CT Lung Screening Patient Triage

Value Proposition

COVID-19 has disrupted the full spectrum of cancer care, including oncologic imaging. In particular, cancer screening efforts have been significantly impaired. Stay at home orders and associated triaging of outpatient imaging exams have resulted in screening appointments being delayed or temporarily canceled. Having survived the peak of the first wave of COVID-19 cases, healthcare facilities are now able to resume scheduling routine healthcare procedures, including screening exams. Algorithms to risk stratify patients will inform scheduling priority for delayed screening and follow-up.  

Narrative(s)

A radiology practice is ready to resume their CT lung screening activities. Having placed low-dose computed tomography (LDCT) exams on hold in anticipation of the COVID-19 pandemic surge, they have a backlog of delayed and deferred screening exams including initial screening exams (baseline/prevalence scans), annual scans (incidence scans), and early follow-up imaging exams for findings identified on the prior prevalence and incidence scans. The practice plans to call patients and schedule them for their imaging exams. Lung cancer is the leading cause of cancer deaths. When identified outside of screening, lung cancer is most often found at a late stage, with five-year survival rates of less than 10%. It is important to prioritize the resumption of screening for patients at highest risk for having lung cancer, in order to maximize impact on downstream survival. 

Workflow Description

Compile a list of CT lung screening program participants that have pending, delayed, or deferred exams. Most screening programs store information in a relational database and/or have a searchable electronic health record. New lung cancer screening enrollees with pending baseline screening orders will need to be added to the list if not already in the database.

Patients will be triaged for return to screening based on their relative risk of lung cancer as compared to the rest of the screening program cohort. The use of lung cancer risk prediction models (e.g., PLCOm2012) is recommended to integrate new participants into the existing participant cohort. Risk assessment variables include Lung-RADS® category, pulmonary nodule malignancy risk score (e.g., using PanCan Malignancy Probability Model) if applicable, lung cancer risk score (e.g., PLCOm2012 or other model depending on available patient data), duration of delay, radiologist report text mining for keywords tied to risk (e.g., nodules with spiculation, ground glass nodule that doubles in size in 1 year, enlarged lymph nodes, nodules of certain size, new nodules, nodules that are growing in size, etc.). There are options for simple and more comprehensive AI models/algorithms depending on the cohort size, cohort complexity, and available data.

The simplest model would be to risk stratify based on just the Lung-RADS score. However, that makes it difficult to integrate new screening program participants without putting them as the lowest risk. It also assumes everyone in a given Lung-RADS category to be at equivalent risk. For small cohorts, this may be the best approach as they can be scheduled based on patient preference for available screening appointments. However, for large cohorts, this may result in delaying the return to screening for higher-risk participants resulting in later stage cancer detection and worse outcomes. 

Inputs

Primary Outputs

Lung-RADS Category and Score by Patient

Lung Cancer Risk Model Assessment

Triage List according to Lung Cancer Risk Information

Input screening site CT lung screening open appointment schedule, populate with patient name and contact info by triage order with the earliest appointment for the first patient on the triage list, and so forth. 

Generate automatic portal mail, e-mail message, text, or phone messages for patients with their scheduled appointment date and time, instructions on coming in for screening, and the option to change appointment time or date or opt-out of screening.   

1. https://brocku.ca/lung-cancer-screening-and-risk-prediction/risk-calculators/ 

2. González Maldonado S, Delorme S, Hüsing A, et al. Evaluation of Prediction Models for Identifying Malignancy in Pulmonary Nodules Detected via Low-Dose Computed Tomography. JAMA Netw Open. 2020;3(2):e1921221. doi:10.1001/jamanetworkopen.2019.21221

3. Katki HA, Kovalchik SA, Petito LC, et al. Implications of Nine Risk Prediction Models for Selecting Ever-Smokers for Computed Tomography Lung Cancer Screening. Ann Intern Med. 2018;169(1):10‐19. doi:10.7326/M17-2701

4. Mazzone PJ, Silvestri GA, Patel S, Kanne JP, Kinsinger LS, Wiener RS, Soo Hoo G, Detterbeck FC. Screening for Lung Cancer: CHEST Guideline and Expert Panel Report. Chest. 2018;153(4):954-985. doi:10.1016/j.chest.2018.01.016

5. Tammemägi MC. Improving Implementation of Lung Cancer Screening With Risk Prediction Models. Ann Intern Med. 2018;169(1):54‐55. doi:10.7326/M18-0986

6. https://www.acr.org/-/media/ACR/Files/RADS/Lung-RADS/LungRADSAssessmentCategoriesv1-1.pdf?la=en 

7. https://www.acr.org/-/media/ACR/Files/RADS/Lung-RADS/LungRADSAssessmentCategoriesv1-1.pdf?la=en 

8. https://www.acr.org/-/media/ACR/Files/RADS/Lung-RADS/LungRADSAssessmentCategoriesv1-1.pdf?la=en 

No known related public datasets at this time,  please alert us if you know of any.