AI-assisted colonoscopy increases polyp and adenoma detection in routine screening

Embargoed for release until 5:00 p.m. ET on Monday 21 October 2024    

@Annalsofim         
Below please find summaries of new articles that will be published in the next issue of Annals of Internal Medicine. The summaries are not intended to substitute for the full articles as a source of information. This information is under strict embargo and by taking it into possession, media representatives are committing to the terms of the embargo not only on their own behalf, but also on behalf of the organization they represent.         
----------------------------         

1. AI-assisted colonoscopy increases polyp and adenoma detection in routine screening 

Abstract: https://www.acpjournals.org/doi/10.7326/ANNALS-24-02604

URL goes live when the embargo lifts           

A systematic review of randomized clinical trials (RCTs) comparing computer-aided detection (CADe) system-enhanced colonoscopy and conventional colonoscopy found that CADe (also known as artificial Intelligence- or AI-assisted) colonoscopies may increase overall detection of colonic polyps and adenomas, or precancerous growths, with a small increase in procedure time. Equivocal results were found regarding detection of advanced colonic neoplasia (ACN), with a small increase in ACN detection rate but no difference in ACN detected per colonoscopy. The findings are published in Annals of Internal Medicine.  

Researchers from Yale University and Mass General Brigham, Harvard School of Medicine comprehensively searched several large scientific research databases for RCTs comparing colonic lesion detection with standard colonoscopy versus AI-assisted colonoscopy with polyp detection (CADe) systems. The authors compared average adenoma per colonoscopy (APC) and ACN per colonoscopy for both screening methods. Secondarily, they compared adenoma detection rate (ADR), adenoma miss rate (AMR), and ACN detection rate between the two colonoscopy types. They found that AI-assisted colonoscopy found more polyps and precancerous growths in the colon than conventional colonoscopy. However, AI-assisted colonoscopy detected marginally more serious growths (ACNs) than conventional colonoscopy but was no better than the conventional method at finding ACNs per colonoscopy.  

The researchers note that there are no clear differences in benefit for detecting adenomas across different AI systems for CADe, and that there was an increase in benefit for providers with lower adenoma detection rate or those without a prior fecal immunochemical test result. They suggest that future studies focus on interval post colonoscopy colorectal cancer and may consider a study design that randomizes colonoscopists rather than patients.    

Media contacts: For an embargoed PDF, please contact Angela Collom at acollom@acponline.org. To speak with corresponding author Dennis L. Shung, MD, MHS, PhD, please email Colleen Moriarty at colleen.moriarty@yale.edu.  

----------------------------    

2. Study suggests FIT may be an efficient way to screen younger patients for colorectal cancer

Abstract: https://www.acpjournals.org/doi/10.7326/M24-0743      

URL goes live when the embargo lifts   

A retrospective cohort study found that younger patients will complete colorectal cancer screening using fecal immunochemical tests (FIT) and follow up with a colonoscopy as well as people who are 50 years old or older. These findings suggest that FIT is a very efficient way to get people screened and support newer guidelines recommending that patients start screening for colorectal cancer at age 45. The study is published in Annals of Internal Medicine.

Researchers from Kaiser Permanente Northern California, Washington and Colorado studied patient health records for 213,928 patients aged 45 to 49 and 53,804 patients aged 50 to compare FIT screening completion and outcomes between the two groups. Patients included in the analysis had no underlying bowel condition or previous cancer diagnosis and were undergoing FIT for the first time. The researchers analyzed the data for FIT completion within 3 months, FIT positivity, follow-up colonoscopy completion within 3 months of positive FIT result, findings at the follow-up colonoscopy, and false-positive FIT results for both age groups. They found that overall FIT completion was slightly higher in patients aged 45 to 49 versus 50. For all health systems combined, FIT positivity was slightly lower in patients in the younger age group, and the frequency of completion of a follow-up colonoscopy after a positive FIT test was not substantially different between age groups. For those who completed a follow-up colonoscopy, detection of any adenoma was lower among those aged 45 to 49 than 50. The authors note that this is the largest population-based report to date of FIT screening completion and yield among patients aged 45 to 49 years. The findings show that the yield of colonoscopy after a positive FIT are similar between both groups, despite an overall risk of CRC that is lower among those aged 45-49.

Media contacts: For an embargoed PDF, please contact Angela Collom at acollom@acponline.org. To speak with corresponding author Theodore R. Levin, MD, please email Sue Rochman at Susan.L.Rochman@kp.org.

----------------------------         

3. Leveraging care transitions in hospital medicine identifies the rate of diagnostic errors  

Researchers observed that missing electronic health record data as a common source of error

Abstract: https://www.acpjournals.org/doi/10.7326/ANNALS-24-00563  

URL goes live when the embargo lifts         

A single-center study analyzed a program leveraging hospitalist end-of-week care transitions to better identify diagnostic errors and possible causes in hospitalized patients. The data revealed that using care transitions to identify diagnostic errors is a feasible process and yields a lower rate of attributable errors than in other studies. The study is published in Annals of Internal Medicine.

Between April 2019 and August 2021, researchers developed a program at Northwestern University’s 897-bed teaching hospital to leverage end-of-week care transitions to identify diagnostic errors. The researchers aimed to determine a more optimal approach to identifying diagnostic errors in hospital medicine patients. Thirty-six hospitalists assuming care completed an electronic questionnaire on day two of their service week, asking whether any diagnoses for 1 to 2 randomly selected patients had changed and for a description of the original and new diagnoses. Cases in which a diagnostic change had occurred were independently reviewed to determine whether the change was due to a diagnostic error. Diagnostic errors were labeled as one of four categories: missed information from the electronic health record (EHR), failure to order a test or consultation, failure to consider a diagnosis or a missed examination finding. The researchers found that among 1,011 patients, 184 (18.2%) had a diagnostic change and 59 (5.8%) had a diagnostic change associated with a diagnostic error. The most common cause of error was missed information from the EHR. The percentage of patients with a diagnostic error was lower than described in other studies which used patient outcomes (ex: ICU transfers, deaths, readmissions) as a trigger. This study demonstrates that examining errors at transitions within a similar level of care provides an accurate estimate of diagnostic errors. 

Media contacts: For an embargoed PDF, please contact Angela Collom at acollom@acponline.org. To speak with corresponding author Gopi J. Astik, MD, MS, please email Jennifer Delacruz at jen.delacruz@nm.org.

----------------------------    

4. Online “Safety in Dementia” tool helps caregivers prepare for decisions about firearm access

Abstract: https://www.acpjournals.org/doi/10.7326/ANNALS-24-00763  

URL goes live when the embargo lifts       

A prospective 2-group randomized trial found that the Safety in Dementia (SiD) online decision aid increased preparation for decision making about firearm access for those caring for people with Alzheimer disease and related dementias (ADRD). The SiD decision aid is broken down into 5 major sections that address general safety in dementia, firearms, driving, home safety, and additional resources and then has a specific firearms section with scenarios and solutions for caregivers. According to the researchers, use of this type of resource in clinical or community settings may support caregivers and people with ADRD in avoiding firearm injury or death. The study is published in Annals of Internal Medicine.

Those caring for persons with ADRD have little guidance on how to manage firearm access and cite concerns about when and how to take action, including legal concerns related to firearm transfer and ownership. They also express concerns about how to avoid angering or upsetting the person with ADRD while trying to prevent firearm-related injury to the person with ADRD or those around them. A survey of caregivers revealed that they would accept counseling to address these issues, but few reported ever having it.

Researchers from the National Institute on Aging and National Institutes of Health sought to determine if the SiD decision aid would improve caregivers’ preparation related to firearm access for their charge with ADRD. They assigned 500 caregivers of community-dwelling adults with ADRD and firearm access to either access to the SiD decision aid or a control information website. Based on a post-intervention caregiver questionnaire, the authors assessed preparation for decision making about firearm access and self-reported action to reduce firearm access for the person with ADRD. The researchers found that those in the SiD intervention group had significantly greater preparation for decision making than those in the control group. However, the SiD had no significant effect on caregivers’ self-reported action to reduce access at short-term follow-up. The caregiver participants in both groups reported high levels of suicidality, depressive symptoms, and potentially dangerous behaviors among this population of people with ADRD who have firearm access.  The researchers note that this study represents an important step forward in both describing and providing ADRD caregivers with effective tools to navigate decisions around firearm access in people with ADRD. The study also adds to the current understanding of suicidal ideation, depressive symptoms and behavior issues in people with ADRD who have firearm access. The results reinforce the notion that web-based decision-aids could offer the potential for broad dissemination of firearm safety resources for ADRD caregivers without compromising on efficacy.

Media contacts: For an embargoed PDF, please contact Angela Collom at acollom@acponline.org. To speak with corresponding author Marian E. Betz, MD, MPH, please email marian.betz@cuanschutz.edu.

---------------------------- 

Also new in this Issue:

Prescription for Change: Health Care Professionals and Advocacy for Farm Bill Reform

Christina Badaracco, MPH, RD, LDN; Farshad F. Marvasti, MD, MPH; Jaclyn Albin, MD; and Olivia Thomas, MS, RD, LDN

Ideas and Opinions

Abstract: https://www.acpjournals.org/doi/10.7326/ANNALS-24-02145