Skip to content

A Randomized Controlled Trial on the Impact of AI-Assisted Decision-Making on Healthcare Quality and Efficiency

A Randomized Controlled Trial on the Impact of AI-Assisted Decision-Making on Healthcare Quality and Efficiency

Status
Active, not recruiting
Phases
Unknown
Study type
Interventional
Source
ChiCTR
Registry ID
ChiCTR2500112977
Enrollment
Unknown
Registered
2025-11-21
Start date
2025-12-01
Completion date
Unknown
Last updated
2026-09-07

For informational purposes only — not medical advice. Sourced from public registries and may not reflect the latest updates. Terms

Conditions

Chronic disease risk identification and health management in health check-up populations

Interventions

LLM-assisted physician group:Junior physicians use the LLM to assist completion of standardized health check-up reports and review and revise the model-generated content.
Online-search-assisted group:Use conventional online search to assist completion of the standardized health check-up report, without an LLM tool.
International online-search-assisted group:Complete the assigned standardized reporting tasks within the study site using conventional online search only, without an LLM tool.
International LLM-assisted physician group:Complete the assigned standardized reporting tasks within the study site using LLM assistance, with review and revision by the junior physician.
Conventional-report-first sequence group:First review the conventional report based on the participant's own health check-up results and complete the single-report assessment
after a 30-minute interval, review and assess the LLM-assisted report, followed by a direct comparison of the two reports.
LLM-assisted-report-first sequence group:First review the LLM-assisted report based on the participant's own health check-up results and complete the single-report assessment
after a 30-minute interval, review and assess the conventional report, followed by a direct comparison of the two reports.
LLM-assisted report group:Receive an LLM-assisted physician report
undergo blinded telephone follow-up after report delivery.
Conventional-report group:Receive a physician-completed report according to the routine workflow of the health-check centre and undergo blinded telephone follow-up after report delivery.
senior raters:Evaluate and score based on the report received.

Sponsors

Tsinghua University
Lead Sponsor

Eligibility

Sex/Gender
All
Age
18 Years to No maximum

Inclusion criteria

Inclusion criteria: 1. Component 1 senior physicians: deputy-senior professional title or above, or at least 5 years of professional experience related to health check-ups; written informed consent. 2. Component 2 China junior physicians: early-career physicians with no more than 3 years of work experience at a participating centre; written informed consent. Senior-physician raters must hold a deputy-senior professional title or above. 3. Component 3 international junior physicians: eligible junior physicians according to the requirements of the participating site; written informed consent. 4. Components 4 and 5 participants: adults aged 30–70 years undergoing routine health check-ups at the Urumqi Health Management Center; written informed consent.

Exclusion criteria

Exclusion criteria: Doctor: substantial involvement in a similar AI pilot during the previous 3 months; unable to complete all study tasks; major concurrent factors likely to interfere with the study. Health-check population: unstable major disease; evident cognitive impairment or inability to complete follow-up; participation in another intervention study that may interfere with this study.

Design outcomes

Primary

MeasureTime frame
Overall quality of health check-up interpretation reports;Four report-quality domain scores (clinical accuracy, report quality, humanistic care, and long-term planning);Sixteen individual report-quality item scores;Participant ratings of health check-up interpretation reports across eight dimensions;Direct paired preference between conventional and LLM-assisted report formats;Number of correctly recalled baseline-documented abnormality categories;

Secondary

MeasureTime frame
Four report-quality domain scores(clinical accuracy, report quality, humanistic care and long-term planning);Overall lifestyle modification;Increased exercise;Improved diet;Weight-control or weight-reduction action;Smoking reduction or cessation;Reduced alcohol intake;Active seeking of health-related information;Hospital follow-up or medical consultation;Initiation or adjustment of medication;Perceived change in overall health;Change in body weight;Perceived impact of the previous health check-up report on subsequent actions;Change in attention to health;Willingness to undergo regular health check-ups in the future;

Countries

China

Contacts

Public ContactDi Qian

Vanke School of Public Health, Tsinghua University

qiandi@tsinghua.edu.cn+86 138 1191 1050

Outcome results

None listed

Source: ChiCTR (via WHO ICTRP) · Data processed: Sep 19, 2026