Automatic system boosts adherence to colorectal post-polypectomy surveillance guidelines
Click Here to Manage Email Alerts
Key takeaways:
- An automatic system was up to 99.77% accurate in identifying patients for surveillance after colorectal polypectomy.
- The system was associated with improved physician adherence to guidelines and reduced staff workload.
An automatic surveillance system that accurately identified patients after colorectal polypectomy and assigned surveillance intervals by risk improved guideline adherence among physicians and reduced staff workload, according to data.
“More than half of patients with a colorectal polyp resected do not receive surveillance colonoscopy within the recommended time,” Lianlian Wu, MD, of the department of gastroenterology at Renmin Hospital of Wuhan University, and colleagues wrote in JAMA Network Open. “Accurate identification of patients with different risk stratification and timely reminders is essential for increasing the surveillance rate of patients after polypectomy and thereby improving early diagnosis.”
They continued: “In the field of AI in gastrointestinal endoscopy, there are numerous studies focusing on computer-aided lesion detection and diagnosis, while few studies pay attention to automatic surveillance for patients.”
In a diagnostic/prognostic study, Wu and colleagues collected endoscopic and pathological reports from 47,544 patients who underwent colonoscopy at three different hospitals, which would be used to train and test an automatic surveillance (AS) system.
The test set included 16,106 patients (mean age, 51.9 years; 47.75% women), who were divided among internal (n = 9,583), external 1 (n = 4,792) and external 2 (n = 1,731) sets. Researchers also conducted a multi-reader, multi-case trial ([MRMC] = 105) to evaluate use of the AS system and physician adherence to surveillance guidelines.
Other study outcomes included the system’s accuracy in identifying patients after polypectomy and assigning surveillance intervals according to risk, and whether the system could successfully contact patients and lessen the workload for health care providers.
According to results, the system had an overall accuracy of 99.91% (95% CI, 99.83-99.95) in the internal test, 99.54% (95% CI, 99.3-99.7) in the external 1 test and 99.77% (95% CI, 99.41-99.91) in the external 2 test when identifying patients for surveillance. Accuracy for stratifying patient risk levels and assigning surveillance intervals was 99.3% (95% CI, 98.67-99.63), 98.89% (95% CI, 98.33-99.27) and 98.56% (95% CI, 95.86-99.51), respectively.
In the MRMC trial, the mean accuracy and efficiency of physicians were significantly increased when using the AS system vs. not using the system (98.67% vs. 78.1%).
Moreover, results of a prospective trial of 88 patients showed the AS system successfully informed 93.18% of patients and was associated with reduced burden of follow-up time (0 hours vs. 2.86 hours).
“The system was associated with improved adherence to guidelines among physicians and reduced time spent by physicians and nurses, and it played a role in informing patients,” Wu and colleagues concluded. “These findings suggest that the system may be associated with improved early diagnosis of premalignant polyps and reductions in unnecessary over-examination and the workload of health care workers.”
Collapse
Read more about
Click Here to Manage Email Alerts