ARTICLE. Editorial Policies of Pediatric Journals. of publication bias and selective reporting of study results 1 and continued

Transcription

1 ARTICLE Editorial Policies of Pediatric Journals Survey of Instructions for Authors Joerg J. Meerpohl, MD; Robert F. Wolff, MD; Charlotte M. Niemeyer, MD; Gerd Antes, PhD; Erik von Elm, MD, MSc Objective: To study whether specific recommendations aiming to improve publication practice were included in author instructions of pediatric journals. Design: We identified 69 journals in the subject category pediatrics of the Journal Citation Report 2007 that publish original research articles. From the journals online author instructions, we extracted information regarding endorsement of the Uniform Requirements for Manuscripts (URM) of the International Committee of Medical Journal Editors and of 5 major reporting guidelines such as the Consolidated Standards of Reporting Trials () statement, disclosure of conflicts of interest, and trial registration. Two investigators collected data independently. Results: The URM were mentioned in author instructions of 38 of the 69 journals (55%). Endorsement of reporting guidelines was low: was referred to most frequently (14 journals; 20%); each of the other 4 reporting guidelines was mentioned in less than 10% of author instructions. Fifty-four journals (78%) explicitly required authors to disclose conflicts of interest, and 16 (23%) either recommended or required trial registration. The odds of endorsing the URM increased by 2.25 (95% confidence interval [CI], ) per additional impact factor point. Similarly, the odds increased by 2.32 (95% CI, ) for requiring disclosure of conflicts of interest and by 3.66 (95% CI, ) for requiring trial registration. Conclusions: Many pediatric journals do not include recommendations that aim to improve publication practice in their author instructions. About one-fifth of journals do not require authors to disclose conflicts of interest on manuscript submission and more than three-quarters do not require/recommend trial registration. Arch Pediatr Adolesc Med. 2010;164(3): Author Affiliations: German Cochrane Center, Institute of Medical Biometry and Medical Informatics (Drs Meerpohl, Wolff, Antes, and von Elm), and Division of Pediatric Hematology and Oncology, Department of Pediatrics (Drs Meerpohl and Niemeyer), University Medical Center Freiburg, Freiburg, Germany; Kleijnen Systematic Reviews Ltd, York, England (Dr Wolff); and Swiss Paraplegic Research, Nottwil (Dr von Elm). THERE IS ACCUMULATING EVIdence of publication bias and selective reporting of study results 1 and continued concerns regarding undisclosed conflicts of interest of authors 2 and cases of scientific misconduct. 3 Nonreporting of study results has been identified in several fields of medical research including pediatrics. 4 Different recommendations have been proposed to bring about change in the reporting and publication practice. First, the International Committee of Medical Journal Editors established the Uniform Requirements for Manuscripts Submitted to Biomedical Journals (URM). 5 This document covers issues such as ethical conduct and reporting of biomedical research, preparation and publishing of manuscripts, and editorial policies. To date, it has been endorsed by more than 700 journals worldwide. 6 Second, specific reporting guidelines such as the Consolidated Standards of Reporting Trials () statement aim at helping authors improve the completeness and accuracy of publications of clinical trials. 7 Their bottom-line message is that all relevant information about a study should be made available so that readers can assess its validity. Third, authors but also journal editors and reviewers potentially have financial ties or personal interests in conflict with an article getting published or not. 8,9 For instance, authors of articles favoring the use of calcium-channel blockers were reported to be more likely to have financial relationships with the drugs manufacturers than those with a neutral or critical point of view. 10 The journals policies range from the requirement of full disclosure of conflicts of interest (eg, on the article s title page) to absence of any requirement. Finally, the problem of publication bias has been widely recognized and described in detail over the last 15 years. 1 Registration of all clinical trials prior to patient enrollment has 268

2 been advocated as an important first step to tackle it Several international and national registries such as ClinicalTrials.gov ( or the German Clinical Trials Register ( have been established and can be used free of charge to register or search ongoing and completed trials. A meta-register called the International Clinical Trials Registry Platform incorporating data of all accredited trial registries is being set up under the auspices of the World Health Organization ( 14 The current editorial practice in pediatrics has received only little attention until now. We wondered to what extent pediatric journals formally endorse existing recommendations that aim to improve publication practice. We set out to study the coverage of 4 aspects: whether journals (1) recommend use of the URM; (2) recommend major reporting guidelines; (3) have editorial policies on conflicts of interest; or (4) require trial registration. METHODS We accessed the Journal Citation Report 2007 (Science Edition) 15 on August 5, 2008, through the Institute for Scientific Information Web of Knowledge Web site and identified all journals listed in the subject category pediatrics. We excluded journals that do not publish original research articles because several of the reporting guidelines of interest only apply to original research. From the journals Web sites, we downloaded the author instructions available in August 2008 and determined the geographical location of the editorial office. Two of us ( J.J.M. and R.F.W.) read each document and used relevant key words in electronic full-text searches to identify information on the following aspects: v Endorsement of the URM. 5 v Editorial policies for disclosure of conflicts of interest. v Requirement of trial registration. v Endorsement of the following 5 reporting guidelines and related explanatory articles: 16,17 and extensions; Strengthening the Reporting of Observational Studies in Epidemiology (STROBE) 18,19 ; Standards for Reporting of Diagnostic Accuracy (STARD) 20,21 ; Meta-analysis of Observational Studies in Epidemiology (MOOSE) 22 ; and Quality of Reporting of Meta-analyses (QUOROM). 23 The PRISMA statement (a revised version of the QUOROM statement) was published in July For each item, we distinguished whether it was required (ie, a submitted manuscript would not be accepted if the item was not considered), recommended (ie, its use or fulfillment ought to be considered), or not mentioned. Two of us ( J.J.M. and R.F.W.) extracted and categorized information independently. Discrepancies occurred in 22 of 3588 extracted items (0.6%). They were resolved by rechecking the respective Web sites and discussion among the investigators. Geographical location of journal editorial offices was grouped into North America (United States and Canada), United Kingdom, Europe without the United Kingdom, and other countries. To determine any associations between the journals impact factor scores 25 (used as a continuous variable) and the presence or absence of journal recommendations, we calculated odds ratios (ORs) using logistic regression models. Statistical analyses were performed using Stata software (release 10.0; StataCorp, College Station, Texas). RESULTS We identified 78 journals in the subject category pediatrics. Eight journals were excluded because they do not publish original research articles. The journals Biology of the Neonate and Neonatology (and their instructions) were identical; we included only the first title. Our final study sample comprised 69 pediatric journals. Their impact factor ranged from to For 35 journals (51%), the editorial office was located in North America; for 14 (20%), in the United Kingdom; for 15 (22%), in other European countries; and for the remaining 5 (7%), in Australia or India. The editorial offices of all 6 journals with an impact factor of more than 3 were located in North America. The URM were mentioned in the author instructions of 38 journals (55%). Of these, 18 referenced the URM Web site only; 6, a print version only; and 8, both the Web site and the print version (Table 1). Six did not include any reference to the URM. If author instructions mentioned URM, they mostly did so in the section on manuscript preparation and references. Often, it was not clear whether the entire document was meant or only a part. Impact factor was associated with recommendation of the URM (OR, 2.25; 95% confidence interval [CI], for each additional 1-point increase in impact factor). The author instructions of 54 journals (78%) included a statement on disclosure of conflicts of interest (Table 1). Journals with a higher impact factor were more likely to have a formal policy on conflicts of interest (OR, 2.32 per impact factor point; 95% CI, ). Presence of a conflict of interest policy showed little variation by geographic location (Table 1). When analyzing the author instructions, it was not always clear what authors were expected to do, eg, whether their potential conflicts of interest should only be disclosed in the covering letter to the editor or should be published as part of the article. Trial registration was mentioned by 16 journals (23%); 11 required that a trial be registered as a condition of publication and 5 simply recommended registration. Seven journals suggested suitable trial registries (eg, Recommendations for trial registration were least common in journals in Europe without the United Kingdom (7%) and most common among journals in the United Kingdom (43%), but the variation was not statistically significant (Table 1). Further, journals with a higher impact factor were more likely to mention trial registration (OR, 3.66 per impact factor point; 95% CI, ). The statement was referenced by 14 journals, of which 3 required authors to follow this reporting guideline (Table 2). Use of the checklist and flowchart was recommended by 8 and 6 journals, respectively. Some journals gave no Web or printed citation for material. Other reporting guidelines, such as STROBE for observational studies, MOOSE for meta-analyses of observational studies, QUOROM for meta-analyses of trials, and STARD for diagnostic test accuracy studies, were mentioned only rarely (Table 2). None of the journals provided a reference to the detailed explanatory articles accompanying the, STARD, or STROBE statements. 269

3 Table 1. Advice Given to Authors by Pediatric Journals No. (%) Geographical Location Impact Factor All (N=69) NA (n=35) UK (n=14) EU (n=15) Other (n=5) 1 (n=27) (n=23) (n=13) Advice about use of the Uniform Requirements for Manuscripts Mentioned 38 (55) 21 (60) 7 (50) 6 (40) 4 (80) 11 (41) 12 (52) 11 (85) 4 (67) Web URL given 26 (68) 13 (62) 5 (71) 4 (66) 4 (80) 9 (33) 6 (26) 7 (54) 4 (67) Print version referenced 14 (37) 7 (33) 2 (29) 4 (66) 1 (20) 5 (19) 4 (17) 5 (39) 0 Not mentioned 31 (45) 14 (40) 7 (50) 9 (60) 1 (20) 16 (59) 11 (48) 2 (15) 2 (33) Advice about disclosure of conflicts of interest Mentioned 54 (78) 28 (80) 10 (71) 12 (80) 4 (80) 20 (74) 15 (65) 13 (100) 6 (100) Not mentioned 15 (22) 7 (20) 4 (29) 3 (20) 1 (20) 7 (26) 8 (35) 0 (0) 0 Advice about clinical trial registration Mentioned 16 (23) 7 (20) 6 (43) 1 (7) 2 (40) 3 (11) 3 (13) 6 (46) 4 (67) Required 11 (69) 5 (71) 4 (67) 0 (0) 2 (100) 2 (67) 2 (67) 5 (83) 2 (50) Recommended 5 (31) 2 (29) 2 (33) 1 (100) 0 (0) 1 (33) 1 (33) 1 (17) 2 (50) Registry suggested 7 (44) 5 (71) 1 (17) 1 (100) 0 (0) 1 (33) 1 (33) 2 (33) 3 (75) Not mentioned 53 (77) 28 (80) 8 (57) 14 (93) 3 (60) 24 (89) 20 (87) 7 (54) 2 (33) Abbreviations: EU, Europe without UK; NA, North America (United States and Canada); UK, United Kingdom. 3 (n=6) Table 2. Endorsement of Reporting Guidelines in the Author Instructions of Pediatric Journals Extensions No. (%) Paper MOOSE QUOROM STARD STARD Paper STROBE STROBE Paper Mentioned 14 (20) 12 (17) a 12 b (17) 3 (4) 4 (6) 4 (6) 0 (0) 3 (4) 3 (4) Required 3 (21) 0 (0) 0 (0) 1 (33) 1 (25) 1 (25) 0 (0) 1 (33) 0 (0) Recommended 11 (79) 0 (0) 0 (0) 2 (67) 3 (75) 3 (75) 0 (0) 2 (67) 0 (0) Print version referenced 5 (36) c 0 (0) 0 (0) 1 (33) 1 (25) 2 (50) d 0 (0) 2 (67) e 0 (0) Web URL given 12 (86) 0 (0) 0 (0) 3 (100) f 4 (100) g 4 (100) h 0 (0) 1 (33) 0 (0) Not mentioned 55 (80) 57 (83) 57 (83) 66 (96) 65 (94) 65 (94) 69 (100) 66 (96) 66 (96) Abbreviations:, Consolidated Standards of Reporting Trials 16,17 ; MOOSE, Meta-analysis of Observational Studies in Epidemiology 22 ; QUOROM, Quality of Reporting of Meta-analyses 23 ; STARD, Standards for Reporting of Diagnostic Accuracy 20,21 ; STROBE, Strengthening the Reporting of Observational Studies in Epidemiology. 18,19 a Extensions are not mentioned explicitly but can be found by accessing the Web site ( b papers are not mentioned explicitly but can be accessed by following links to or EQUATOR Web sites ( c One journal references an old version of the statement published in JAMA in d Both journals reference published print version of the STARD initiative. e Two journals link to the online article published in BMJ ( but not to STROBE Web site. f Can be accessed by following links to or EQUATOR Web sites. g Can be accessed by following link to Web site. h Two journals reference Web site, which includes link to STARD statement; 1 links to EQUATOR network, and the other, to the main Web site of Annals of Internal Medicine ( COMMENT We analyzed the author instructions of pediatric journals for recommendations aiming to improve publication practice. The endorsement of several reporting guidelines and other recommendations was low. However, journals with a higher impact factor were more likely to have adopted some recommendations (eg, the URM). Recently, the content of author instructions of medical journals with high impact factors was analyzed by Hopewell et al. 26 The URM and the statement were endorsed by 42% and 38% of 165 journals, respectively. We found the proportion of pediatric journals referring to the URM to be larger (55%), while the proportion of journals endorsing the statement (23%) was lower. 26 Schriger et al 27 compared the methodological and statistical content of the author instructions of high-profile journals in 33 clinical categories; most of these instructions gave no advice about statistical methods (61%) or publication guidelines such as URM (59%) or (78%). Matarese 28 compared the quality of author instructions and editorial leadership of Italian and British journals and reported that Italian journals scored worse (eg, none of the Italian journals recommended trial registration). The importance of good publication practice, including registration of clinical trials, in children has been recognized and efforts taken to translate it into editorial prac- 270

4 tice However, few pediatric journals recommended trial registration to their authors despite the ongoing discussion about this topic and the benefit of supplementary trial information made available in registries. 11,32,33 Further, most pediatric journals ask authors to disclose any conflicts of interest. But a substantial number of journals still neither have a published policy on conflicts of interest nor require authors to disclose any. The journals policies vary greatly: some require authors to mention any potential conflicts of interest in the submission letter without conveying this information to readers while others publish it as part of the article. Only a few pediatric journals provided information on potential conflicts of interest of their editors or reviewers, which is consistent with data from gastroenterology journals. 9 However, we did not search the author instructions and Web sites of the journals specifically for this aspect. Given the crucial role of reviewers and editors in the dissemination of research evidence, journals may consider disclosure of potential conflicts of interest of these additional groups involved in the publication process. Deficiencies in reports from clinical pediatric research have been found in various pediatric subspecialties including pediatric urology, pediatric orthopedics, and drug trials Generic reporting guidelines have been developed by groups of scientists, methodologists, and journal editors with the aim to improve the quality of research reports. There is now first evidence of a positive impact of such guidelines on reporting quality. 7,38 To date, only a few pediatric journals endorse major reporting guidelines. Possible reasons are that, first, for a particular journal, a specific recommendation may not be applicable (eg, QUOROM for a journal not publishing systematic reviews of randomized controlled trials). However, if a journal publishes original clinical research, at least part of the included reporting guidelines are relevant. Second, reporting guidelines may be perceived as an attempt to standardize research reporting. Consequently, journals may choose to abstain from requiring their authors to follow them. However, rather than producing boilerplate manuscripts, many authors would likely use these recommendations in a sensible way to ensure the completeness and accuracy of their manuscripts before submission. There are some limitations to our study. Not all pediatric journals are indexed in the Journal Citation Report, and consequently, our sample may not be representative of all pediatric publications. Next, we took the impact factor as an indicator of the profile of pediatric journals, bearing in mind its multiple weaknesses. 25,39,40 Finally, we did not survey journal editors on whether and how they actually implement and monitor their policies or whether they have any implicit rules or policies that are not reflected by author instructions. CONCLUSIONS We found that current international recommendations aiming to improve publication practice have not yet been implemented widely in pediatric journals. Adherence to the URM and to reporting guidelines such as has a potential to improve the completeness and clarity of biomedical publications. Further, the registration of trials as a prerequisite for publication of ensuing reports and the disclosure of conflicts of interests represent solutions to tackle the problem of publication bias and to enhance transparency. Accepted for Publication: October 10, Correspondence: Joerg J. Meerpohl, MD, German Cochrane Center, Institute of Medical Biometry and Medical Informatics, University Medical Center Freiburg, Stefan- Meier-Strasse 26, Freiburg, Germany Author Contributions: Study concept and design: Meerpohl, Wolff, Niemeyer, and von Elm. Acquisition of data: Meerpohl and Wolff. Analysis and interpretation of data: Meerpohl, Wolff, Antes, and von Elm. Drafting of the manuscript: Meerpohl. Critical revision of the manuscript for important intellectual content: Meerpohl, Wolff, Niemeyer, Antes, and von Elm. Statistical analysis: Meerpohl, Wolff, and von Elm. Administrative, technical, and material support: Antes. Study supervision: Antes and von Elm. Financial Disclosure: None reported. Additional Information: Dr von Elm is one of the authors of the STROBE statement and academic editor of PLoS ONE. Dr Niemeyer is academic editor of Pediatric Blood and Cancer. REFERENCES 1. Dwan K, Altman DG, Arnaiz JA, et al. Systematic review of the empirical evidence of study publication bias and outcome reporting bias. PLoS One. 2008; 3(8):e Bekelman JE, Li Y, Gross CP. Scope and impact of financial conflicts of interest in biomedical research: a systematic review. JAMA. 2003;289(4): Titus SL, Wells JA, Rhoades LJ. Repairing research integrity. Nature. 2008;453 (7198): Hetherington J, Dickersin K, Chalmers I, Meinert CL. Retrospective and prospective identification of unpublished controlled trials: lessons from a survey of obstetricians and pediatricians. Pediatrics. 1989;84(2): Uniform requirements for manuscripts submitted to biomedical journals: writing and editing for biomedical publication. International Committee of Medical Journal Editors Web site. Accessed August 6, Journals following URM. International Committee of Medical Journal Editors Web site. Accessed August 8, Plint AC, Moher D, Morrison A, et al. Does the checklist improve the quality of reports of randomised controlled trials? a systematic review. Med J Aust. 2006;185(5): Smith R. Beyond conflict of interest: transparency is the key. BMJ. 1998;317(7154): Bhargava N, Qureshi J, Vakil N. Funding source and conflict of interest disclosures by authors and editors in gastroenterology specialty journals. Am J Gastroenterol. 2007;102(6): Stelfox HT, Chua G, O Rourke K, Detsky AS. Conflict of interest in the debate over calcium-channel antagonists. N Engl J Med. 1998;338(2): De Angelis C, Drazen JM, Frizelle FA, et al; International Committee of Medical Journal Editors. Clinical trial registration: a statement from the International Committee of Medical Journal Editors. Croat Med J. 2004;45(5): Bonati M, Pandolfini C. Trial registration, the ICMJE statement, and paediatric journals. Arch Dis Child. 2006;91(1): Rivara FP. Registration of clinical trials. Arch Pediatr Adolesc Med. 2005;159(7): Gülmezoglu AM, Pang T, Horton R, Dickersin K. WHO facilitates international collaboration in setting standards for clinical trial registration. Lancet. 2005; 365(9474): Journal Citation Report ISI Web of Knowledge Web site. Accessed August 5, Moher D, Schulz KF, Altman D; Group (Consolidated Standards of Reporting Trials). The statement: revised recommendations for improving the quality of reports of parallel-group randomized trials. JAMA. 2001;285 (15):

5 17. Altman DG, Schulz KF, Moher D, et al; GROUP (Consolidated Standards of Reporting Trials). The revised statement for reporting randomized trials: explanation and elaboration. Ann Intern Med. 2001;134(8): von Elm E, Altman DG, Egger M, Pocock SJ, Gotzsche PC, Vandenbroucke JP; STROBE Initiative. The Strengthening the Reporting of Observational Studies in Epidemiology (STROBE) statement: guidelines for reporting observational studies. Ann Intern Med. 2007;147(8): Vandenbroucke JP, von Elm E, Altman DG, et al; STROBE initiative. Strengthening the Reporting of Observational Studies in Epidemiology (STROBE): explanation and elaboration. Ann Intern Med. 2007;147(8):W Bossuyt PM, Reitsma JB, Bruns DE, et al; Standards for Reporting of Diagnostic Accuracy. Towards complete and accurate reporting of studies of diagnostic accuracy: the STARD Initiative. Ann Intern Med. 2003;138(1): Bossuyt PM, Reitsma JB, Bruns DE, et al; Standards for Reporting of Diagnostic Accuracy. The STARD statement for reporting studies of diagnostic accuracy: explanation and elaboration. Ann Intern Med. 2003;138(1):W Stroup DF, Berlin JA, Morton SC, et al. Meta-analysis of observational studies in epidemiology: a proposal for reporting. Meta-analysis of Observational Studies in Epidemiology (MOOSE) group. JAMA. 2000;283(15): Moher D, Cook DJ, Eastwood S, Olkin I, Rennie D, Stroup DF. Improving the quality of reports of meta-analyses of randomised controlled trials: the QUOROM statement. quality of reporting of meta-analyses. Lancet. 1999;354(9193): Moher D, Liberati A, Tetzlaff J, Altman DG; PRISMA Group. Preferred reporting items for systematic reviews and meta-analyses: the PRISMA statement. BMJ. 2009;339:b2535. doi: /bmj.b Garfield E. The history and meaning of the journal impact factor. JAMA. 2006;295 (1): Hopewell S, Altman DG, Moher D, Schulz KF. Endorsement of the statement by high impact factor medical journals: a survey of journal editors and journal Instructions to Authors. Trials. 2008;9(1): Schriger DL, Arora S, Altman DG. The content of medical journal instructions for authors. Ann Emerg Med. 2006;48(6): , 749. e1-e Matarese V. Relationship between quality and editorial leadership of biomedical research journals: a comparative study of Italian and UK journals. PLoS One. 2008; 3(7):e Bonati M, Impicciatore P, Pandolfini C. Registering clinical trials: register of clinical trials in children must be set up. BMJ. 2000;320(7245): Pandolfini C, Bonati M, Rossi V, et al. The DEC-net European register of paediatric drug therapy trials: contents and context. Eur J Clin Pharmacol. 2008; 64(6): Santoro E, Rossi V, Pandolfini C, Bonati M. DEC-net: the development of the European register of clinical trials on medicines for children. Clin Trials. 2006; 3(4): Sim I, Chan AW, Gulmezoglu AM, Evans T, Pang T. Clinical trial registration: transparency is the watchword. Lancet. 2006;367(9523): Zarin DA, Ide NC, Tse T, Harlan WR, West JC, Lindberg DA. Issues in the registration of clinical trials. JAMA. 2007;297(19): Dulai SK, Slobogean BL, Beauchamp RD, Mulpuri K. A quality assessment of randomized clinical trials in pediatric orthopaedics. J Pediatr Orthop. 2007;27 (5): Raad JM, Bellinger S, McCormick E, Roberts MC, Steele RG. Brief report: reporting practices of methodological information in four journals of pediatric and child psychology. J Pediatr Psychol. 2008;33(7): Standing JF, Khaki ZF, Wong IC. Poor formulation information in published pediatric drug trials. Pediatrics. 2005;116(4):e559-e Welk B, Afshar K, MacNeily AE. Randomized controlled trials in pediatric urology: room for improvement. J Urol. 2006;176(1): , discussion Moher D, Jones A, Lepage L; Group (Consolitdated Standards for Reporting of Trials). Use of the statement and quality of reports of randomized trials: a comparative before-and-after evaluation. JAMA. 2001;285 (15): Dong P, Loh M, Mondry A. The impact factor revisited. Biomed Digit Libr. 2005; 2: Hernán MA. Impact factor: a call to reason. Epidemiology. 2009;20(3): , discussion One time, why I don t know, I asked a donor who tested positive for THC [tetrahydrocannabinol] why he wanted a copy of his test results. He said that he had purchased a product to clean his urine. With proof of a failed test he could get double his money back. Anonymous 272