Uptake of core outcome sets by clinical trialists in China: a protocol

Background The concept of core outcome sets (COS) has been introduced in China for about 10 years. In recent years, some Chinese researchers also committed to developing COS, though the majority of COS are ongoing. However, there were more than 500 published COS for research in the COMET database by 2020. The extent of availability of COS for the top 25 diseases with the highest burden in China is unknown. In addition, the uptake of COS in clinical trials for these diseases is unknown, along with the knowledge, perceptions, and views of the clinical trialist community in China on the use of COS in relation to choosing outcomes for their research. Methods The main burden of disease in China will be identified. Then we will search the COMET database to identify if there are ongoing or completed relevant COS research A COS published since 2012 would be preferred to one published before 2012 for the analysis of COS uptake if one meets the eligibility criteria. We will extract scopes of published eligible COS, including condition, population, interventions, and core outcomes. Then we will search the Chinese Clinical Trial Registry using disease names for each disease that has a published COS. We will assess the overlap in scope between clinical trials and COS. Then we will conduct an online survey and semi-structured interviews to identify the knowledge and perceptions of COS among primary investigators of included clinical trials. Discussion This research will fill in gaps between COS and the burden of disease in China. Understanding clinical trialists’knowledge and perceptions of COS may help dissemination and application of COS in the future. Trial registration This research is registered in Core Outcome Measures in Effectiveness: https://www.comet-initiative.org/Studies/Details/2563.


Background
A core outcome set (COS) is an agreed standardized set of outcomes that should be measured and reported, as a minimum, in all clinical trials in specific areas of health or health care. 1 The concept of COS was introduced in China in 2013. 2 Chinese researchers have registered more than 80 COS for research in the Core Outcome Measures in Effectiveness Trials (COMET) database by 2022.
A search of the COMET database shows that more than 500 completed COS studies have been published by 2021.The number of COS varies by disease category.The top five disease categories are cancer, rheumatology, neurology, heart and circulation, orthopaedics and trauma. 3However, the number of candidates vary by disease category.In addition, there are more than one COS for some conditions, while no COS exists in some other important conditions.Whether there are gaps between COS research and the burden of disease in China remains unclear.
There are some benefits to using COS in clinical trials, which include improving the consistency of outcome reporting, increasing the relevance of outcomes measured to decision-makers, and helping to identify potential selective outcome reporting bias.However, previous research has shown that COS uptake is low in most research areas. 4,5Research shows that clinical trialists' awareness and understanding of COS may facilitate the use of COS. 6 However, Chinese clinical trialists' awareness, understanding and using of COS are unknown.
The aims of this study are to examine: (1) whether the top 25 diseases with the highest burden in China have relevant COS; (2) the use of COS in clinical trials registered in the Chinese Clinical Trial Registry for these 25 diseases; and (3) views of these clinical trialists on the knowledge, perceptions, and use of COS in relation to choosing outcomes, as well as barriers and facilitators to uptake of COS.

Mapping of burden of disease in China to COS 2.1.1 Identification of diseases for study
The top 25 causes of disability-adjusted life-years (DALY) in China in 2019 have been identified from the tool of Global Burden of Disease (GBD). 7Two researchers (RQ and XF) will also discuss subtypes of disease according to practice guidelines or textbook of internal medicine if necessary.The initial list of diseases for the study is shown in Table 1.REVISED Amendments from Version 2 A few errors have been changed.
Any further responses from the reviewers can be found at the end of the article

Identification of relevant COS
We will manually search the COMET database and identify all COS for diseases/subtypes in the diseases list.We will identify if there are ongoing or completed relevant COS research.The study began on April 7 th , 2023 and is ongoing.

Inclusion criteria for COS
(1) The scope appears relevant to a disease in the diseases list; (2) COS for research should have included patient participants; RQ will identify COS from the COMET database, and XF will check the results.Any discrepancies will be resolved by consensus discussion or by consulting a third reviewer (PW).RQ and PW will discuss all the COS for the 25 causes of DALY and determine which COS are eligible.The criterion related to the inclusion of patient participants follows previous work examining uptake of specific COS, and reflects the impact of patient participation on the choice of core outcomes. 8e aim of COS is to promote a single unified approach to evaluating outcomes in a given condition throughout the world, however only 30% of COS published up to the end of 2021 have included participants from anywhere in Asia, 8 and the majority of COS developed in China are focused on traditional Chinese medicine (TCM).To explore whether the uptake of COS is influenced by whether it was developed by Chinese researchers, we will additionally include a COS published in a Chinese language journal if one meets the inclusion criteria above.

Data extraction
We will describe the number of ongoing and published COS for the highest burden of disease in China.Full texts will be retrieved for published COS only.
Previous COMET systematic reviews, and ongoing inclusion of published COS by COMET, will be used for details of published COS relevant to this work.Data extraction will include the intended use of recommendations, disease category, disease name, population characteristics (age, sex), interventions, geographical locations of COS developers, geographical locations of COS participants will be extracted, and whether patients participated.

Data analysis
Descriptive analysis will be used for general characteristics of COS.

Exploring uptake of COS in trialists in China 2.2.1 Identification of clinical trials in the Chinese Clinical Trial Registry
We will search the Chinese Clinical Trial Registry (https://www.chictr.org.cn/historyversionpub.aspx)according to the disease name extracted for the published COS.

Inclusion and exclusion criteria for clinical trials
Inclusion criteria for clinical trials: (1) Randomised and non-randomised clinical trials; (2) The objectives of clinical trials are for disease therapy; (3) The prospective registration time should be in 2021 and 2022.
Exclusion criteria for clinical trials: The clinical trials are to assess the mechanisms of effect of interventions.
If the disease name in the COS is broader, we will further search subtypes of disease.RQ and XF will discuss the list of subtypes of disease, any disagreement will be discussed with a third investigator (HS).XF, WW, ZC, and SL will search for clinical trials from the Chinese Clinical Trial Registry and apply the inclusion and exclusion criteria to identify the eligible clinical trials.RQ will check the results.

Comparison of COS with clinical trials
Inclusion criteria: (1) An eligible COS exists published since 2012.
(2) The number of clinical trials with matching scope to an eligible COS is not less than 40.
A COS published since 2012 would be preferred to one published before 2012 for the analysis of COS uptake if one meets the eligibility criteria.A COS developed by Chinese researchers may be significant for Chinese clinical trialists.If there is an eligible COS developed by Chinese researchers, it will be included.If it does not meet the inclusion criteria for comparison of COS with clinical trials, then the prospective registration time for clinical trials will be expanded.
An online survey will be sent to trialists who registered clinical trials which are relevant to a specific COS.We anticipate a 50% non-response rate, thus if the number of trials is too small, it will be difficult to obtain a reasonable spread of perspectives from trialists about the use of the relevant COS.It was felt that responses from 20 trialists would be helpful based on a typical number of participants involved in qualitative research, thus requiring an anticipated 40 registered trials.

Data extraction and analysis
Data extraction for COS will include: author, title, publication time, condition, population, intervention, outcomes, and outcome measurement instruments (if applicable).
Contact details of trialists (including applicant's name, the registration contact person's email and telephone number, the primary investigator's name, the primary investigator's email and telephone number) will be extracted, because they will be invited to participate in the online survey.The other data extracted will include: title, the primary investigator's institutions, study type, disease name, population, intervention, outcomes, and outcome measurement instruments.
For comparing trial outcomes and COS, only published COS will be used.If there are more than two COS for a specific disease, the differences in scope between these COS will be discussed by two reviewers.
We will compare the overlap in scope between clinical trials and COS according to the framework used in previous research. 9The framework is shown in Table 2.
We will calculate the median percentages of outcomes in each clinical trial that were specific matches, general matches, and non-matches with outcomes in each relevant COS.We will calculate the proportion of outcomes that overlap between the COS and clinical trials for each disease.A survey will be sent to clinical trialists who have registered or completed a clinical trial that has overlapped scope between the trial and COS.We will identify the primary investigator's name, email, and telephone number when we extract eligible clinical trials from the Chinese Clinical Trial Registry.We will send a Chinese translation of the COS deemed relevant to the trialists' trials with questionnaire.When the primary investigators cannot be contacted via emails, the applicants for registration will be approached.We will examine trialists' knowledge and awareness of COS.An online survey will be sent to clinical trialists, with one of three versions of the survey: 1) clinical trialists who report a full COS in their clinical trials; 2) clinical trialists who report only a few of outcomes in the COS; and 3) clinical trialists who do not report any outcome in the COS.
The survey will include both closed and open-ended questions.General information, such as gender, age, participants' role (clinicians or researcher), specialty (traditional Chinese medicine, integrative medicine, or Western medicine), work experience, geographical area, professional qualification, and educational background will be collected.We will refer to the facilitators and barriers identified in previous research, including using of COS and clinical practice guidelines, 10,11 to develop open and closed questions to ask all participants.MK, PW, and RQ will discuss the questions in the questionnaires.The survey and other tools will be distributed into Chinese.RQ will translate tools into Chinese and the translation will then be sent to several researchers who have experience in COS research from a Chinese authors' team and we will ask them to give comments.RQ, SL, ZC, and WW will discuss and determine the translation before distribution.
The structured questions in the survey are shown in extended data.
At the end of each survey, we will ask the trialists if they would like to attend a semi-structured interview.

Data analysis
Closed questions will be analyzed using descriptive statistics.Ratings for the barriers and facilitators to uptake will be summarized in terms of the median, range, and interquartile range.For open questions, thematic analysis will be used.All free text will be extracted in to a word document which RQ will read multiple times to organize the text into some initial themes and the author team will review the data and its assignment to the themes identified.The reason why the trialists used or did not use COS will be analyzed by content analysis.New suggested potential barriers and facilitators will be mapped to different themes.

Semi-structured interviews 2.3.2.1 The process of semi-structured interviews
We will invite trialists who agree to be interviewed to attend an in-person or online interview, if saturation is reached, the interviews will be completed.The COS publication, translated into Chinese, will be sent to them before the interview.The topic guide is as follows: 1. How many years of experience do you have in the design, management, or analysis clinical trials?
2. How many clinical trials did you participate in the design, management, or analysis?

Data analysis
General characteristics of interviewees will be presented using descriptive methods.The trialist's awareness of, and decisions to search for and use a COS will be described, and the reasons why trialists used or did not use COS will be analyzed by content analysis.

Dissemination
The finding of this research will be disseminated through conferences, peer-reviewed publication, and social media.

Discussion
COS have been introduced in China for about 10 years.Chinese trialists' knowledge and awareness of using COS are important to COS developers in China.This study will provide information on the proportion of trialists in China who registered clinical trials in the Chinese Clinical Trial Registry in the areas with the highest burden of disease, and how the outcomes they chose compare to existing COS.
COS are very important for reducing heterogeneity of outcome reporting, potential bias, and research waste when they are used in clinical trials.There are hundreds of COS published in the world.Chinese researchers are also developing COS in some health areas, whether there is a gap between COS and the burden of disease in China remains unclear.This research will provide evidence relevant to COS researchers and clinical trialists.The authors state that the two aims of this protocol are to: Determine whether ongoing and completed COS are relevant to the disease burden in China.

1.
Find out whether clinical trialists in China use published COS in their trials.2.
They have identified the top 25 disease categories in China, based on the number of disabilityadjusted life years (DALYs).They plan to search the COMET database to find ongoing and published Core Outcome Sets (COS) that have been added since 2012, that are relevant to these 25 disease categories.They will also search the Chinese Clinical Trial Registry to identify clinical trials (registered 2021-2022) in the 25 disease categories that list outcomes that match the outcomes in the diseasespecific COS.This will give an indication of whether ongoing and completed COS are relevant to the disease burden in China.
For further analysis, individual clinical trials will be analysed to see whether an appropriate COS exists for the disease in question.Each trial also has to meet the inclusion criteria of there being at least 40 clinical trials that match to each specific COS.Lead investigators of the selected clinical trials will then be surveyed, and some will be interviewed, on their views of COS.This is an ambitious study that will provide an interesting insight into COS uptake in China.It is actually three studies in one: Mapping of burden of disease in China to published COS. 1.
Exploring uptake of COS in trials registered on the Chinese clinical trials registry.

2.
Mixed methods study to establish awareness, barriers and facilitators to COS uptake (including both an initial survey and subsequent interview study). 3.

1.
It might have been helpful to describe these three aspects separately, with specific aims, objectives and methods for each.Not all of the studies aims and objectives are clearly articulated in the existing two aims.
The authors suggest that specific COS may be required to evaluate interventions testing Chinese Medicine and that uptake of COS developed in China might be higher than uptake of COS developed internationally.I am a little worried that such a stance goes against the very essence of what COS initiatives are trying to do.Namely to promote a single unified approach to evaluating outcomes in a given condition throughout the world.I am not yet convinced that a COS specific to Chinese Medicine is helpful or required.
The approach of focussing on the top 25 diseases by burden of disease in China is interesting and should help to focus the study, although this still represents a lot of work.It is not clear why only COS published since 2012 are included.If a disease category has a gold-standard COS that was published pre-2012 (e.g the work of the OMERACT group), this protocol would count the disease category as having no COS, which seems a shame.
It is proposed that the COS identified through a search of the COMET database will be evaluated for quality.Whilst this is a laudable ambition, there was very little detail provided as to exactly how this would be done and what the cut-off for meeting the quality standard might be.Rather than excluding COS based on their quality status, I wonder if it would be more useful to include all identified COS, and to report the quality ratings descriptively.At least this information would then be in the public domain and could avoid wasted effort of people repeating these quality assessments.One of the specified inclusion criteria for study 2 (uptake of COS in clinical trials and studies) required that there be "not less than 40 trials for a given condition".It might be helpful to add an explanation as to why this was included.If trials are only included in cases where at least 40 of them match a specific COS, the authors are not capturing up to 39 trials per disease category.
Similarly, in the section describing data extraction, it might be helpful to signpost to the fact that study 3 will require the contact details of clinical trialist who have been identified in study 2. Otherwise, the reader is left wondering why this information was being extracted from the registry data.
In the data extraction section, it states that 'data will be retrieved from previous COMET systematic reviews' but it is not clear how this information will be used in the protocol.
We have suggested a few minor wording amendments to improve readability: Background ○ Suggest amend to: "There are some benefits to using COS in clinical trials, which include improving the consistency of outcome reporting, increasing the relevance of outcomes measured to decision-makers, and helping to identify potential selective outcome reporting bias."articulated in the existing two aims.
Author response 1.The three aims were listed in the final paragraph of the Background section, and the methods for each are described in separate subsections of the Methods section.

Reviewer comment 4.
It is proposed that the COS identified through a search of the COMET database will be evaluated for quality.Whilst this is a laudable ambition, there was very little detail provided as to exactly how this would be done and what the cut-off for meeting the quality standard might be.Rather than excluding COS based on their quality status, I wonder if it would be more useful to include all identified COS, and to report the quality ratings descriptively.At least this information would then be in the public domain and could avoid wasted effort of people repeating these quality assessments.
Author response 4. We have clarified that we will use patient participation in COS development as a marker of methodological quality, with the rationale provided.
Reviewer comment 5.One of the specified inclusion criteria for study 2 (uptake of COS in clinical trials and studies) required that there be "not less than 40 trials for a given condition".It might be helpful to add an explanation as to why this was included.If trials are only included in cases where at least 40 of them match a specific COS, the authors are not capturing up to 39 trials per disease category.
Author response 5.The online survey will be sent to trialists who registered clinical trials which are relevant to a specific COS.We anticipate a 50% non-response rate, thus if the number of trials is too small, it will be difficult to obtain a reasonable spread of perspectives from trialists about the use of the relevant COS.It was felt that responses from 20 trialists would be helpful based on a typical number of participants involved in qualitative research, thus requiring an anticipated 40 trials This rationale has been explained in the methods (2.2.3).
Reviewer comment 6.Similarly, in the section describing data extraction, it might be helpful to signpost to the fact that study 3 will require the contact details of clinical trialist who have been identified in study 2. Otherwise, the reader is left wondering why this information was being extracted from the registry data.
Author response 6.This has been explained in 2.2.4.

Reviewer comment 7.
In the data extraction section, it states that 'data will be retrieved from previous COMET systematic reviews' but it is not clear how this information will be used in the protocol.
Author response 7. Previous COMET systematic reviews, and ongoing inclusion of published COS by COMET, will be used for details of published COS relevant to this work.Data extraction will include the intended use of recommendations, disease category, disease name, population characteristics (age, sex), interventions, geographical locations of COS developers, geographical locations of COS participants will be extracted, and whether patients participated.This has been explained in 2.1.4.

Reviewer comment 8.
We have suggested a few minor wording amendments to improve readability: Background Suggest amend to: "There are some benefits to using COS in clinical trials, which include improving the consistency of outcome reporting, increasing the relevance of outcomes measured to decision-makers, and helping to identify potential selective outcome reporting bias."Methods Confusing typo in survey section says 'A survey will be sent to clinical trialists who will conduct a clinical trial' -should this be 'A survey will be sent to clinical trialists who have conducted a clinical trial'?
Author response 8.We have made these changes in the manuscript.
Reviewer comment 9.In the survey, there is a question about whether the respondent is a working in Traditional Chinese Medicine, Integrative Medicine or Western Medicine.Integrative Medicine is not mentioned anywhere in the protocol.
Author response 9.The text has been amended.

Competing Interests:
No competing interests were disclosed.
Reviewer Report 30 November 2023 In addition to the literature review, the study also designed a survey for trial investigators, which is a commendable survey to understand further the current situation of knowledge and application of COS among clinical trial investigators in China, which is also conducive to the development and promotion of COS.
From the structured questions in the survey, it should be possible to analyze why the investigators used or did not use COS, so it is recommended to consider this analysis in the statistical analysis section.
Is the rationale for, and objectives of, the study clearly described?Yes

Are sufficient details of the methods provided to allow replication by others? Yes
Are the datasets clearly presented in a useable and accessible format?

Not applicable
Competing Interests: No competing interests were disclosed.
Reviewer Expertise: Evidence-based medicine research in traditional medicine; clinical epidemiology; COS development I confirm that I have read this submission and believe that I have an appropriate level of expertise to confirm that it is of an acceptable scientific standard.

©
2024 Thomas K et al.This is an open access peer review report distributed under the terms of the Creative Commons Attribution License, which permits unrestricted use, distribution, and reproduction in any medium, provided the original work is properly cited.Kim Thomas 1 Centre of Evidence Based Dermatology, School of Medicine, University of Nottingham, Nottingham, UK 2 Centre of Evidence Based Dermatology, School of Medicine, University of Nottingham, Nottingham, UK 3 Centre of Evidence Based Dermatology, School of Medicine, University of Nottingham, Nottingham, UK Emma Campbell 1 Centre of Evidence Based Dermatology, School of Medicine, University of Nottingham, Nottingham, UK 2 Centre of Evidence Based Dermatology, School of Medicine, University of Nottingham, Nottingham, UK 3 Centre of Evidence Based Dermatology, School of Medicine, University of Nottingham, Nottingham, UK Uptake of core outcome sets by clinical trialists in China: a protocol Thank you for asking for a review of this interesting study protocol.Overall, the protocol is well written.

Table 1 .
The initial list of diseases for the study (ranked by number of disability-adjusted life-years).

Table 2 .
The framework for assessing overlap in scope between clinical trials and COS.
comment 2. The authors suggest that specific COS may be required to evaluate interventions testing Chinese Medicine and that uptake of COS developed in China might be higher than uptake of COS developed internationally.I am a little worried that such a stance goes against the very essence of what COS initiatives are trying to do.Namely to promote a single unified approach to evaluating outcomes in a given condition throughout the world.I am not yet convinced that a COS specific to Chinese Medicine is helpful or required.Author response 2. We did not intend to suggest that, and have included the rationale for including a COS developed by Chinese researchers in the protocol.The approach of focussing on the top 25 diseases by burden of disease in China is interesting and should help to focus the study, although this still represents a lot of work.It is not clear why only COS published since 2012 are included.If a disease category has a gold-standard COS that was published pre-2012 (e.g the work of the OMERACT group), this protocol would count the disease category as having no COS, which seems a shame.Author response 3.All COS will be analyzed in the mapping of burden of disease in China to COS.A COS published since 2012 would be preferred to one published before 2012 for the analysis of COS uptake if one meets the eligibility criteria.It has been clarified in the methods.