al. A risk model for lung cancer incidence. Cancer Prev Res (Phila) 2012;5:834-46.
Tammemagi MC, Katki HA, Hocking WG, et al. Selection criteria for lung-cancer screening. N Engl J Med 2013;368:728-36.
Marcus MW, Chen Y, Raji OY, Duffy SW, Field JK. LLPi: Liverpool lung project risk prediction model for lung cancer incidence. Cancer Prev Res (Phila) 2015;8:570-5.
Wilson DO, Weissfeld J. A simple model for predicting lung cancer occurrence in a lung cancer screening program: the Pittsburgh predictor. Lung Cancer 2015;89:31-37.
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;315:2300-11. doi: 10.1001/jama.2016.6255.
Katki HA, Kovalchik SA, Petito LC, Cheung LC, Jacobs E, Jemal A, Berg CD, Chaturvedi AK. Implications of nine risk prediction models for selecting ever-smokers for computed tomography lung cancer screening. Ann Intern Med. 2018;doi: 10.7326/M17-2701.

age <- c(66,58,75,72,56)
bmi <- c(23,28,26,27,24)
cpd <- c(36,36,40,24,40)
emp <- c(0,1,1,0,1)
fam.lung.trend <- c(0,2,0,2,0)
female <- c(0,1,0,1,0)
smkyears <- c(43,37,45,42,29)
qtyears <- c(0,0,9,6,6)
race <- c(0,1,2,2,3)
edu6 <- c(3,5,4,5,5)
asb <- c(0,0,0,0,0)
pneu <- c(0,0,0,0,0)
prior.cancer <- c(0,0,0,0,0)
fam.cancer.onset <- c(0,1,0,2,0)
dust <- c(0,0,0,0,0)
fam.cancer <- c(0,1,0,1,0)
fam.smoke.cancer <- c(0,1,0,1,0)
no.hayfever <- c(1,1,1,1,1)
asian <- c(0,0,0,0,1)
islander <- c(0,0,0,0,0)
indian <- c(0,0,0,0,0)

persons <- data.frame(age,
                      female,
                      smkyears,
                      qtyears,
                      cpd,
                      race,
                      emp,
                      fam.lung.trend,
                      bmi,
                      edu6,
                      asb,
                      pneu,
                      prior.cancer,
                      fam.cancer.onset,
                      dust,
                      fam.cancer,
                      fam.smoke.cancer,
                      no.hayfever,
                      asian,
                      islander,
                      indian)

persons_predictions <- lcmodels(persons)
persons_predictions