- Email: [email protected]
Trial registration in pediatric surgery trials
Journal of Pediatric Surgery xxx (2017) xxx–xxx
Contents lists available at ScienceDirect
Journal of Pediatric Surgery journal homepage: www.elsevier.com/locate/jpedsurg
Trial registration in pediatric surgery trials☆,☆☆ Sana Rokhsefat a, Deanna E. Morra a, Martin Offringa a,b, Lisa M. Askie c,d, Lauren E. Kelly a,e,⁎ a
Child Health Evaluative Sciences, The Hospital for Sick Children, Toronto, Canada University of Toronto, Faculty of Medicine, Toronto, Canada c University of Sydney, Sydney Medical School, Sydney, Australia d University of Sydney, NHMRC Clinical Trials Centre, Sydney, Australia e Seneca, School of Biological Sciences and Applied Chemistry, Toronto, Canada b
a r t i c l e
i n f o
Article history: Received 4 May 2017 Received in revised form 29 August 2017 Accepted 16 October 2017 Available online xxxx Key words: Trial registration Outcome measures Pediatric surgery trials
a b s t r a c t Background: Prospective clinical trial registration serves to increase transparency and to mitigate selective reporting bias. An assessment of adult surgical trials revealed poor trial registration practice with incomplete provision of information in registries and inconsistent information in the corresponding publication. The extent and completeness of pediatric surgical trial registration are unknown. We aimed to determine the proportion and adequacy of clinical trial registration in pediatric surgery trials published in 2014. Methods: Using sensitive search strategies in MEDLINE, abstracts and full-texts of prospective pediatric intervention studies published in 2014 were screened in duplicate. Pediatric surgical trials were included. Clinical trial registration numbers obtained from publications were searched in trial registries. Data were extracted based on WHO 20-item minimum data set to determine the completeness of registration data. The proportion of registered trials was recorded and registration data were compared to reported data in the corresponding publication. Results: Our search and abstract screening identified 3375 articles for full text review. Following coding, a total of 54 pediatric surgical trials were included and analyzed; 28% (15/54) of which published a registration number. In trials which reported a registration number, 40% (6/15) were retrospectively registered and 40% (6/15) had made changes to their registered primary and/or secondary outcome measures. One included published trial reported an incorrect registration number. Conclusions: Analysis of pediatric surgery trials published in 2014 revealed a poor prospective trial registration rate and incomplete registration data. Our study supports future initiatives for improved registration behaviors in pediatric surgery trials to ensure high-quality, transparent, reproducible evidence is generated. Study type: Therapeutic (clinical trials), level II. © 2017 Elsevier Inc. All rights reserved.
Pediatric surgical practice must be guided by a reliable evidencebase including randomized, comparative effectiveness trials. Reporting bias in clinical trials, including publication bias (preferential publication of positive studies) and selective reporting bias, can distort the validity of medical literature and undermine clinical decision-making [1,2]. Prospective trial registries emerged as a strategy to deter selective reporting by increasing transparency and promoting accountability
Abbreviations: ICMJE, International Committee of Medical Journal Editors; StaR Child Health, Standards for Research in Child Health; WHO, World Health Organization. ☆ Funding: This research did not receive any specific grant from funding agencies in the public, commercial, or not-for-profit sectors. ☆☆ Disclosure: Dr. Lisa M. Askie is the Manager of the Australian New Zealand Clinical Trials Registry; the other authors have no conflicts of interest relevant to this article to disclose. ⁎ Corresponding author at: The Hospital for Sick Children, Toronto, Canada, M5G 0A4. Tel.: +1 416 813 7654x301025. E-mail addresses: [email protected], [email protected] (L.E. Kelly).
through public cataloguing of clinical trials [3] prior to their commencement. The International Committee of Medical Journal Editors (ICMJE) implemented a policy in 2005, mandating registration of clinical trials as a prerequisite for publication in their respective journals which was adopted by eleven peer-reviewed scientific journals [3]. Their intent was to reduce reporting bias, prevent unnecessary duplication and provide an information portal for patients and the public [4,5]. The effectiveness of prospective trial registration, particularly in its role in increasing transparency and accountability, relies on the provision of complete and accurate information [6,7]. In the past, the completeness of information provided in prospective trial registries has been variable [6] and reporting bias remains an ongoing challenge [8]. The assessment of information provided in trial registries has been predominantly focused on adult trials and the completeness of pediatric surgical trial registration has yet to be investigated. The purpose of this paper is to investigate the proportion of pediatric surgical trials that have been registered and to compare information recorded in
https://doi.org/10.1016/j.jpedsurg.2017.10.049 0022-3468/© 2017 Elsevier Inc. All rights reserved.
Please cite this article as: Rokhsefat S, et al, Trial registration in pediatric surgery trials, J Pediatr Surg (2017), https://doi.org/10.1016/ j.jpedsurg.2017.10.049
2
S. Rokhsefat et al. / Journal of Pediatric Surgery xxx (2017) xxx–xxx
clinical trial registries with the information reported in corresponding publications. The objectives of this study are to: (1) assess the proportion of registered pediatric surgical trials and the completeness of information provided in trial registries based on the WHO 20-Item Minimum Data Set; and (2) determine if information reported in clinical trial registries matches the information presented in the corresponding publication.
than 21 years were excluded. Data from each published full-text article were collected and managed using REDCap a secure, web-based application designed to support data capture, hosted at The Hospital for Sick Children [9]. Data were collected by six independent reviewers, with a sample completed in duplicate to assess consistency of data collection. Random spot checks for consistency between reviewers were completed weekly by the Senior Project Manager.
1. Methods
1.2. Data extraction and analysis
1.1. Study selection and search strategy
The proportion of registered trials, and the completeness of registered data versus reported data in a corresponding publication were determined. Using the WHO Trial Registration Data Set (Appendix 2) [10], two investigators (SR, DM) independently manually extracted all data into an Excel (Microsoft, Redmond, United States of America) database. The WHO Trial Registration Data Set was chosen as it is the internationally agreed standard for the minimum information required for a trial to be considered fully registered [10]. Completeness of WHO Trial Registration Data Set was classified as a) did not meet criteria, b) partially met criteria or c) fully met criteria. “Did not meet criteria” was defined as registered information that did not match reported published
A MEDLINE search was conducted using a sensitive and specific search strategy (Appendix 1) restricted to 2014 publications (Fig. 1). The search strategy employed MeSH and non-MeSH terms that are relevant to pediatrics, clinical trials and study protocols. Publications of abstracts and titles were retrieved and screened independently by two reviewers in Endnote. Abstracts pertaining to prospective, interventional trials in children were selected for full-text assessment, and relevant full-text publications were reviewed. Studies without an intervention, without patients younger than 18 or with a mean age greater
Fig. 1. Modified PRISMA [23] flow diagram of prospective pediatric surgical trials published in 2014.
Please cite this article as: Rokhsefat S, et al, Trial registration in pediatric surgery trials, J Pediatr Surg (2017), https://doi.org/10.1016/ j.jpedsurg.2017.10.049
S. Rokhsefat et al. / Journal of Pediatric Surgery xxx (2017) xxx–xxx
information and/or was not clear. For example, if the registered primary outcome was pain but the publication did not report any pain measures. “Partially met criteria” was defined as trials that partially reported or published data items. For example, if an outcome measurement tool was planned to be assessed at 6 and 12 months but only 12-month outcome data were reported in the publication this would constitute a partial criteria match. “Fully met criteria” was defined as trials that fully matched reported (published) with registered (recorded) data items. Retrospective registration was defined as studies which recruited patients before the recorded date of complete trial registration. Disagreements regarding data entries were resolved through discussion with the senior author. If the authors referred to an additional publication which included study details, this information was retrieved and considered as reported (published) information. Journals following the ICMJE recommendations were identified on their website [11] and classified as mandating clinical trial registration. To avoid confusion with registered and nonregistered trials, all information included in clinical trial registries will be referred to as recorded while data obtained from publications will be termed reported. The analysis of completeness of reporting and selection bias is descriptive and results are presented as totals and percentage of all included trials. 2. Results Fifty-four prospective pediatric surgical trials published in 2014 were identified (Fig. 1) and included in this analysis. Descriptive and methodological features included in trials are reported in Table 1. There were no disagreements which required third party review with the senior author. Only 28% (15/54) of trials reported a registration number. There were 17 trials published in journals which mandate registration; 25.6% (10/39) of trials that did not report a registration number were published in a journal that mandates registration. The demographics of included studies are described in Table 1. The majority of trials that reported a registration number were registered with ClinicalTrials.gov; 60% (9/15) (Table 2). Of the registered trials, 40% (6/15) were retrospectively registered and 40% (6/15) made changes to their registered primary and/or secondary outcome measures (Table 2) after registration. Discrepancies between information recorded in the registration record and information reported in a corresponding journal article are found in Table 3. The majority 93.3% (14/15), of trials reported accurate registration numbers; one Table 1 Characteristics of published clinical trials reporting a Clinical Trial Registration Number versus those without a reported Registration Number.
Continent of Recruitment Africa Asia Australia/New Zealand Europe North America South America Journal Registration Mandated Registration Not Mandated Comparison Group Placebo Active Control Standard of Care None Historical Control No treatment control Blinded study design Randomized
Registered No. (%) of Trials (n = 15)
Nonregistered No. (%) of Trials (n = 39)
0 (0) 2 (13.3) 1 (6.7) 6 (40) 4 (26.7) 3 (20)
2 (5.1) 20 (51.3) 1 (2.6) 7 (17.9) 8 (20.5) 1 (2.6)
7 (46.7) 8 (53.3)
10 (25.6) 29 (74.3)
1 (6.7) 9 (60) 4 (26.7) 0 (0) 0 (0) 0 (0) 6 (40) 13 (86.7)
0 (0) 14 (35.9) 10 (25.6) 10 (25.6) 1 (2.6) 4 (10.2) 8 (20.5) 16 (41.0)
○ 3 of the registered trials recruited participants from more than one country. ○ 1 unregistered trial recruited participants from more than one country.
3
Table 2 Clinical trial registry, retrospective registration and changes. Clinical Trial Registry
Registered vs. Reported Data: No. (%) of Trials (N = 15)
ClinicalTrials.gov (NCT) The International ISRCTN Netherlands Trial Registry (NTR) Brazilian Clinical Trials Registry (ReBec) Clinical Trials Registry—India (CTRI) Retrospective Registration Retrospective changes to primary and secondary outcome
9 (60) 2 (13.3) 2 (13.3) 1 (6.7) 1 (6.7) 6 (40) 6 (40)
trial reported an incorrect registration number. With respect to outcome reporting, only 20% (3/15) of trials had registered a primary outcome that matched the primary outcome reported in the journal article (Table 3). Most included pediatric surgery trials, 40% (6/15), recorded primary outcomes which partially matched the publication and 40% (6/15) did not have a registered primary outcome that matched the reported outcome. Examination of primary outcome reporting discrepancies (Table 4) revealed that 13.3% (2/15) of trials published a primary outcome described as secondary outcome in the registry, and 13.3% (2/15) of trials had a registered primary outcome reported as a secondary outcome in the corresponding publication. One trial published a primary outcome not recorded in the registry, another had a registered primary outcome which was not reported in the publication. Discrepancies in secondary outcome reporting revealed 33.3% (5/15) of the registered trials recorded a secondary outcome that did not match the reported secondary outcome (Table 4). The majority of trials, 60% (9/15) registered a secondary outcome that partially matched the secondary outcome published. One trial reported a secondary outcome that was not registered (Table 4). One trial published a secondary outcome described as primary outcome in registry and another trial registered a secondary outcome which was reported as primary outcome in the publication (Table 4). 3. Discussion Over a decade has passed since the International Committee of Medical Journal Editors (ICMJE) mandated registration as prerequisite for publication of clinical trials [3] however the proportion of publications reporting a registration number and the provision of complete registration data remains variable [5]. To our knowledge, this study is the first to compare data recorded in clinical trial registries with the corresponding publication in pediatric surgery trials. The vulnerability of children coupled with the potential of serious adverse consequences associated with pediatric surgeries such as permanent disability and/or death makes complete and timely trial registration and updates a particularly important ethical requirement. Surgical education has historically been guided by apprenticeship style teaching engrained in institutional practice [4]. Complete trial registration data ensure that those involved in health care decision-making have access to publically available information on ongoing and completed studies. Analysis of pediatric surgical trials published in 2014 revealed that only 28% of trials reported a trial registration number. Our findings are similar to an analysis of surgical RCTs published in 2009 and 2010 from 10 high-impact factor surgical journals, in which 44% (111/246) of included studies were registered [12]. This is concerning as most high-impact surgical journals have adopted the ICMJE mandate for trial registration [11]. Registration before enrollment of the first participant is a scientific, moral and ethical responsibility in adherence with standards for quality clinical trials and transparency. Conversely, in a previous report on 698 clinical trials in 5 general medical journals which mandate registration, 89% of trials that began in 2008 were registered appropriately [13]. In our study 25.6% (10/39) of pediatric
Please cite this article as: Rokhsefat S, et al, Trial registration in pediatric surgery trials, J Pediatr Surg (2017), https://doi.org/10.1016/ j.jpedsurg.2017.10.049
4
S. Rokhsefat et al. / Journal of Pediatric Surgery xxx (2017) xxx–xxx
Table 3 Discrepancies between registered and reported data. WHO Dataset Item Item 1: Primary Registry and Trial Identifying Number
Item 2: Date of Registration in Primary Registry
Item 3: Secondary Identifying Numbers
Item 4: Source(s) of Monetary or Material Support
Item 5: Primary Sponsor
Item 6: Secondary Sponsor(s)
Item 7: Contact for Public Queries
Item 8: Contact for Scientific Queries
Item 9: Public Title
Item 10: Scientific Title
Item 11: Countries of Recruitment
Item 12: Condition
Item 13: Intervention(s)
Item 14: Key inclusion or exclusion criteria
Item 15: Study type
Item 16: Date of First Enrollment
Item 17: Target Sample Size
Item 18: Recruitment Status
Item 19: Primary outcome
Item 20: Secondary outcome(s)
surgical trials were published in journals which mandate registration but without a registration number listed in the manuscript. These data highlight variability in enforcement and accountability by journals with regard to implementing a trial registration mandate. Although trial registration is required by the major Journals, there is limited accountability for maintaining and monitoring the completeness of registration. Ideally, trial data provided in registries should be cross-referenced with data analysis plans during manuscript preparation prior to publication, mitigating selective outcome reporting bias. Any unreported outcomes or changes to registered outcomes should be documented in the final publication with justification for the
Registered vs. Reported Data: No. (%) of Trials (N = 15) Did not meet criteria Partially met criteria Fully met criteria Did not meet criteria Partially met criteria Fully met criteria Did not meet criteria Partially met criteria Fully met criteria Did not meet criteria Partially met criteria Fully met criteria Did not meet criteria Partially met criteria Fully met criteria Did not meet criteria Partially met criteria Fully met criteria Did not match Partially matched Fully matched Did not meet criteria Partially met criteria Fully met criteria Did not meet criteria Partially met criteria Fully met criteria Did not meet criteria Partially met criteria Fully met criteria Did not meet criteria Partially met criteria Fully met criteria Did not meet criteria Partially met criteria Fully met criteria Did not meet criteria Partially met criteria Fully met criteria Did not meet criteria Partially met criteria Fully met criteria Did not meet criteria Partially met criteria Fully met criteria Did not meet criteria Partially met criteria Fully met criteria Did not meet criteria Partially met criteria Fully met criteria Did not meet criteria Partially met criteria Fully met criteria Did not meet criteria Partially met criteria Fully met criteria Did not meet criteria Partially met criteria Fully met criteria
1 (6.7) 0 (0) 14 (100) 14 (93.3) 0 (0) 1 (6.7) 11 (73.3) 1 (6.7) 3 (20) 11 (73.3) 0 (0) 4 (26.7) 7 (46.6) 0 (0) 8 (53.3) 14 (93.3) 0 (0) 1 (6.7) 15 (100) 0 (0) 0 (0) 9 (60) 0 (0.0) 6 (40) 15 (100) 0 (0) 0 (0) 4 (26.7) 10 (66.7) 1 (6.7) 3 (20) 0 (0) 12 (80) 0 (0) 5 (33.3) 10 (66.7) 0 (0) 1 (6.7) 14 (93.3) 1 (6.7) 10 (66.7) 4 (26.7) 0 (0) 6 (40) 9 (60) 8 (53.3) 4 (26.7) 3 (20) 15 (100) 0 (0) 0 (0) 7 (46.6) 0 (0) 8 (53.3) 6 (40) 6 (40) 3 (20) 5 (33.3) 9 (60) 1 (6.7)
registered protocol deviation. Trial data should be published even in the absence of registration as an ethical commitment to the study participants. These trials could include a statement justifying the timing or absence of trial registration. An inherent challenge with trial registration is determining who must provide the dedicated resources for checking compliance. A potential solution is to have key governing bodies, such as, ethics committees, institutions and funders also involved in the enforcement of prospective trial registration owing to their key role early in the design of trials [12]. As ethics committees review protocols before recruitment has begun, requiring trial registration before approving trials is a key step in
Please cite this article as: Rokhsefat S, et al, Trial registration in pediatric surgery trials, J Pediatr Surg (2017), https://doi.org/10.1016/ j.jpedsurg.2017.10.049
S. Rokhsefat et al. / Journal of Pediatric Surgery xxx (2017) xxx–xxx
5
Table 4 Discrepancies in primary outcome reporting in registered information versus publication. Discrepancy
Primary Outcome No. (%) of Trials (N = 15)
Secondary Outcome(s) No. (%) of Trials (N = 15)
Reported outcome not in registry Partially reported outcome Published outcome described as secondary outcome in registry Registered outcome reported as secondary outcome in text Registered outcome not reported Nonregistered follow-up outcomes No secondary outcomes Reported outcomes matched the registry
1 (6.7) 6 (40) 2 (13.3) 2 (13.3) 1 (6.7) 3 (20) – 3 (20)
1 (6.7) 8 (53.3) 1 (6.7) 1 (6.7) 3 (20) 2 (13.3) 1 (6.7) 1 (1)
increasing registration compliance. A comprehensive approach is required by all stakeholders to improve the rates of prospective trial registration. The information provided in registries compared to the corresponding journal article was often incomplete. One value of trial registration is that public cataloguing of information offers the opportunity for journal editors to identify and prevent selective reporting bias [11] but this is dependent on the provision of complete trial registration data. Providing complete information in registries should no longer remain voluntary, as registry data are of limited use when registry entries simply state “Not provided.” Software advances now present a feasible solution, and can include autolinking of large international registries with the corresponding publications, allowing for improved interpretation. A recent review identified that automated linkage between registration and corresponding publications has not increased over time [15]. Ensuring data interfaces are user-friendly may encourage investigators to input more than the minimum required information. Mandated linkage of publications directly to the registration may provide incentives for updating and maintaining registry information. The ANZCTR (Australian and New Zealand Clinical Trial Registry) recently added a feature to link a full protocol to a registration record [16]. Guidelines for investigators with respect to when to upload trial results (without violating manuscript publishing) may facilitate timely registry data updates. Improvements to clinical trial registry operations such as these may improve registry behaviors. Trial registration may be viewed as an administrative duty that can be completed at any time. If so, this limits the value of prospective trial registration [14]. In our study, 40% of trials that reported a registration number were retrospectively registered. This is similar to retrospective trial registration rates reported by others [7,14]. While there is a slight possibility that researchers may still not be aware of trial registration or not realize their study meets the definition for trial registration [16], retrospective registration may be problematic for several reasons including registration only to satisfy journal requirements or by allowing changes to the outcome measured or reported after reviewing the data. Outcome measures that are intended but not reported [17,18] may be related to a lack of statistical significance [19] or potentially related to inappropriate outcome selection prior to the initiation of the trial. A study which examined the effects of selective outcome reporting on meta-analysis showed that half of trials relevant to a systematic review did not contribute to the important patient outcomes of the meta-analysis [20]. Interestingly, unconventional methods have been introduced to address selective outcome reporting. In the United Kingdom, the COMPare group is actively monitoring and reporting discrepancies in trials in an effort to increase accountability, with variable success [16,21]. The United States Department of Health and Human Services recently proposed tightening rules for ClinicalTrials.gov data submission, requiring all data to be reported regardless of FDA approval [22]. A limitation of this paper is the small sample size and the inability to evaluate data included in future publications. For example, if long-term outcome assessment was recorded in the clinical trial registry but no information was provided on such a planned future analysis in the
publication, this would have been considered a partial match. The low number of included trials which recruited patients from American and European countries may limit the generalizability of our findings to research conducted in these areas. There is accumulating evidence highlighting poor compliance with trial registration requirements, yet the scientific community has been slow in responding to the need for processes to improve registration behaviors. This paper serves as a snapshot which identified signals of poor registration rates and frequently incomplete trial registration data in surgical pediatric trials. Valid research into reporting improvements over time is challenging with respect to selecting a sample, identifying surgical trials in child health and extracting data. These data serve as a baseline to measure improvements in pediatric surgical trial registration behaviors moving forward. 4. Conclusion To ensure that pediatric surgery is guided by reliable evidence, clinical research must be useful. Clinical trial registration should be complete and consistent with the corresponding publication(s). Our analysis of surgical pediatric trials published in 2014 reveals poor prospective trial registration rates and incomplete registration data. Future initiatives should evaluate implementation strategies to improve trial registration behavior in pediatric surgical trials. Appendix 1. Study search strategy MEDLINE (1946 to March Week 3 2015, Ovid Interface) — March 26, 2015. 1. exp. infant/ or exp. child/ or adolescent/ or minors/ or exp. pediatrics/. 2. (infan$ or newborn$ or baby or babies or neonat$ or preterm$ or child$ or schoolchild$ or school$ or preschool$ or toddler$ adoles$ or teen$ or kid or kids or boy$ or girl$ or minor$ or pubert$ or pubescen$ or prepubescen$ or pediatric or pediatrics or pediatric or pediatrics or nursery or kindergar$ or highschool$ or youth or youths).mp. 3. “18 years or younger”.ti,ab. 4. exp. clinical trials as topic/ or exp. clinical trial/ or rct.ti,ab. 5. (clinical-trial$ or clinical-stud$ or controlled-trial$ or controlledstud$ or multicentre-trial$ or multicentre-stud$ or multicentertrial$ or multicenter-stud$).ti,ab. 6. Exp clinical protocols/ or protocol$.mp. 7. (1 or 2 or 3) and (4 or 5 or 6). 8. animal/ not human/. 9. 7 not 8. 10. limit 9 to (yr = “2014” and English) Appendix 2. WHO Trial Registration Data Set (Version 1.2.1) [10] The minimum amount of trial information that must appear in a register in order for a given trial to be considered fully registered. There are currently 20 items in the WHO Trial Registration Data Set. It is sometimes referred to as the TRDS.
Please cite this article as: Rokhsefat S, et al, Trial registration in pediatric surgery trials, J Pediatr Surg (2017), https://doi.org/10.1016/ j.jpedsurg.2017.10.049
6
S. Rokhsefat et al. / Journal of Pediatric Surgery xxx (2017) xxx–xxx
1. Primary Registry and Trial Identifying Number Name of Primary Registry, and the unique ID number assigned by the Primary Registry to this trial. 2. Date of Registration in Primary Registry Date when trial was officially registered in the Primary Registry. 3. Secondary Identifying Numbers Other identifiers besides the Trial Identifying Number allocated by the Primary Registry, if any. These include: a. The Universal Trial Number (UTN) b. Identifiers assigned by the sponsor (record Sponsor name and Sponsor-issued trial number (e.g. protocol number)) c. Other trial registration numbers issued by other Registries (both Primary and Partner Registries in the WHO Registry Network, and other registries) d. Identifiers issued by funding bodies, collaborative research groups, regulatory authorities, ethics committees / institutional review boards, etc. All secondary identifiers will have 2 elements: an identifier for the issuing authority (e.g. NCT, ISRCTN, ACTRN) plus a number. There is no limit to the number of secondary identifiers that can be provided. 4. Source(s) of Monetary or Material Support Major source(s) of monetary or material support for the trial (e.g. funding agency, foundation, company, institution). 5. Primary Sponsor The individual, organization, group or other legal entity which takes responsibility for initiating, managing and/or financing a study. The Primary Sponsor is responsible for ensuring that the trial is properly registered. The Primary Sponsor may or may not be the main funder. 6. Secondary Sponsor(s) Additional individuals, organizations or other legal persons, if any, that have agreed with the primary sponsor to take on responsibilities of sponsorship. A secondary sponsor may have agreed to: a. take on all the responsibilities of sponsorship jointly with the primary sponsor; or b. form a group with the Primary Sponsor in which the responsibilities of sponsorship are allocated among the members of the group; or c. act as the Primary Sponsor's legal representative in relation to some or all of the trial sites. 7. Contact for Public Queries Email address, telephone number and postal address of the contact who will respond to general queries, including information about current recruitment status. 8. Contact for Scientific Queries There must be clearly assigned responsibility for scientific leadership to a named Principal Investigator. The PI may delegate responsibility for dealing with scientific enquiries to a scientific contact for the trial. This scientific contact will be listed in addition to the PI. The contact for scientific queries must therefore include: a. Name and title, email address, telephone number, postal address and affiliation of the Principal Investigator, and; b. Email address, telephone number, postal address and affiliation of the contact for scientific queries about the trial (if applicable). The details for the scientific contact may be generic (that is, there does not need to be a named individual): e.g. a generic email address for research team members qualified to answer scientific queries. 9. Public Title Title intended for the lay public in easily understood language. 10. Scientific Title Scientific title of the study as it appears in the protocol submitted for funding and ethical review. Include trial acronym if available.
11. Countries of Recruitment The countries from which participants will be, are intended to be, or have been recruited at the time of registration. 12. Health Condition(s) or Problem(s) Studied Primary health condition(s) or problem(s) studied (e.g., depression, breast cancer, medication error). If the study is conducted in healthy human volunteers belonging to the target population of the intervention (e.g. preventive or screening interventions), enter the particular health condition(s) or problem(s) being prevented. 13. Intervention(s) For each arm of the trial record a brief intervention name plus an intervention description. Intervention Name: For drugs use generic name; for other types of interventions provide a brief descriptive name. a. For investigational new drugs that do not yet have a generic name, a chemical name, company code or serial number may be used on a temporary basis. As soon as the generic name has been established, update the associated registered records accordingly. b. For nondrug intervention types, provide an intervention name with sufficient detail so that it can be distinguished from other similar interventions. Intervention Description: Must be sufficiently detailed for it to be possible to distinguish between the arms of a study (e.g. comparison of different dosages of drug) and/or among similar interventions (e.g. comparison of multiple implantable cardiac defibrillators). For example, interventions involving drugs may include dosage form, dosage, frequency and duration. If the intervention is one or more drugs then use the International Non-Proprietary Name for each drug if possible (not brand/trade names). For an unregistered drug, the generic name, chemical name, or company serial number is acceptable. If the intervention consists of several separate treatments, list them all in one line separated by commas (e.g. “low-fat diet, exercise”). For controlled trials, the identity of the control arm should be clear. The control intervention(s) is/are the interventions against which the study intervention is evaluated (e.g. placebo, no treatment, active control). If an active control is used, be sure to enter in the name(s) of that intervention, or enter “placebo” or “no treatment” as applicable. For each intervention, describe other intervention details as applicable (dose, duration, mode of administration, etc). 14. Key Inclusion and Exclusion Criteria Inclusion and exclusion criteria for participant selection, including age and sex. Other selection criteria may relate to clinical diagnosis and comorbid conditions; exclusion criteria are often used to ensure patient safety. If the study is conducted in healthy human volunteers not belonging to the target population (e.g. a preliminary safety study), enter “healthy human volunteer”. 15. Study Type Study type consists of: a. Type of study (interventional or observational) b. Study design including: i. Method of allocation (randomized/nonrandomized) ii. Masking (is masking used and, if so, who is masked) iii. Assignment (single arm, parallel, crossover or factorial) iv. Purpose c. Phase (if applicable) For randomized trials: the allocation concealment mechanism and sequence generation will be documented. 16. Date of First Enrollment Anticipated or actual date of enrolment of the first participant. 17. Target Sample Size Number of participants that this trial plans to enroll in total.
Please cite this article as: Rokhsefat S, et al, Trial registration in pediatric surgery trials, J Pediatr Surg (2017), https://doi.org/10.1016/ j.jpedsurg.2017.10.049
S. Rokhsefat et al. / Journal of Pediatric Surgery xxx (2017) xxx–xxx
18. Recruitment Status Recruitment status of this trial: a. Pending: participants are not yet being recruited or enrolled at any site b. Recruiting: participants are currently being recruited and enrolled c. Suspended: there is a temporary halt in recruitment and enrolment d. Complete: participants are no longer being recruited or enrolled e. Other 19. Primary Outcome(s) Outcomes are events, variables, or experiences that are measured because it is believed that they may be influenced by the intervention. The Primary Outcome should be the outcome used in sample size calculations, or the main outcome(s) used to determine the effects of the intervention(s). Most trials should have only one primary outcome. For each primary outcome provide: a. The name of the outcome (do not use abbreviations) b. The metric or method of measurement used (be as specific as possible) The timepoint(s) of primary interest. Example: Outcome Name: Depression Metric/method of measurement: Beck Depression Score Timepoint: 18 weeks following end of treatment 20. Key Secondary Outcomes Secondary outcomes are outcomes which are of secondary interest or that are measured at timepoints of secondary interest. A secondary outcome may involve the same event, variable, or experience as the primary outcome, but measured at timepoints other than those of primary interest. As for primary outcomes, for each secondary outcome provide: a. The name of the outcome (do not use abbreviations) b. The metric or method of measurement used (be as specific as possible) c. The timepoint(s) of interest References [1] Dickersin K, Chalmers I. Recognizing, investigating and dealing with incomplete and biased reporting of clinical research: from Francis bacon to the WHO. J R Soc Med 2011;104(12):532–8. [2] Crawford JM, Briggs CL, Engeland CG. Publication bias and its implications for evidence-based clinical decision making. J Dent Educ 2010;746:593–600.
7
[3] De Angelis C, Drazen JM, Frizelle FA, et al. Clinical trial registration: a statement from the International Committee of Medical Journal Editors. N Engl J Med 2004;351:1250–1. [4] The International Committee of Medical Journal Editors. Clinical trials registration 2016. http://icmje.org/about-icmje/faqs/clinical-trials-registration/, Accessed date: 23 September 2016. [5] Ostlie DJ, St Peter SD. The current state of evidence-based pediatric surgery. J Pediatr Surg 2010;45:1940–6. [6] Viergever RF, Karam G, Reis A, et al. The quality of registration of clinical trials: still a problem. PLoS One 2014;9:12–20. [7] Ross JS, Mulvey GK, Hines EM, et al. Trial publication after registration in ClinicalTrials.Gov: a cross-sectional analysis. PLoS Med 2009:6. [8] Huić M, Marušić M, Marušić A. Completeness and changes in registered data and reporting bias of randomized controlled trials in ICMJE journals after trial registration policy. PLoS One 2011;6(9):e25258. [9] Harris PA, Taylor R, Thielke R, et al. Research electronic data capture (REDCap)-a metadata-driven methodology and workflow process for providing translational research informatics support. J Biomed Inform 2009;42:377–81. [10] World Health Organization. WHO trial registration data set (version 1.2.1). Int Clin trials Regist Platf (ICTRP) 2016. http://www.who.int/ictrp/en/, Accessed date: 23 September 2016. [11] The International Committee of Medical Journal Editors. Journals following the ICMJE recommendations. [n.d.] http://icmje.org/about-icmje/faqs/icmje-recommendations/ #5, Accessed date: 19 December 2016. [12] Killeen S, Sourallous P, Hunter IA, et al. Registration rates, adequacy of registration, and a comparison of registered and published primary outcomes in randomized controlled trials published in surgery journals. Ann Surg 2014;259: 193–6. [13] Huser V, Cimino JJ. Evaluating adherence to the International Committee of Medical Journal Editors' policy of mandatory, timely clinical trial registration. J Am Med Inform Assoc 2013;20:e169-74. [14] Harriman SL, Patel J. When are clinical trials registered? An analysis of prospective versus retrospective registration. Trials 2016;17:187. [15] Bashir R, Bourgeois FT, Dunn AG. A systematic review of the processes used to link clinical trial registrations to their published results. Syst Rev 2017;6(1):123. [16] Askie L. Trial registration records, updates, and protocols. Lancet 2016;341. [17] Dwan K, Gamble C, Williamson PR, et al. Systematic review of the empirical evidence of study publication bias and outcome reporting bias – an updated review. PLoS One 2013;8(7):e66844. [18] Chan An-Wen, Altman Douglas G. Identifying outcome reporting bias in randomised trials on PubMed: review of publications and survey of authors. BMJ 2005; 330(7494):753. [19] Hartling L, Hamm M, Klassen T, et al. Standard 2: containing risk of bias. Pediatrics 2012;129:S124–31. [20] Furukawa TA, Watanabe N, Omori IM, et al. Association between unreported outcomes and effect size estimates in Cochrane meta-analyses. JAMA 2007;297: 468–70. [21] Goldacre B, Drysdale H, Powell-Smith A, et al. The COMPare trials project 2016. www.COMParetrials.org, Accessed date: 23 July 2016. [22] Taichman DB, BAckuc J, Baethge C, et al. Sharing clinical trial data — a proposal from the International Committee of Medical Journal Editors. How to encourage trial reporting. NEJM 2016;374:384–6. [23] Moher D, Liberati A, Tetzlaff J, et al. Preferred reporting items for systematic reviews and meta-analyses: the PRISMA statement. PLoS Med 2009;6(7):e1000097.
Please cite this article as: Rokhsefat S, et al, Trial registration in pediatric surgery trials, J Pediatr Surg (2017), https://doi.org/10.1016/ j.jpedsurg.2017.10.049
Contents lists available at ScienceDirect
Journal of Pediatric Surgery journal homepage: www.elsevier.com/locate/jpedsurg
Trial registration in pediatric surgery trials☆,☆☆ Sana Rokhsefat a, Deanna E. Morra a, Martin Offringa a,b, Lisa M. Askie c,d, Lauren E. Kelly a,e,⁎ a
Child Health Evaluative Sciences, The Hospital for Sick Children, Toronto, Canada University of Toronto, Faculty of Medicine, Toronto, Canada c University of Sydney, Sydney Medical School, Sydney, Australia d University of Sydney, NHMRC Clinical Trials Centre, Sydney, Australia e Seneca, School of Biological Sciences and Applied Chemistry, Toronto, Canada b
a r t i c l e
i n f o
Article history: Received 4 May 2017 Received in revised form 29 August 2017 Accepted 16 October 2017 Available online xxxx Key words: Trial registration Outcome measures Pediatric surgery trials
a b s t r a c t Background: Prospective clinical trial registration serves to increase transparency and to mitigate selective reporting bias. An assessment of adult surgical trials revealed poor trial registration practice with incomplete provision of information in registries and inconsistent information in the corresponding publication. The extent and completeness of pediatric surgical trial registration are unknown. We aimed to determine the proportion and adequacy of clinical trial registration in pediatric surgery trials published in 2014. Methods: Using sensitive search strategies in MEDLINE, abstracts and full-texts of prospective pediatric intervention studies published in 2014 were screened in duplicate. Pediatric surgical trials were included. Clinical trial registration numbers obtained from publications were searched in trial registries. Data were extracted based on WHO 20-item minimum data set to determine the completeness of registration data. The proportion of registered trials was recorded and registration data were compared to reported data in the corresponding publication. Results: Our search and abstract screening identified 3375 articles for full text review. Following coding, a total of 54 pediatric surgical trials were included and analyzed; 28% (15/54) of which published a registration number. In trials which reported a registration number, 40% (6/15) were retrospectively registered and 40% (6/15) had made changes to their registered primary and/or secondary outcome measures. One included published trial reported an incorrect registration number. Conclusions: Analysis of pediatric surgery trials published in 2014 revealed a poor prospective trial registration rate and incomplete registration data. Our study supports future initiatives for improved registration behaviors in pediatric surgery trials to ensure high-quality, transparent, reproducible evidence is generated. Study type: Therapeutic (clinical trials), level II. © 2017 Elsevier Inc. All rights reserved.
Pediatric surgical practice must be guided by a reliable evidencebase including randomized, comparative effectiveness trials. Reporting bias in clinical trials, including publication bias (preferential publication of positive studies) and selective reporting bias, can distort the validity of medical literature and undermine clinical decision-making [1,2]. Prospective trial registries emerged as a strategy to deter selective reporting by increasing transparency and promoting accountability
Abbreviations: ICMJE, International Committee of Medical Journal Editors; StaR Child Health, Standards for Research in Child Health; WHO, World Health Organization. ☆ Funding: This research did not receive any specific grant from funding agencies in the public, commercial, or not-for-profit sectors. ☆☆ Disclosure: Dr. Lisa M. Askie is the Manager of the Australian New Zealand Clinical Trials Registry; the other authors have no conflicts of interest relevant to this article to disclose. ⁎ Corresponding author at: The Hospital for Sick Children, Toronto, Canada, M5G 0A4. Tel.: +1 416 813 7654x301025. E-mail addresses: [email protected], [email protected] (L.E. Kelly).
through public cataloguing of clinical trials [3] prior to their commencement. The International Committee of Medical Journal Editors (ICMJE) implemented a policy in 2005, mandating registration of clinical trials as a prerequisite for publication in their respective journals which was adopted by eleven peer-reviewed scientific journals [3]. Their intent was to reduce reporting bias, prevent unnecessary duplication and provide an information portal for patients and the public [4,5]. The effectiveness of prospective trial registration, particularly in its role in increasing transparency and accountability, relies on the provision of complete and accurate information [6,7]. In the past, the completeness of information provided in prospective trial registries has been variable [6] and reporting bias remains an ongoing challenge [8]. The assessment of information provided in trial registries has been predominantly focused on adult trials and the completeness of pediatric surgical trial registration has yet to be investigated. The purpose of this paper is to investigate the proportion of pediatric surgical trials that have been registered and to compare information recorded in
https://doi.org/10.1016/j.jpedsurg.2017.10.049 0022-3468/© 2017 Elsevier Inc. All rights reserved.
Please cite this article as: Rokhsefat S, et al, Trial registration in pediatric surgery trials, J Pediatr Surg (2017), https://doi.org/10.1016/ j.jpedsurg.2017.10.049
2
S. Rokhsefat et al. / Journal of Pediatric Surgery xxx (2017) xxx–xxx
clinical trial registries with the information reported in corresponding publications. The objectives of this study are to: (1) assess the proportion of registered pediatric surgical trials and the completeness of information provided in trial registries based on the WHO 20-Item Minimum Data Set; and (2) determine if information reported in clinical trial registries matches the information presented in the corresponding publication.
than 21 years were excluded. Data from each published full-text article were collected and managed using REDCap a secure, web-based application designed to support data capture, hosted at The Hospital for Sick Children [9]. Data were collected by six independent reviewers, with a sample completed in duplicate to assess consistency of data collection. Random spot checks for consistency between reviewers were completed weekly by the Senior Project Manager.
1. Methods
1.2. Data extraction and analysis
1.1. Study selection and search strategy
The proportion of registered trials, and the completeness of registered data versus reported data in a corresponding publication were determined. Using the WHO Trial Registration Data Set (Appendix 2) [10], two investigators (SR, DM) independently manually extracted all data into an Excel (Microsoft, Redmond, United States of America) database. The WHO Trial Registration Data Set was chosen as it is the internationally agreed standard for the minimum information required for a trial to be considered fully registered [10]. Completeness of WHO Trial Registration Data Set was classified as a) did not meet criteria, b) partially met criteria or c) fully met criteria. “Did not meet criteria” was defined as registered information that did not match reported published
A MEDLINE search was conducted using a sensitive and specific search strategy (Appendix 1) restricted to 2014 publications (Fig. 1). The search strategy employed MeSH and non-MeSH terms that are relevant to pediatrics, clinical trials and study protocols. Publications of abstracts and titles were retrieved and screened independently by two reviewers in Endnote. Abstracts pertaining to prospective, interventional trials in children were selected for full-text assessment, and relevant full-text publications were reviewed. Studies without an intervention, without patients younger than 18 or with a mean age greater
Fig. 1. Modified PRISMA [23] flow diagram of prospective pediatric surgical trials published in 2014.
Please cite this article as: Rokhsefat S, et al, Trial registration in pediatric surgery trials, J Pediatr Surg (2017), https://doi.org/10.1016/ j.jpedsurg.2017.10.049
S. Rokhsefat et al. / Journal of Pediatric Surgery xxx (2017) xxx–xxx
information and/or was not clear. For example, if the registered primary outcome was pain but the publication did not report any pain measures. “Partially met criteria” was defined as trials that partially reported or published data items. For example, if an outcome measurement tool was planned to be assessed at 6 and 12 months but only 12-month outcome data were reported in the publication this would constitute a partial criteria match. “Fully met criteria” was defined as trials that fully matched reported (published) with registered (recorded) data items. Retrospective registration was defined as studies which recruited patients before the recorded date of complete trial registration. Disagreements regarding data entries were resolved through discussion with the senior author. If the authors referred to an additional publication which included study details, this information was retrieved and considered as reported (published) information. Journals following the ICMJE recommendations were identified on their website [11] and classified as mandating clinical trial registration. To avoid confusion with registered and nonregistered trials, all information included in clinical trial registries will be referred to as recorded while data obtained from publications will be termed reported. The analysis of completeness of reporting and selection bias is descriptive and results are presented as totals and percentage of all included trials. 2. Results Fifty-four prospective pediatric surgical trials published in 2014 were identified (Fig. 1) and included in this analysis. Descriptive and methodological features included in trials are reported in Table 1. There were no disagreements which required third party review with the senior author. Only 28% (15/54) of trials reported a registration number. There were 17 trials published in journals which mandate registration; 25.6% (10/39) of trials that did not report a registration number were published in a journal that mandates registration. The demographics of included studies are described in Table 1. The majority of trials that reported a registration number were registered with ClinicalTrials.gov; 60% (9/15) (Table 2). Of the registered trials, 40% (6/15) were retrospectively registered and 40% (6/15) made changes to their registered primary and/or secondary outcome measures (Table 2) after registration. Discrepancies between information recorded in the registration record and information reported in a corresponding journal article are found in Table 3. The majority 93.3% (14/15), of trials reported accurate registration numbers; one Table 1 Characteristics of published clinical trials reporting a Clinical Trial Registration Number versus those without a reported Registration Number.
Continent of Recruitment Africa Asia Australia/New Zealand Europe North America South America Journal Registration Mandated Registration Not Mandated Comparison Group Placebo Active Control Standard of Care None Historical Control No treatment control Blinded study design Randomized
Registered No. (%) of Trials (n = 15)
Nonregistered No. (%) of Trials (n = 39)
0 (0) 2 (13.3) 1 (6.7) 6 (40) 4 (26.7) 3 (20)
2 (5.1) 20 (51.3) 1 (2.6) 7 (17.9) 8 (20.5) 1 (2.6)
7 (46.7) 8 (53.3)
10 (25.6) 29 (74.3)
1 (6.7) 9 (60) 4 (26.7) 0 (0) 0 (0) 0 (0) 6 (40) 13 (86.7)
0 (0) 14 (35.9) 10 (25.6) 10 (25.6) 1 (2.6) 4 (10.2) 8 (20.5) 16 (41.0)
○ 3 of the registered trials recruited participants from more than one country. ○ 1 unregistered trial recruited participants from more than one country.
3
Table 2 Clinical trial registry, retrospective registration and changes. Clinical Trial Registry
Registered vs. Reported Data: No. (%) of Trials (N = 15)
ClinicalTrials.gov (NCT) The International ISRCTN Netherlands Trial Registry (NTR) Brazilian Clinical Trials Registry (ReBec) Clinical Trials Registry—India (CTRI) Retrospective Registration Retrospective changes to primary and secondary outcome
9 (60) 2 (13.3) 2 (13.3) 1 (6.7) 1 (6.7) 6 (40) 6 (40)
trial reported an incorrect registration number. With respect to outcome reporting, only 20% (3/15) of trials had registered a primary outcome that matched the primary outcome reported in the journal article (Table 3). Most included pediatric surgery trials, 40% (6/15), recorded primary outcomes which partially matched the publication and 40% (6/15) did not have a registered primary outcome that matched the reported outcome. Examination of primary outcome reporting discrepancies (Table 4) revealed that 13.3% (2/15) of trials published a primary outcome described as secondary outcome in the registry, and 13.3% (2/15) of trials had a registered primary outcome reported as a secondary outcome in the corresponding publication. One trial published a primary outcome not recorded in the registry, another had a registered primary outcome which was not reported in the publication. Discrepancies in secondary outcome reporting revealed 33.3% (5/15) of the registered trials recorded a secondary outcome that did not match the reported secondary outcome (Table 4). The majority of trials, 60% (9/15) registered a secondary outcome that partially matched the secondary outcome published. One trial reported a secondary outcome that was not registered (Table 4). One trial published a secondary outcome described as primary outcome in registry and another trial registered a secondary outcome which was reported as primary outcome in the publication (Table 4). 3. Discussion Over a decade has passed since the International Committee of Medical Journal Editors (ICMJE) mandated registration as prerequisite for publication of clinical trials [3] however the proportion of publications reporting a registration number and the provision of complete registration data remains variable [5]. To our knowledge, this study is the first to compare data recorded in clinical trial registries with the corresponding publication in pediatric surgery trials. The vulnerability of children coupled with the potential of serious adverse consequences associated with pediatric surgeries such as permanent disability and/or death makes complete and timely trial registration and updates a particularly important ethical requirement. Surgical education has historically been guided by apprenticeship style teaching engrained in institutional practice [4]. Complete trial registration data ensure that those involved in health care decision-making have access to publically available information on ongoing and completed studies. Analysis of pediatric surgical trials published in 2014 revealed that only 28% of trials reported a trial registration number. Our findings are similar to an analysis of surgical RCTs published in 2009 and 2010 from 10 high-impact factor surgical journals, in which 44% (111/246) of included studies were registered [12]. This is concerning as most high-impact surgical journals have adopted the ICMJE mandate for trial registration [11]. Registration before enrollment of the first participant is a scientific, moral and ethical responsibility in adherence with standards for quality clinical trials and transparency. Conversely, in a previous report on 698 clinical trials in 5 general medical journals which mandate registration, 89% of trials that began in 2008 were registered appropriately [13]. In our study 25.6% (10/39) of pediatric
Please cite this article as: Rokhsefat S, et al, Trial registration in pediatric surgery trials, J Pediatr Surg (2017), https://doi.org/10.1016/ j.jpedsurg.2017.10.049
4
S. Rokhsefat et al. / Journal of Pediatric Surgery xxx (2017) xxx–xxx
Table 3 Discrepancies between registered and reported data. WHO Dataset Item Item 1: Primary Registry and Trial Identifying Number
Item 2: Date of Registration in Primary Registry
Item 3: Secondary Identifying Numbers
Item 4: Source(s) of Monetary or Material Support
Item 5: Primary Sponsor
Item 6: Secondary Sponsor(s)
Item 7: Contact for Public Queries
Item 8: Contact for Scientific Queries
Item 9: Public Title
Item 10: Scientific Title
Item 11: Countries of Recruitment
Item 12: Condition
Item 13: Intervention(s)
Item 14: Key inclusion or exclusion criteria
Item 15: Study type
Item 16: Date of First Enrollment
Item 17: Target Sample Size
Item 18: Recruitment Status
Item 19: Primary outcome
Item 20: Secondary outcome(s)
surgical trials were published in journals which mandate registration but without a registration number listed in the manuscript. These data highlight variability in enforcement and accountability by journals with regard to implementing a trial registration mandate. Although trial registration is required by the major Journals, there is limited accountability for maintaining and monitoring the completeness of registration. Ideally, trial data provided in registries should be cross-referenced with data analysis plans during manuscript preparation prior to publication, mitigating selective outcome reporting bias. Any unreported outcomes or changes to registered outcomes should be documented in the final publication with justification for the
Registered vs. Reported Data: No. (%) of Trials (N = 15) Did not meet criteria Partially met criteria Fully met criteria Did not meet criteria Partially met criteria Fully met criteria Did not meet criteria Partially met criteria Fully met criteria Did not meet criteria Partially met criteria Fully met criteria Did not meet criteria Partially met criteria Fully met criteria Did not meet criteria Partially met criteria Fully met criteria Did not match Partially matched Fully matched Did not meet criteria Partially met criteria Fully met criteria Did not meet criteria Partially met criteria Fully met criteria Did not meet criteria Partially met criteria Fully met criteria Did not meet criteria Partially met criteria Fully met criteria Did not meet criteria Partially met criteria Fully met criteria Did not meet criteria Partially met criteria Fully met criteria Did not meet criteria Partially met criteria Fully met criteria Did not meet criteria Partially met criteria Fully met criteria Did not meet criteria Partially met criteria Fully met criteria Did not meet criteria Partially met criteria Fully met criteria Did not meet criteria Partially met criteria Fully met criteria Did not meet criteria Partially met criteria Fully met criteria Did not meet criteria Partially met criteria Fully met criteria
1 (6.7) 0 (0) 14 (100) 14 (93.3) 0 (0) 1 (6.7) 11 (73.3) 1 (6.7) 3 (20) 11 (73.3) 0 (0) 4 (26.7) 7 (46.6) 0 (0) 8 (53.3) 14 (93.3) 0 (0) 1 (6.7) 15 (100) 0 (0) 0 (0) 9 (60) 0 (0.0) 6 (40) 15 (100) 0 (0) 0 (0) 4 (26.7) 10 (66.7) 1 (6.7) 3 (20) 0 (0) 12 (80) 0 (0) 5 (33.3) 10 (66.7) 0 (0) 1 (6.7) 14 (93.3) 1 (6.7) 10 (66.7) 4 (26.7) 0 (0) 6 (40) 9 (60) 8 (53.3) 4 (26.7) 3 (20) 15 (100) 0 (0) 0 (0) 7 (46.6) 0 (0) 8 (53.3) 6 (40) 6 (40) 3 (20) 5 (33.3) 9 (60) 1 (6.7)
registered protocol deviation. Trial data should be published even in the absence of registration as an ethical commitment to the study participants. These trials could include a statement justifying the timing or absence of trial registration. An inherent challenge with trial registration is determining who must provide the dedicated resources for checking compliance. A potential solution is to have key governing bodies, such as, ethics committees, institutions and funders also involved in the enforcement of prospective trial registration owing to their key role early in the design of trials [12]. As ethics committees review protocols before recruitment has begun, requiring trial registration before approving trials is a key step in
Please cite this article as: Rokhsefat S, et al, Trial registration in pediatric surgery trials, J Pediatr Surg (2017), https://doi.org/10.1016/ j.jpedsurg.2017.10.049
S. Rokhsefat et al. / Journal of Pediatric Surgery xxx (2017) xxx–xxx
5
Table 4 Discrepancies in primary outcome reporting in registered information versus publication. Discrepancy
Primary Outcome No. (%) of Trials (N = 15)
Secondary Outcome(s) No. (%) of Trials (N = 15)
Reported outcome not in registry Partially reported outcome Published outcome described as secondary outcome in registry Registered outcome reported as secondary outcome in text Registered outcome not reported Nonregistered follow-up outcomes No secondary outcomes Reported outcomes matched the registry
1 (6.7) 6 (40) 2 (13.3) 2 (13.3) 1 (6.7) 3 (20) – 3 (20)
1 (6.7) 8 (53.3) 1 (6.7) 1 (6.7) 3 (20) 2 (13.3) 1 (6.7) 1 (1)
increasing registration compliance. A comprehensive approach is required by all stakeholders to improve the rates of prospective trial registration. The information provided in registries compared to the corresponding journal article was often incomplete. One value of trial registration is that public cataloguing of information offers the opportunity for journal editors to identify and prevent selective reporting bias [11] but this is dependent on the provision of complete trial registration data. Providing complete information in registries should no longer remain voluntary, as registry data are of limited use when registry entries simply state “Not provided.” Software advances now present a feasible solution, and can include autolinking of large international registries with the corresponding publications, allowing for improved interpretation. A recent review identified that automated linkage between registration and corresponding publications has not increased over time [15]. Ensuring data interfaces are user-friendly may encourage investigators to input more than the minimum required information. Mandated linkage of publications directly to the registration may provide incentives for updating and maintaining registry information. The ANZCTR (Australian and New Zealand Clinical Trial Registry) recently added a feature to link a full protocol to a registration record [16]. Guidelines for investigators with respect to when to upload trial results (without violating manuscript publishing) may facilitate timely registry data updates. Improvements to clinical trial registry operations such as these may improve registry behaviors. Trial registration may be viewed as an administrative duty that can be completed at any time. If so, this limits the value of prospective trial registration [14]. In our study, 40% of trials that reported a registration number were retrospectively registered. This is similar to retrospective trial registration rates reported by others [7,14]. While there is a slight possibility that researchers may still not be aware of trial registration or not realize their study meets the definition for trial registration [16], retrospective registration may be problematic for several reasons including registration only to satisfy journal requirements or by allowing changes to the outcome measured or reported after reviewing the data. Outcome measures that are intended but not reported [17,18] may be related to a lack of statistical significance [19] or potentially related to inappropriate outcome selection prior to the initiation of the trial. A study which examined the effects of selective outcome reporting on meta-analysis showed that half of trials relevant to a systematic review did not contribute to the important patient outcomes of the meta-analysis [20]. Interestingly, unconventional methods have been introduced to address selective outcome reporting. In the United Kingdom, the COMPare group is actively monitoring and reporting discrepancies in trials in an effort to increase accountability, with variable success [16,21]. The United States Department of Health and Human Services recently proposed tightening rules for ClinicalTrials.gov data submission, requiring all data to be reported regardless of FDA approval [22]. A limitation of this paper is the small sample size and the inability to evaluate data included in future publications. For example, if long-term outcome assessment was recorded in the clinical trial registry but no information was provided on such a planned future analysis in the
publication, this would have been considered a partial match. The low number of included trials which recruited patients from American and European countries may limit the generalizability of our findings to research conducted in these areas. There is accumulating evidence highlighting poor compliance with trial registration requirements, yet the scientific community has been slow in responding to the need for processes to improve registration behaviors. This paper serves as a snapshot which identified signals of poor registration rates and frequently incomplete trial registration data in surgical pediatric trials. Valid research into reporting improvements over time is challenging with respect to selecting a sample, identifying surgical trials in child health and extracting data. These data serve as a baseline to measure improvements in pediatric surgical trial registration behaviors moving forward. 4. Conclusion To ensure that pediatric surgery is guided by reliable evidence, clinical research must be useful. Clinical trial registration should be complete and consistent with the corresponding publication(s). Our analysis of surgical pediatric trials published in 2014 reveals poor prospective trial registration rates and incomplete registration data. Future initiatives should evaluate implementation strategies to improve trial registration behavior in pediatric surgical trials. Appendix 1. Study search strategy MEDLINE (1946 to March Week 3 2015, Ovid Interface) — March 26, 2015. 1. exp. infant/ or exp. child/ or adolescent/ or minors/ or exp. pediatrics/. 2. (infan$ or newborn$ or baby or babies or neonat$ or preterm$ or child$ or schoolchild$ or school$ or preschool$ or toddler$ adoles$ or teen$ or kid or kids or boy$ or girl$ or minor$ or pubert$ or pubescen$ or prepubescen$ or pediatric or pediatrics or pediatric or pediatrics or nursery or kindergar$ or highschool$ or youth or youths).mp. 3. “18 years or younger”.ti,ab. 4. exp. clinical trials as topic/ or exp. clinical trial/ or rct.ti,ab. 5. (clinical-trial$ or clinical-stud$ or controlled-trial$ or controlledstud$ or multicentre-trial$ or multicentre-stud$ or multicentertrial$ or multicenter-stud$).ti,ab. 6. Exp clinical protocols/ or protocol$.mp. 7. (1 or 2 or 3) and (4 or 5 or 6). 8. animal/ not human/. 9. 7 not 8. 10. limit 9 to (yr = “2014” and English) Appendix 2. WHO Trial Registration Data Set (Version 1.2.1) [10] The minimum amount of trial information that must appear in a register in order for a given trial to be considered fully registered. There are currently 20 items in the WHO Trial Registration Data Set. It is sometimes referred to as the TRDS.
Please cite this article as: Rokhsefat S, et al, Trial registration in pediatric surgery trials, J Pediatr Surg (2017), https://doi.org/10.1016/ j.jpedsurg.2017.10.049
6
S. Rokhsefat et al. / Journal of Pediatric Surgery xxx (2017) xxx–xxx
1. Primary Registry and Trial Identifying Number Name of Primary Registry, and the unique ID number assigned by the Primary Registry to this trial. 2. Date of Registration in Primary Registry Date when trial was officially registered in the Primary Registry. 3. Secondary Identifying Numbers Other identifiers besides the Trial Identifying Number allocated by the Primary Registry, if any. These include: a. The Universal Trial Number (UTN) b. Identifiers assigned by the sponsor (record Sponsor name and Sponsor-issued trial number (e.g. protocol number)) c. Other trial registration numbers issued by other Registries (both Primary and Partner Registries in the WHO Registry Network, and other registries) d. Identifiers issued by funding bodies, collaborative research groups, regulatory authorities, ethics committees / institutional review boards, etc. All secondary identifiers will have 2 elements: an identifier for the issuing authority (e.g. NCT, ISRCTN, ACTRN) plus a number. There is no limit to the number of secondary identifiers that can be provided. 4. Source(s) of Monetary or Material Support Major source(s) of monetary or material support for the trial (e.g. funding agency, foundation, company, institution). 5. Primary Sponsor The individual, organization, group or other legal entity which takes responsibility for initiating, managing and/or financing a study. The Primary Sponsor is responsible for ensuring that the trial is properly registered. The Primary Sponsor may or may not be the main funder. 6. Secondary Sponsor(s) Additional individuals, organizations or other legal persons, if any, that have agreed with the primary sponsor to take on responsibilities of sponsorship. A secondary sponsor may have agreed to: a. take on all the responsibilities of sponsorship jointly with the primary sponsor; or b. form a group with the Primary Sponsor in which the responsibilities of sponsorship are allocated among the members of the group; or c. act as the Primary Sponsor's legal representative in relation to some or all of the trial sites. 7. Contact for Public Queries Email address, telephone number and postal address of the contact who will respond to general queries, including information about current recruitment status. 8. Contact for Scientific Queries There must be clearly assigned responsibility for scientific leadership to a named Principal Investigator. The PI may delegate responsibility for dealing with scientific enquiries to a scientific contact for the trial. This scientific contact will be listed in addition to the PI. The contact for scientific queries must therefore include: a. Name and title, email address, telephone number, postal address and affiliation of the Principal Investigator, and; b. Email address, telephone number, postal address and affiliation of the contact for scientific queries about the trial (if applicable). The details for the scientific contact may be generic (that is, there does not need to be a named individual): e.g. a generic email address for research team members qualified to answer scientific queries. 9. Public Title Title intended for the lay public in easily understood language. 10. Scientific Title Scientific title of the study as it appears in the protocol submitted for funding and ethical review. Include trial acronym if available.
11. Countries of Recruitment The countries from which participants will be, are intended to be, or have been recruited at the time of registration. 12. Health Condition(s) or Problem(s) Studied Primary health condition(s) or problem(s) studied (e.g., depression, breast cancer, medication error). If the study is conducted in healthy human volunteers belonging to the target population of the intervention (e.g. preventive or screening interventions), enter the particular health condition(s) or problem(s) being prevented. 13. Intervention(s) For each arm of the trial record a brief intervention name plus an intervention description. Intervention Name: For drugs use generic name; for other types of interventions provide a brief descriptive name. a. For investigational new drugs that do not yet have a generic name, a chemical name, company code or serial number may be used on a temporary basis. As soon as the generic name has been established, update the associated registered records accordingly. b. For nondrug intervention types, provide an intervention name with sufficient detail so that it can be distinguished from other similar interventions. Intervention Description: Must be sufficiently detailed for it to be possible to distinguish between the arms of a study (e.g. comparison of different dosages of drug) and/or among similar interventions (e.g. comparison of multiple implantable cardiac defibrillators). For example, interventions involving drugs may include dosage form, dosage, frequency and duration. If the intervention is one or more drugs then use the International Non-Proprietary Name for each drug if possible (not brand/trade names). For an unregistered drug, the generic name, chemical name, or company serial number is acceptable. If the intervention consists of several separate treatments, list them all in one line separated by commas (e.g. “low-fat diet, exercise”). For controlled trials, the identity of the control arm should be clear. The control intervention(s) is/are the interventions against which the study intervention is evaluated (e.g. placebo, no treatment, active control). If an active control is used, be sure to enter in the name(s) of that intervention, or enter “placebo” or “no treatment” as applicable. For each intervention, describe other intervention details as applicable (dose, duration, mode of administration, etc). 14. Key Inclusion and Exclusion Criteria Inclusion and exclusion criteria for participant selection, including age and sex. Other selection criteria may relate to clinical diagnosis and comorbid conditions; exclusion criteria are often used to ensure patient safety. If the study is conducted in healthy human volunteers not belonging to the target population (e.g. a preliminary safety study), enter “healthy human volunteer”. 15. Study Type Study type consists of: a. Type of study (interventional or observational) b. Study design including: i. Method of allocation (randomized/nonrandomized) ii. Masking (is masking used and, if so, who is masked) iii. Assignment (single arm, parallel, crossover or factorial) iv. Purpose c. Phase (if applicable) For randomized trials: the allocation concealment mechanism and sequence generation will be documented. 16. Date of First Enrollment Anticipated or actual date of enrolment of the first participant. 17. Target Sample Size Number of participants that this trial plans to enroll in total.
Please cite this article as: Rokhsefat S, et al, Trial registration in pediatric surgery trials, J Pediatr Surg (2017), https://doi.org/10.1016/ j.jpedsurg.2017.10.049
S. Rokhsefat et al. / Journal of Pediatric Surgery xxx (2017) xxx–xxx
18. Recruitment Status Recruitment status of this trial: a. Pending: participants are not yet being recruited or enrolled at any site b. Recruiting: participants are currently being recruited and enrolled c. Suspended: there is a temporary halt in recruitment and enrolment d. Complete: participants are no longer being recruited or enrolled e. Other 19. Primary Outcome(s) Outcomes are events, variables, or experiences that are measured because it is believed that they may be influenced by the intervention. The Primary Outcome should be the outcome used in sample size calculations, or the main outcome(s) used to determine the effects of the intervention(s). Most trials should have only one primary outcome. For each primary outcome provide: a. The name of the outcome (do not use abbreviations) b. The metric or method of measurement used (be as specific as possible) The timepoint(s) of primary interest. Example: Outcome Name: Depression Metric/method of measurement: Beck Depression Score Timepoint: 18 weeks following end of treatment 20. Key Secondary Outcomes Secondary outcomes are outcomes which are of secondary interest or that are measured at timepoints of secondary interest. A secondary outcome may involve the same event, variable, or experience as the primary outcome, but measured at timepoints other than those of primary interest. As for primary outcomes, for each secondary outcome provide: a. The name of the outcome (do not use abbreviations) b. The metric or method of measurement used (be as specific as possible) c. The timepoint(s) of interest References [1] Dickersin K, Chalmers I. Recognizing, investigating and dealing with incomplete and biased reporting of clinical research: from Francis bacon to the WHO. J R Soc Med 2011;104(12):532–8. [2] Crawford JM, Briggs CL, Engeland CG. Publication bias and its implications for evidence-based clinical decision making. J Dent Educ 2010;746:593–600.
7
[3] De Angelis C, Drazen JM, Frizelle FA, et al. Clinical trial registration: a statement from the International Committee of Medical Journal Editors. N Engl J Med 2004;351:1250–1. [4] The International Committee of Medical Journal Editors. Clinical trials registration 2016. http://icmje.org/about-icmje/faqs/clinical-trials-registration/, Accessed date: 23 September 2016. [5] Ostlie DJ, St Peter SD. The current state of evidence-based pediatric surgery. J Pediatr Surg 2010;45:1940–6. [6] Viergever RF, Karam G, Reis A, et al. The quality of registration of clinical trials: still a problem. PLoS One 2014;9:12–20. [7] Ross JS, Mulvey GK, Hines EM, et al. Trial publication after registration in ClinicalTrials.Gov: a cross-sectional analysis. PLoS Med 2009:6. [8] Huić M, Marušić M, Marušić A. Completeness and changes in registered data and reporting bias of randomized controlled trials in ICMJE journals after trial registration policy. PLoS One 2011;6(9):e25258. [9] Harris PA, Taylor R, Thielke R, et al. Research electronic data capture (REDCap)-a metadata-driven methodology and workflow process for providing translational research informatics support. J Biomed Inform 2009;42:377–81. [10] World Health Organization. WHO trial registration data set (version 1.2.1). Int Clin trials Regist Platf (ICTRP) 2016. http://www.who.int/ictrp/en/, Accessed date: 23 September 2016. [11] The International Committee of Medical Journal Editors. Journals following the ICMJE recommendations. [n.d.] http://icmje.org/about-icmje/faqs/icmje-recommendations/ #5, Accessed date: 19 December 2016. [12] Killeen S, Sourallous P, Hunter IA, et al. Registration rates, adequacy of registration, and a comparison of registered and published primary outcomes in randomized controlled trials published in surgery journals. Ann Surg 2014;259: 193–6. [13] Huser V, Cimino JJ. Evaluating adherence to the International Committee of Medical Journal Editors' policy of mandatory, timely clinical trial registration. J Am Med Inform Assoc 2013;20:e169-74. [14] Harriman SL, Patel J. When are clinical trials registered? An analysis of prospective versus retrospective registration. Trials 2016;17:187. [15] Bashir R, Bourgeois FT, Dunn AG. A systematic review of the processes used to link clinical trial registrations to their published results. Syst Rev 2017;6(1):123. [16] Askie L. Trial registration records, updates, and protocols. Lancet 2016;341. [17] Dwan K, Gamble C, Williamson PR, et al. Systematic review of the empirical evidence of study publication bias and outcome reporting bias – an updated review. PLoS One 2013;8(7):e66844. [18] Chan An-Wen, Altman Douglas G. Identifying outcome reporting bias in randomised trials on PubMed: review of publications and survey of authors. BMJ 2005; 330(7494):753. [19] Hartling L, Hamm M, Klassen T, et al. Standard 2: containing risk of bias. Pediatrics 2012;129:S124–31. [20] Furukawa TA, Watanabe N, Omori IM, et al. Association between unreported outcomes and effect size estimates in Cochrane meta-analyses. JAMA 2007;297: 468–70. [21] Goldacre B, Drysdale H, Powell-Smith A, et al. The COMPare trials project 2016. www.COMParetrials.org, Accessed date: 23 July 2016. [22] Taichman DB, BAckuc J, Baethge C, et al. Sharing clinical trial data — a proposal from the International Committee of Medical Journal Editors. How to encourage trial reporting. NEJM 2016;374:384–6. [23] Moher D, Liberati A, Tetzlaff J, et al. Preferred reporting items for systematic reviews and meta-analyses: the PRISMA statement. PLoS Med 2009;6(7):e1000097.
Please cite this article as: Rokhsefat S, et al, Trial registration in pediatric surgery trials, J Pediatr Surg (2017), https://doi.org/10.1016/ j.jpedsurg.2017.10.049