**Background:** The Chinese government mandates that Food for Special Medical Purposes (FSMP) designed for specific diseases must undergo clinical trials before registration approval. However, as of November 3, 2022, only 1 of 92 registered FSMP products was categorized as whole-nutrient FSMP (disease specified), highlighting a supply-demand imbalance. Clinical trials for FSMP are characterized by long durations (6–90 days, with common durations of 7, 14, 28, 84, and 90 days) and highly diverse outcome selections. A systematic review by the authors identified 161 outcomes from method sections of enrolled studies, which, combined with 9 indices from CFDA files, resulted in an outcome pool of 170 outcomes. Only 54 (31.8%) outcomes appeared 3 or more times, and only 3 outcomes (fasting blood glucose, postprandial blood glucose, and triglyceride levels) appeared in at least half of the enrolled studies (≥23 times). This diversity leads to research waste and complicates regulatory approval. Core outcome sets (COS) — the minimum sets of outcomes to be measured and reported — offer a solution to standardize outcome selection, improve comparability, and optimize data use. This protocol uses T2DM as an example to develop COS for FSMP clinical trials.
**Methods:** The study follows Core Outcome Set-STAndardised Protocol Items (COS-STAP) and Core Outcome Set-STAndards for Development (COS-STAD) guidelines. The COS development is divided into 3 phases. Phase 1: Generate a list of relevant outcomes from a systematic review of published studies (from 6 medical databases: CNKI, Wanfang, VIP, PubMed, Ovid-Medline, and Cochrane Library), regulatory documentation (from SAMR and other international bodies), and qualitative stakeholder interviews. Outcomes will be categorized into 4 domains: safety, nutritional adequacy, special medical effects, and others. Phase 2: Conduct at least 2 rounds of Delphi surveys among stakeholders (patients, clinical dietitians, physicians, COS researchers, journal editors, FSMP manufacturers, and regulatory representatives). Stakeholders will score each outcome on a 9-point Likert scale (1–3 = not important, 4–6 = important, 7–9 = critical). Consensus definitions are provided: 'consensus in' requires ≥70% of any stakeholder group scoring 7–9 and <15% scoring 1–3; 'consensus out' requires ≥70% scoring 1–3 and <15% scoring 7–9. Phase 3: Hold a face-to-face or online consensus meeting with key stakeholders to finalize the COS. The target sample size is approximately 100 professional participants and 30 patient participants (3:1 ratio), with 15–25 key stakeholders invited to the consensus meeting. Alternative plans are prepared for Likert scale selection (3-point scale if 9-point fails), additional Delphi rounds (if >50% of outcomes remain), and streamlined scoring groups (if subgroups have <30 persons).
**Key Results:** This is a study protocol; no results are reported. The systematic review has been completed, and the study is ongoing at phase 1. The protocol is registered with COMET (registration number 1547, registered March 23, 2020). The study is expected to be completed by the end of 2024.
**Clinical Implications:** The COS for FSMP-T2DM is intended to serve as a multistakeholder agreed tool for regulators in pre- and postmarket management of FSMP. Premarket, it will guide manufacturers in formulation development and serve as an assessment criterion for clinical trial institutions. Regulators will not approve FSMP registration unless clinical trial reports show acceptable COS results. Postmarket, real-world data on COS will help clinicians select the most suitable FSMP for patients. If successful, this approach could be extended to FSMP for other diseases and potentially lead to development of technical specifications, guidelines, or standards for the FSMP industry in China.