MISCONDUCT:how we investigate cases of misconduct (Brian Wansink case)
Cochrane Style Basics: new version released
Details of the CES closure for Christmas and New Year 2018-19
ME Support cover for the festive season
1. EPPR: Supplemental material
We have recently received several queries about the possibility of publishing supplemental data/files for a Cochrane Review. At the current time we do not have the functionality to publish supplemental data/files alongside a Cochrane Review in the CDSR and so we have devised a process that must be followed when there is a need to publish supplemental data/files:
The implementation phase of the policy will end on 31st December 2018, when the policy becomes mandatory. All MEs have now been contacted individually to assess whether any additional assistance is required in order to implement the policy, and to confirm compliance with the policy will be achieved. Pleasecontact ME Support if you have any queries related to the implementation of the policy.
3. When does an amendment need a new citation?
When amending a Cochrane Review or Protocol, an important question to ask is: does the change need a New citation Event?
A New citation Event means that the new version has its own DOI, and the previous version will remain intact and visible forever. A New citation Event will be delivered to PubMed and other external databases.
Any change that does not have a New citation Event will permanently overwrite the existing version on the Cochrane Library. If you do not create a New citation Event, any changes made to the review Abstract will not be reflected in the Abstract on PubMed.
For the integrity of the scientific record, it is important that citation versions do not change significantly. However, creating a New citation Event for every change could confuse people searching for and citing the review.
Examples of changes that would NOT need a New citation Event are:
Minor typographical errors that do not impact the data, findings or interpretation.
A Comment added to the Feedback section AND with no other changes to the Review or Protocol.
Studies removed from 'Studies awaiting classification'.
New search performed, but studies not fully incorporated into the results of the review (i.e. added to ‘Studies awaiting classification’.
In a Protocol: changes to information presented in the background section that will not otherwise affect the eligibility criteria.
4. REVMAN WEB - Automatic tagging of review versions
When the review is being worked on, a version of the review will be automatically tagged every 15 minutes. This will be seen in the History tab in the Dashboard. See below.
This function is for emergency back-up, but it is hoped that in future it could be used to revert to an earlier version (similar to what we currently do in Archie).
The RevMan Web team is currently piloting this but in future the time may change from 15 minutes to a longer interval. The team will also be reviewing how to deal with the number of Automatically tagged versions (before they start to self-delete).
Authors should still be encouraged to check in a draft to Archie after each session.
5. MISCONDUCT: how we investigate cases of misconduct (Brian Wansink case)
In October 2018 concerns were raised by Public Health and Health Systems Network regarding a food marketing researcher named Brian Wansink, as several of his publications were retracted by high-profile journals. In September 2018, three JAMA journals retracted six studies, adding to the seven previous retractions of his work; it is thought that further retractions may follow.
An investigation by Wansink’s institution, Cornell, concluded that Wansink was guilty of the: “misreporting of research data, problematic statistical techniques, failure to properly document and preserve research results, and inappropriate authorship.” A JAMA editorial responded that “This raises concerns of—at the least—endemic sloppiness and—at worst—a disproportionate focus on creating compelling narratives rather than establishing firm observations".
We wanted to find out if there was an impact on any Cochrane Reviews and, if so, to take any relevant actions. These are the questions that we asked:
Are any of the retracted Wansink articles included in a published Cochrane Review?
Do any of the retracted Wansink articles appear in a Cochrane Review (CR) or Protocol?
Do any other articles authored by B Wansink appear in a published or draft CR or Protocol? If yes, have there been concerns raised about any of these articles (e.g. post-publication notices, such as an expression or concern, retraction notices, etc.)?
What did we find?
1. Are any of the retracted Wansink articles included in a published Cochrane Review? No
2. Do any of the retracted Wansink articles appear in a Cochrane Review or Protocol*? Yes = 1 (Listed as an excluded study [Portion, package or tableware size for changing selection and consumption of food, alcohol and tobacco; 10.1002/14651858.CD011045.pub2)]).
3. Are there any other articles authored by B Wansink in a draft or published Cochrane Review or Protocol? If yes, have there been concerns raised about any of these articles? Yes:
5 protocols with additional reference each
2 reviews with 1 additional reference each
1 review with 1 included study (not currently retracted) and 1 additional reference
1 review with 13 included studies (none currently retracted), 9 excluded studies (including 1 retracted), 3 awaiting assessment (none currently retracted; note "Accepted into the review and awaiting full integration"), and 3 additional references
1 review with 1 excluded study (not currently retracted) and 1 awaiting assessment (status unclear, as it is a conference abstract)
1 review with 2 excluded studies (not currently retracted)
What actions have been taken?
Searches and checks have been completed; and meeting set up with affected CRGs/Networks:
4 CRGs: EPOC, Heart, Metabolic and Endocrine Disorders, and Public Health
3 Networks: Long-term Conditions and Ageing Network 1, Public Health Network, and Health Systems, Circulation and Breathing Network
The Public Health CRG added notes to two published Reviews (‘Nutritional labelling for healthier food or non-alcoholic drink purchasing and consumption’and‘Portion, package or tableware size for changing selection and consumption of food, alcohol and tobacco’) to acknowledge that we are aware of the retractions (and that more could follow) but that the two Cochrane Reviews are not affected at present.
Next steps Now (CRG actions)
For the excluded study that has been retracted (Wansink B, Cheney MM. Super bowls: serving bowl size and food consumption. JAMA 2005;293(14):1727‐8. 10.1001/jama.293.14.1727):
add a citation to the retraction notice to study ID in the Review AND add a note to Characteristics Excluded Studies (to state that the study has been retracted and why)
link the retraction notice to the study in CRS (and display in CENTRAL)
2. CRGs to monitor draft protocols and draft updates for citations to Wansink articles.
Next (EPP team actions)
Continue to monitor retractions(two further retractions in October)
Monthly check of Wansink articles in CRs for post-publication notices (plus check for any newly retracted Wansink articles)
If any included articles are retracted, follow Cochrane policy (in draft) for impact on Reviews and CRS/CENTRAL.
If anyone has any questions about this case, or any other cases of misconduct in studies, please contact Bryony Urquhart, Editor, Editorial Policy (email@example.com).
We hope this updated resource will be helpful for authors and editors. Please share with authors, both new and experienced, especially at the start of the review stage or when an update is about to start, or perhaps when sending out reminders to authors about deadlines. We are working with colleagues in the CET to make sure this new version of Cochrane Style Basics is made available in the resources and websites that authors and editors use.
John Hilton (firstname.lastname@example.org) and Elizabeth Royle (email@example.com), on behalf of theCochrane Style Manual Working Group.
7. Details of the Copy-edit Support (CES) closure for Christmas and New Year 2018-19
CES will close on:
Thursday 20 December 2018 at 5 pm GMT CES will reopen on:
Thursday 3 January 2019 at 9 am GMT Last day for assigning:
Thursday 20 December 2018, probably completed by 10 am GMT MEs starting holidays early: Any MEs who will not be available to put permissions in place on Thursday 20 December should let Elizabeth Royle know when they intend to stop work and so that this can be taken into account and appropriate arrangements put in place.
Return dates: Assignments allocated in the week before the CES shut-down will have at least 14 days allowed for completion.
URGENT submissions with publication deadlines before 11 January 2019
Please inform Elizabeth Royle of any submissions that need to be published early in January as soon as possible. State the reason for the urgency and the required publication date.Please mention the urgency clearly and obviously in EVERY message concerning that submission. At present the availability of the CES team is reasonable, but it will reduce as Christmas approaches, and some delays may develop. You are encouraged to send in your submissions as early as possible, as this will help CES to return them to you in a timely manner.
We say farewell to Henning Keinke Andersen who was the Managing Editor for the Cochrane Colorectal Cancer Group. Henning was a longstanding ME and a familiar face at Colloquia and ME meetings over many years. He has taken up a position outside Cochrane and we wish him well.
9. ME Support cover for the festive season
The ME Support team will be working up to 21 December 2018. We will be back at work on 2 January 2019. If you have an urgent query, please include ‘Urgent’ in the subject field of your email firstname.lastname@example.org,and we will endeavour to respond as soon as possible.