The Cost-Effectiveness of AI-Assisted Colonoscopy as a Primary or Secondary Screening Test in a Population-Based...

Area of Science:

• Health Economics and Outcomes Research
• Gastroenterology and Oncology
• Medical Technology Assessment

Background:

• Colorectal cancer (CRC) is a leading global cancer, necessitating effective screening strategies.
• Conventional colonoscopy, while effective, is limited by physician experience.
• Artificial intelligence (AI)-assisted colonoscopy shows promise for improved CRC detection but its cost-effectiveness requires evaluation.

Purpose of the Study:

• To evaluate the cost-effectiveness of various colorectal cancer (CRC) screening strategies in an Asian population.
• To compare strategies including no screening, fecal immunochemical test (FIT) followed by conventional or AI-assisted colonoscopy, and direct colonoscopy or AI-assisted colonoscopy.

Main Methods:

• A hypothetical Asian population model was used, with screening starting at age 50.
• Cost-effectiveness was assessed using incremental cost-effectiveness ratios (ICERs) and metrics like life years saved.
• A Markov model decision analysis incorporated data from literature and government sources.

Main Results:

• FIT followed by AI-assisted colonoscopy demonstrated the lowest ICER (US $122,539) compared to other strategies.
• This strategy prevented more CRC cases (4.1%) and saved more life years (308 y) than FIT followed by conventional colonoscopy.
• AI-assisted colonoscopy alone dominated conventional colonoscopy, showing a negative ICER of -US $39,040.

Conclusions:

• For an Asian population, fecal immunochemical test (FIT) followed by AI-assisted colonoscopy is the most cost-effective CRC screening strategy.
• This strategy offers superior value compared to no screening, FIT with conventional colonoscopy, and direct colonoscopy or AI-assisted colonoscopy.