AdvanCing Health through Intelligent Evidence and liVing guidelines Enhancement
Scope & Vision
Research Interests
-
Health research methods on clinical guideline development, evidence synthesis (effectiveness on multiple interventions), and cohort studies
-
Incorporating AI technology in living guideline development and evidence synthesis
-
Health research on rare diseases and chronic diseases (e.g., insomnia, chronic pain)
-
Alternative medicine (TCM, acupuncture, et al)
Projects
-
Rapid recommendations for chronic Insomnia, PTSD, Chronic Fatigue, fibromyalgia
-
Loops consideration for indirect evidence in NMA
-
Holiday effect for hospital admission and surgery
-
AI-assisted evidence synthesis and living guidelines
Research vision
Our Team, the AdvanCing Health through Intelligent Evidence and liVing guidelines Enhancement (ACHIEVE), is a research group dedicated to advancing health through intelligent evidence and living guidelines enhancement. Our research aims to bridge the gap between clinical evidence and practice, enhancing patient outcomes and quality of life through reliable evidence synthesis and trustworthy guideline development.​​​
​
We are passionate about advancing health through intelligent evidence and living guidelines enhancement. Our team is committed to conducting high-quality research on clinical guideline development and living evidence synthesis. We aim to make a significant impact in the field of healthcare by providing valuable insights and contributing to evidence-based practices. ACHIEVE stands out for its dedication to excellence and innovation, and we are always looking for opportunities to collaborate with like-minded individuals and organizations.
Publications & News
2025.9​
​Dr Liu and Yao have been recognized among the World's Top 2% Scientists in the latest 'Single Recent Year Impact' metrics, compiled by Stanford University using Scopus data provided by Elsevier through the ICSR Lab (Institute for Scientific Information on Research). This marks Yao's fifth consecutive year receiving this honor since 2021.
​
2025.8
New in Lancet eClinicalMedicine: Our meta-analysis (11 RCTs, 5,505 pts) shows shorter antibiotic courses for bloodstream infections are as effective as longer ones—& cut hospital stay by ~3 days. Less exposure, same outcomes, lower resistance risk.
https://www.thelancet.com/journals/eclinm/article/PIIS2589-5370(25)00329-3/fulltext ​
​
2025.6
Just published an opinion in BMJ. This editorial addresses the use of artificial intelligence (AI) tools in evidence synthesis, especially guiding GRADE (Grading of Recommendations Assessment, Development, and Evaluation) evaluations and recommendation development.
key points:
-
AI assistance should support—not replace—careful expert oversight when applying GRADE frameworks for rating study quality or consistency.
-
All AI-generated steps or outputs must be clear and explainable to evidence users.
-
Users (clinicians, boards, policymakers) must be able to verify and understand how conclusions were reached—even if AI was involved.
​
2025.5
In the sixth paper in the Core GRADE series, we outlined best practices for constructing Summary of Findings (SoF) tables in systematic reviews and clinical guidelines.
SoF tables are crucial for transparently presenting evidence—highlighting key outcomes, effect sizes, certainty levels, and plain‑language explanations
​
2025.2
Our new publication on BMJ: Common interventional procedures for chronic non-cancer spine pain. Our study assessed the effectiveness of commonly used interventional procedures (such as epidural injections and nerve blocks) for chronic non‑cancer spinal pain—both axial (back) and radicular (nerve-related) – published along with a clinical guideline in BMJ in February 2025
​
NMA: https://www.bmj.com/content/388/bmj-2024-079971.full
​
Guideline: https://www.bmj.com/content/388/bmj-2024-079970
​​
2024.9
Liang Yao has been recognized among the World's Top 2% Scientists in the latest 'Single Recent Year Impact' metrics, compiled by Stanford University using Scopus data provided by Elsevier through the ICSR Lab (Institute for Scientific Information on Research). This marks Liang's fourth consecutive year receiving this honor since 2021.
​
https://topresearcherslist.com/Home/Search?AuthFull
​
.png)
This study investigated whether the alignment of strength of recommendations with quality of evidence differs in consensus-based versus evidence-based guidelines. The study included 12 ACC/AHA guidelines that generated 1,434 recommendations and 69 ASCO guidelines that generated 1,094 recommendations. ​
The findings suggest consensus-based guidelines produce more recommendations violating the principles of evidence-based medicine than evidence-based guidelines. Ensuring the appropriate alignment of quality of evidence with the strength of recommendations is key to the development of “trustworthy” guidelines. This work, for the first time, highlighted the importance of assessing the appropriateness of strong recommendations and provided guidance on how to address it.
Yao L, Ahmed MM, Guyatt GH, Yan P, Hui X, Wang Q, Yang K, Tian J, Djulbegovic B. Discordant and inappropriate discordant recommendations in consensus and evidence based guidelines: empirical analysis. BMJ. 2021 Nov 25;375:e066045. doi: 10.1136/bmj-2021-066045. PMID: 34824101; PMCID: PMC8613613.​​
Evidence vs Consensus Guidelines
Inappropriate Strong Recommendations
Some organizations—including the ACC/AHA and ASCO—explicitly classify their guidelines as evidence-based when much of the supporting evidence is deemed to be moderate or high quality, and classify their guidelines as consensus based when it is not.
Our findings suggest when facing low or very low-quality evidence, guidelines should avoid issuing inappropriate discordant recommendations. Abandoning consensus-based guidelines is likely to facilitate this goal.
Yao L, Guyatt GH, Djulbegovic B. Can we trust strong recommendations based on low quality evidence? BMJ. 2021 Nov 25;375:n2833. doi: 10.1136/bmj.n2833. PMID: 34824089; PMCID: PMC8769229.
​
Management of Chronic TMD Pain
To support the guideline development for chronic pain secondary to Temporomandibular Disorders (TMD), I led a team and conducted a complex network meta-analysis with a huge amount of work, involving over 200 RCTs and comparing 59 pharm and non-pharm treatments the same time (see the right figure).
In this network, we were able to propose innovative methodologies including ranking the interventions by considering both statistics and certainty of evidence, considering the MID for interpretation, and calculating the possibility of achieving MID for measuring the outcomes. We finally identified 8 interventions proving effective for patients with TMD chronic pain. Our findings from the network meta-analysis informed 21 recommendations by the guideline panel. This work has become a trusted reference point for decision-makers in TMD
Yao L, Sadeghirad B, Li M, Li J, Wang Q, Crandon HN, Martin G, Morgan R, Florez ID, Hunskaar BS, Wells J, Moradi S, Zhu Y, Ahmed MM, Gao Y, Cao L, Yang K, Tian J, Li J, Zhong L, Couban RJ, Guyatt GH, Agoritsas T, Busse JW. Management of chronic pain secondary to temporomandibular disorders: a systematic review and network meta-analysis of randomised trials. BMJ. 2023 Dec 15;383:e076226. doi: 10.1136/bmj-2023-076226. Erratum in: BMJ. 2024 Jan 30;384:q253. doi: 10.1136/bmj.q253. PMID: 38101924.​
