EditverseCONSORT 2025: what changed and how to bring an RCT manuscript into line
CONSORT 2025 replaces CONSORT 2010 with a 30-item checklist and a new open science section. An item-by-item comparison, and one methods paragraph rewritten to meet it.
Key points
- CONSORT 2025 is a 30-item checklist with a participant flow diagram. Its authors state that it supersedes CONSORT 2010.
- Against CONSORT 2010 the update adds seven items, completely revises three, deletes one and brings in items from the Harms, Outcomes and Non-pharmacological Treatment extensions and from TIDieR.
- A new open science section groups registration, protocol and statistical analysis plan access, data sharing, funding and conflicts of interest as items 2 to 5.
- The items most often absent from a manuscript drafted to the 2010 list are data sharing (4), patient and public involvement (8), how harms were assessed (15), who was analysed and how missing data were handled (21b, 21c), and delivery of the intervention and concomitant care (24a, 24b).
- The item numbers have changed throughout, so a completed 2010 checklist cannot be reused. Fill in the 2025 form from scratch.
What CONSORT 2025 is and what it replaces
The CONSORT 2025 statement was published in April 2025 in the BMJ and co-published in JAMA, The Lancet, Nature Medicine and PLOS Medicine.1 It is a checklist of 30 items and a flow diagram, to be read with a separate explanation and elaboration paper that gives the reasoning and published examples for every item.2
The previous version, CONSORT 2010, had 25 numbered items.3 The 2025 statement says in plain terms that it supersedes the 2010 one. It was updated at the same time as SPIRIT 2025, the guideline for trial protocols, and the wording of the two checklists was aligned so that a protocol and the report of its results describe the same things in the same terms.4
CONSORT 2025 does not prescribe a manuscript structure. The statement asks that each item be addressed somewhere in the article in enough detail, in the format the journal requires, and it recommends submitting a completed checklist that shows where each item is reported.
The changes in one table
The statement summarises its own changes in a box. The table below restates that summary by type of change, with the 2025 item numbers.1
| Type of change | 2025 item | What the item now asks for |
|---|---|---|
| New items (seven) | 4 | Data sharing: where and how de-identified participant data, the data dictionary, statistical code and other materials can be accessed |
| 5b | Financial and other conflicts of interest of the manuscript authors | |
| 8 | How patients or the public were involved in the design, conduct and reporting of the trial | |
| 12b | Where applicable, eligibility criteria for sites and for the individuals delivering the interventions | |
| 15 | How harms were defined and assessed | |
| 21b, 21c | Who is included in each analysis and in which group; how missing data were handled | |
| 24a, 24b | How the intervention and comparator were actually administered; concomitant care received in each group | |
| Completely revised items (three) | 3 | Access to the statistical analysis plan as well as the protocol |
| 10 | Important changes after the trial started, including outcomes or analyses that were not prespecified, with reasons | |
| 26 | For each primary and secondary outcome, by group: number analysed, number with data at the time point, the result, and the effect size with its precision | |
| Deleted item (one) | (2010 item 21) | Generalisability is no longer a separate item. It is now part of trial limitations (item 30) |
| Integrated from extensions | 7, 14, 15, 21a, 23a, 24, 26, 27 | Harms items from CONSORT Harms 2022, outcome items from CONSORT-Outcomes 2022, and intervention delivery items from the Non-pharmacological Treatment extension and TIDieR |
| Restructured | 2 to 5 | New open science section placed straight after the title and abstract |
The integration matters for authors who never used the extensions. Reporting how harms were assessed, defining each outcome down to its measurement variable, analysis metric, method of aggregation and time point, and describing how the intervention was actually delivered were previously asked for in separate documents.5,6,7 They are now in the core checklist that every trial report is checked against.
CONSORT 2010 to CONSORT 2025, item by item
The statement's web appendix 1 holds the authors' own detailed comparison. The table below is our side-by-side reading of the two published checklists, intended as a working map for a manuscript or checklist that was drafted against the 2010 numbering.1,3
| Topic | 2010 | 2025 | What to check in the manuscript |
|---|---|---|---|
| Title and abstract | 1a, 1b | 1a, 1b | Randomised design identified; structured summary |
| Registration | 23 | 2 | Registry name and number, now with the URL and the date of registration |
| Protocol and analysis plan | 24 | 3 | Where both documents can be accessed |
| Data sharing | none | 4 | Access route for data, data dictionary and code |
| Funding and conflicts | 25 | 5a, 5b | Funding and role of funders; authors' conflicts of interest |
| Background and objectives | 2a, 2b | 6, 7 | Objectives now cover both benefits and harms |
| Patient and public involvement | none | 8 | What involvement there was, or a statement that there was none |
| Trial design | 3a | 9 | Type of trial, allocation ratio and framework (superiority, equivalence, non-inferiority, exploratory) |
| Changes after the trial started | 3b, 6b | 10 | One item for changes to methods, outcomes and analyses, with reasons |
| Setting and eligibility | 4b, 4a | 11, 12a, 12b | Setting moved ahead of eligibility; criteria for sites and providers added |
| Intervention and comparator | 5 | 13 | Replicable description; where manuals or other materials can be found |
| Outcomes | 6a | 14 | Variable, metric, aggregation and time point for every prespecified outcome |
| Harms, methods | none | 15 | How harms were defined and whether collection was systematic |
| Sample size | 7a, 7b | 16a, 16b | All assumptions behind the calculation; interim analyses and stopping guidelines |
| Randomisation | 8a, 8b, 9, 10 | 17a, 17b, 18, 19 | Who generated the sequence and how; restriction; concealment; whether enrolling and assigning staff had access to the sequence |
| Blinding | 11a, 11b | 20a, 20b | Who was blinded; how blinding was achieved |
| Statistical methods | 12a, 12b | 21a to 21d | Methods including harms; analysis populations; missing data; additional analyses marked as prespecified or post hoc |
| Participant flow and recruitment | 13a, 13b, 14a, 14b | 22a, 22b, 23a, 23b | Follow-up dates now stated for outcomes of benefit and of harm |
| Delivery of the intervention | part of 5 | 24a, 24b | Who delivered it, adherence, fidelity; concomitant care by group |
| Baseline data | 15 | 25 | Baseline table by group |
| Numbers analysed, outcomes and estimation | 16, 17a, 17b | 26 | Three old items merged into one; absolute and relative effects for binary outcomes |
| Harms, results | 19 | 27 | All harms or unintended events in each group |
| Ancillary analyses | 18 | 28 | Subgroup and sensitivity analyses, prespecified or post hoc |
| Interpretation | 22 | 29 | Consistent with results; benefits weighed against harms |
| Limitations and generalisability | 20, 21 | 30 | Bias, imprecision, generalisability and multiplicity in one item |
One methods paragraph, before and after
The example below is illustrative. The trial, its numbers and its registry entry are invented to show the difference in reporting and do not describe a real study.
Before: adequate against CONSORT 2010 habits
This was a randomised controlled trial in adults with knee osteoarthritis. Patients were randomised 1:1 to supervised exercise or usual care using computer-generated random numbers. The primary outcome was pain at 12 weeks. Adverse events were recorded. A sample of 180 was calculated to give 90% power. Analysis was by intention to treat using a t test. The trial was registered and approved by the ethics committee.
After: written to CONSORT 2025 (illustrative)
This was a two-arm, parallel-group, superiority trial with 1:1 allocation (item 9), conducted in the physiotherapy outpatient departments of three teaching hospitals (item 11). Adults aged 45 years or older with radiographic knee osteoarthritis were eligible (item 12a). Sites had to run at least two supervised classes a week, and the programme was delivered by physiotherapists with at least two years of musculoskeletal practice who had completed a four-hour training session (item 12b). The intervention was a 12-week supervised exercise programme of two 45-minute sessions a week; the comparator was usual care with a written advice leaflet. The session manual is in supplement 2 (item 13). The primary outcome was knee pain on a 0 to 10 numerical rating scale, analysed as change from baseline to week 12 and summarised as the between-group difference in means (item 14). Harms were collected systematically at every visit with a structured checklist of falls, joint swelling and pain flares, and through open questions (item 15). A sample of 180 participants gave 90% power to detect a difference of 1.0 point, assuming a standard deviation of 2.0, two-sided alpha of 0.05 and 10% loss to follow-up (item 16a). The trial statistician, who had no contact with participants, generated the allocation sequence in R using permuted blocks of four and six, stratified by site (items 17a, 17b). Allocation was released through a central web system after baseline assessment, and recruiting staff had no access to the sequence (items 18, 19). Outcome assessors and the analysing statistician were blinded; participants and physiotherapists could not be (item 20a). The primary analysis included all randomised participants in the group to which they were allocated (item 21b), using a linear mixed model adjusted for site and baseline pain (item 21a). Missing outcome data were handled by multiple imputation under a missing-at-random assumption, with a complete-case sensitivity analysis (items 21c, 21d). Two patient partners with knee osteoarthritis reviewed the session schedule and the wording of the outcome questionnaire before recruitment (item 8). The trial was registered before enrolment of the first participant; the registry name, number, URL and date are given at the end of the abstract (item 2). The protocol and statistical analysis plan are in supplement 1 (item 3). After recruitment began, the upper age limit of 80 years was removed to improve accrual; no outcome was changed (item 10).
The second version is longer, and most journals will not accept all of it in the main text. The usual solution is to keep design, eligibility, outcomes, randomisation, blinding and analysis in the Methods and move the manual, the harms checklist and the full sample size reasoning to a supplement that the completed checklist points to.
Bringing an existing draft into line
For a manuscript already written to the 2010 list, the work falls into five steps.
- Build the open science block. Collect the registry record (number, URL, date), the location of the protocol and statistical analysis plan, the data sharing statement, the funding statement with the funder's role, and each author's conflicts of interest. The ICMJE Recommendations, updated January 2026, already require a data sharing statement for trial reports, so items 2 to 5 overlap with what the journal's submission form asks for.8
- Write the harms method. State how harms were defined, who collected them, at which visits and whether collection was systematic. Then check that the harms results (item 27) are reported per group.
- Define the analysis populations and missing data handling. Items 21b and 21c need explicit sentences. The phrase "intention to treat" on its own does not tell a reader who was analysed.
- Describe delivery. Item 24a asks what was actually administered: who delivered it, adherence and fidelity. Item 24b asks for concomitant care in each group. These usually need a short results paragraph and a supplementary table.
- Rebuild the outcome tables. Item 26 asks for the number analysed and the number with available data for every primary and secondary outcome in each group, with effect sizes and their precision, and both absolute and relative effects for binary outcomes.
Patient and public involvement (item 8) cannot be added after the fact. If there was none, say so. The statement asks authors to report what was done and to acknowledge what was not done or was modified.
The abstract needs the same attention as the body. Item 1b asks for a structured summary of design, methods, results and conclusions, and the ICMJE Recommendations ask that trial abstracts carry the items the CONSORT group has identified as essential, with the registration number placed at the end.8,9
Where the trial is being written up from a clinical study report or a statistical output, the same mapping is the first task in our original research manuscript writing service, and a completed CONSORT 2025 checklist with page references is part of the delivery. For a finished draft, medical manuscript editing covers the same check. Both sit within our work on journal manuscripts.
Do journals already require CONSORT 2025?
Policies differ by journal and some may still accept a CONSORT 2010 checklist during a transition period. Read the target journal's instructions for authors before completing the form. Because the 2025 statement supersedes the 2010 one, reporting to the 2025 list is the safer choice where the journal does not specify.
My trial finished before 2025. Do I have to report items such as patient involvement that were never planned?
CONSORT is a reporting guideline. It asks you to describe what happened. Where an activity did not take place, state that it did not, and report every other item in full.
Do the CONSORT extensions still apply?
Yes. The 2025 statement lists extensions for designs such as cluster, crossover, non-inferiority and pragmatic trials, and these continue to be used alongside the main checklist. Check whether the extension that fits your design has itself been updated since 2025.
References
- Hopewell S, Chan AW, Collins GS, Hróbjartsson A, Moher D, Schulz KF, et al. CONSORT 2025 statement: updated guideline for reporting randomised trials. BMJ. 2025;389:e081123. PMID 40228833 · DOI
- Hopewell S, Chan AW, Collins GS, Hróbjartsson A, Moher D, Schulz KF, et al. CONSORT 2025 explanation and elaboration: updated guideline for reporting randomised trials. BMJ. 2025;389:e081124. PMID 40228832 · DOI
- Schulz KF, Altman DG, Moher D; CONSORT Group. CONSORT 2010 statement: updated guidelines for reporting parallel group randomised trials. BMJ. 2010;340:c332. PMID 20332509 · DOI
- Chan AW, Boutron I, Hopewell S, Moher D, Schulz KF, Collins GS, et al. SPIRIT 2025 statement: updated guideline for protocols of randomised trials. BMJ. 2025;389:e081477. PMID 40294953 · DOI
- Junqueira DR, Zorzela L, Golder S, Loke Y, Gagnier JJ, Julious SA, et al. CONSORT Harms 2022 statement, explanation, and elaboration: updated guideline for the reporting of harms in randomised trials. BMJ. 2023;381:e073725. PMID 37094878 · DOI
- Butcher NJ, Monsour A, Mew EJ, Chan AW, Moher D, Mayo-Wilson E, et al. Guidelines for reporting outcomes in trial reports: the CONSORT-Outcomes 2022 extension. JAMA. 2022;328(22):2252-2264. PMID 36511921 · DOI
- Hoffmann TC, Glasziou PP, Boutron I, Milne R, Perera R, Moher D, et al. Better reporting of interventions: template for intervention description and replication (TIDieR) checklist and guide. BMJ. 2014;348:g1687. PMID 24609605 · DOI
- International Committee of Medical Journal Editors. Recommendations for the Conduct, Reporting, Editing, and Publication of Scholarly Work in Medical Journals. Updated January 2026. icmje.org/recommendations
- Hopewell S, Clarke M, Moher D, Wager E, Middleton P, Altman DG, et al. CONSORT for reporting randomised trials in journal and conference abstracts. Lancet. 2008;371(9609):281-3. PMID 18221781 · DOI