现在的位置: 首页 >研究点评, 进展交流>正文
[JAMA Intern Med发表编者按]:自动化并非所有问题的答案
2026年11月19日 ⁄ 研究点评, 进展交流 ⁄ [JAMA Intern Med发表编者按]:自动化并非所有问题的答案已关闭评论

Editor's Note 

Automation Is Not Always the Answer

Raegan W. Durant, Cary P. Gross, Ilana Richman

JAMA Intern Med Published Online: August 10, 2026

doi: 10.1001/jamainternmed.2026.3664

Despite ample evidence for the benefits of lung cancer screening,1 the uptake of low-dose computed tomography (LCDT) for both initial and annual follow-up screening remains low. Among individuals eligible for screening with a negative baseline LDCT finding, adherence to LDCT has been estimated at 61.2% in the time window for a 12-month follow-up, and adherence falls to near 50% in the window for a 24-month follow-up.2 Physicians, scientists, and health system leaders have explored multiple approaches to alerting both patients and physicians to the need for screening, yet many eligible individuals remain unscreened.

In this issue of JAMA Internal Medicine, Wernli et al3 tested an intervention based on stepped reminders for ordering and scheduling follow-up LDCT scans among patients with a normal LDCT finding. The investigators designed a stepped reminder intervention in which a medical assistant would (1) create an LDCT order for the physician to sign and (2) communicate with eligible patients 4 weeks prior to the due date of the next scan, first by patient portal and, if unsuccessful, by letter or phone to encourage them to schedule an LDCT with the radiology department. The comparator intervention (health communication) was composed of a portal message or mailed letter sent 3 weeks after the index scan including an educational video or written content, respectively, on lung cancer screening. In a 2 × 2 factorial analysis, adherence to guideline-concordant annual screening was 46.1% among those randomized to health communication only, 78.1% among stepped reminders only, 72.8% among a combination of both, and 48.7% among those randomized to usual care. These results suggest that the stepped reminders (to both the physician and the patient) were vital to success of the intervention.

Several elements may account for the comparative success of the stepped reminder intervention. First, the intervention used multiple strategies to prompt repeat LDCT, including patient and physician reminders. Second, the intervention was deployed just before the next LDCT due date, which may reflect optimal timing. Last, the intervention made effective use of nonphysician team members. Approximately 30% of patients in the intervention arm who scheduled an LDCT did so after the clinical team member escalated the reminder from portal prompt or letter to a phone call. Furthermore, including clinical team members as part of a stepwise reminder effort allows for calibration and escalation in both an iterative and efficient fashion to achieve the desired screening goal.

With the advent of artificial intelligence, clinical care is likely to become even more reliant on automated reminders and messages to physicians and patients. However, we must remember that options, including varied mode of delivery and timing for prompts and the involvement of other clinical team members, may yield more tailored approaches toward the optimal success with screening uptake—one step at a time.

抱歉!评论已关闭.

×
腾讯微博