Ensure Patient Follow-Up of Radiology Report Recommendations

Purpose

Encourage development and implementation of patient follow-ups in radiology report recommendations.

Tag(s)

 

Panel

Non-Interpretive, Patient Facing

Define-AI ID

22110011

Originator

Patient Facing Panel

Panel Chair

Andrea Borondy Kitts

Panel Reviewers

Alexander J. Towbin, Melissa Davis

License

Creative Commons 4.0

Status Public Comment 
                               

Clinical Implementation


Value Proposition

Lung cancer is the leading cause of cancer deaths in the US and globally.(1,2) Two large randomized clinical trials, National Lung Screening Trial (NLST) and Dutch-Belgium NELSON trial, have shown that lung cancer screening with low dose computed tomography (LDCT) for high risk individuals can reduce mortality from lung cancer by at least 20%.(3,4) The Lung RADS structured reporting categories are predicated on the high risk individual returning for annual screening. Unfortunately adherence rates to annual screening in the US are only 57%.(5)  In the NLST about half of all the lung cancers were found on the follow-up annual screening (incidence scans) so it’s important for individuals to return for their annual screening LDCT to realize the mortality benefit from screening. Individuals who quit smoking were less likely to return for their annual screening scan. 

1.Sung H, Ferlay J, Siegel RL, et al. Global Cancer Statistics 2020: GLOBOCAN Estimates of Incidence and Mortality Worldwide for 36 Cancers in 185 Countries. CA: a cancer journal for clinicians 2021; 71(3): 209-49.

2.Siegel RL, Miller KD, Fuchs HE, Jemal A. Cancer statistics, 2022. CA: a cancer journal for clinicians 2022; 72(1): 7-33.

3.Aberle DR, Adams AM, Berg CD, et al. Reduced lung-cancer mortality with low-dose computed tomographic screening. The New England journal of medicine 2011; 365(5): 395-409.

4. de Koning HJ, van der Aalst CM, de Jong PA, et al. Reduced Lung-Cancer Mortality with Volume CT Screening in a Randomized Trial. The N

5.Lin Y, Fu M, Ding R, Inoue K, Jeon CY, Hsu W, Aberle DR, Prosper AE. Patient Adherence to Lung CT Screening Reporting & Data System-Recommended Screening Intervals in the United States: A Systematic Review and Meta-Analysis. J Thorac Oncol. 2022 Jan;17(1):38-55. doi: 10.1016/j.jtho.2021.09.013. Epub 2021 Oct 6. PMID: 34624528; PMCID: PMC8692358.

Narrative(s)

A 55-year-old male, with a 40 pack year smoking history who quit 10 years ago, underwent recommended low-dose CT lung screening. Results showed a benign finding (LungRADs Category 2), appropriate communication with the referring clinician was made, and the patient was recommended a routine 1 year follow-up for repeat CT. A year later, the patient does not return for a follow-up CT.

Workflow Description

A month before the one year recommended follow-up date the AI algorithm checks for a follow-up appointment. If none is found, the patient and the referring clinician are sent reminders. If found, the AI checks that it was completed as planned one day after the appointment date. For unscheduled/unfulfilled appointments, reminders are sent to the patient and referring clinician at 30, 60 and 90 days after the recommended follow-up date. For appointments overdue more than 90 days, the patient and referring clinician will be added to a contact list and screening site administrative staff will attempt phone contact to schedule the follow-up. 

AI Algorithm Details:

 

Add a code to each CT lung screening exam radiologist report with a required follow-up. The codes will be added to the report electronically by the radiologist from a drop down menu. Codes will be developed for different follow-ups: return scans, 3 month, 6 month, and annual. Most scans will have follow-up required. Scans without follow-up will be for patients who are no longer eligible for screening (ages out, exceeds 15 year quit threshold, is diagnosed with lung cancer). 

Every week, the AI algorithm will search the screening site EMR (or other screening site appropriate electronic database) and create a file with a list of individuals that need follow-up. The AI then searches for scheduled or completed appointments a month before and a month after the follow-up date and checks for appropriate appointment type/department.

Individuals from the follow-up list who attended or scheduled an appointment will be annotated with a follow-up code, completed or pending appointment. Pending appointments will be followed up after the appointment date to ensure completion. Individuals from the follow-up list who have not completed appointments will be contacted via preferred communication method. The referring provider will be sent a notice indicating their patient has a follow-up due and will request the provider to contact the patient or indicate if the follow-up was completed elsewhere. 

This process will be run for currently due recommendations, recommendations  overdue by one month, two months and three months. The individual with overdue recommendations without scheduled appointments will be sent reminders about the follow-up with a link to make an appointment as well as the referring physician. The reminder will include a short rationale for why the follow-up is important. 

Individuals who continue to be overdue for their recommended follow-up for more than 3 months will be added to an action contact list. These individuals will be contacted by screening site administrative staff by phone to schedule the follow-up. If the individual has received follow-up elsewhere a note will be put in the file that follow-up is completed and the individual will be removed from the list. If the individual or their referring clinician opts them out they will be removed from the long term follow-up list and a note put in the file that they opted out. 


Considerations for Dataset Development


Technical Specifications


Inputs 

Radiology Report, LungRADS categories and follow-up and incidental findings categories and follow-up

Procedure

CT lung screening (LDCT)

Views

N/A
Data Type Categorical
Modality CT
Body Region Thorax in LDCT and Surrounding Field of View in LDCT
 Anatomic Focus Lungs
 Pharmaceutical N/A
 Scenario N/A
 

Primary Outputs

Identifier codes for follow-up LDCT scan frequency, identifier codes for incidental finding follow-up recommendations, description of follow-up recommendation, letter, e-mail, text, phone script

RadElement ID N/A
 Definition Identifier code for follow-up recommendations, descriptions of follow-up recommendations and importance of adherence
 Data Type Numerical for identifier code, text for follow-up communication descriptions
 Value Set N/A
 Units N/A

Secondary Outputs

Adherence to follow-up recommendations

RadElement ID

N/A

Definition

Number of recommended follow-ups completed divided by total number of scans with recommended follow-ups

Data Type

Numerical

Value Set

N/A

Units

Ratio

Future Development Ideas


  • Automate follow-up recommendation appointment scheduling when the recommendation is made.

  • Include patient surveys on the reason for delayed adherence and suggestions for making it easier to do the follow-up.