Home page of The Cochrane Collaboration

Cochrane Multiple Sclerosis and Rare Diseases of the Central Nervous System Group

Resources for review authors

Getting involed as a Review Author

If you have expertise in some aspect of multiple sclerosis health care, consider joining our Group. Being part of a Cochrane Review Group provides you with the support and resources you need to undertake a systematic review. You would also be very welcome to visit us personally and discuss aspects of your review with skilled review authors. If you cannot visit we are accessible by telephone or email. Please download the form you will need to submit a proposal for a new Cochrane review:

Here you can find and download a variety of information and support materials to help you develop your protocol. 

Training - face-to-face

Contact Cochrane Centres or Review Groups about local workshops and courses in review production. Some of these events are listed on the Cochrane workshops page.

Training - online

  • Open Learning Materials - learn the steps in convenient online modules which supplement the Cochrane Reviewers' Handbook in helping you gain skills and complete your review.

Training resources provided by other organizations:

 

Methods used in reviews

Search strategies

Access to specialised register by authors

Authors working on a review or a protocol are supported by the Editorial Base in the following ways:
The Trials Search Co-ordinator designs a search strategy for CCTRs in consultation with the authors. This personalised search strategy, together with a general search strategy for the Group, is run every three months following receipt of each new edition of the Cochrane Library.
References including references to abstracts are sent out to authors quarterly. Full articles are sent if authors have difficulties in obtaining them locally.

Additional search strategies

To ensure as many relevant studies as possible are identified, review authors may need to conduct additional searches. For example it may be useful to search the bibliographies of full articles received from the Editorial Base and articles found independently. It may also be useful to search other bibliographic databases. Finally it is often useful to contact trialists and drug companies. The results of these additional searches are sent to the editorial base for inclusion in the MS Group Trials Register.

Study selection

The Editorial Base recommends including only randomised controlled trials (RCTs) of any design (e.g. parallel group, crossover and cluster randomisation) or quasi-randomised trials. If trials using a crossover design are found, authors should seek to assess whether this design is appropriate for the question being investigated; if the design is inappropriate the trial should be excluded.
Quasi-randomised trials may be included if it is felt that they may make a useful contribution (e.g. if most available studies are quasi-randomised, excluding them would discard most of the data.) Often studies are described as "randomised" but the randomisation method is not specified; such trials may in fact be quasi-randomised and it may be necessary to contact the author for clarification.
Observational studies (e.g. cohort or case-control studies) should not be included in meta-analyses or contribute to the results of conclusions of reviews.
Wherever possible, the criteria for deciding whether a given study should be included in a Cochrane Review, should be applied independently by at least two authors. Disagreements about eligibility that cannot be resolved between authors are referred to the Contact Editor. Studies that, in the judgement of the authors, are seriously compromised by weaknesses in design, conduct or analysis should also be excluded. The characteristics of such studies must to be reported in the Excluded Studies table.
Cochrane MS Group policy divides inclusion criteria into four sub-categories:
(1) Type of study
(2) Type of participant
(3) Type of intervention
(4) Type of outcome
These are described in the Cochrane Handbook for Systematic Reviews of Interventions (Handbook 20091). With regard outcomes, their measures and time must always  be clearly defined. For example "Number of patients who progressed at 2 years. Progression is defined as a persistent worsening of at least one point in EDSS (Kurtzke 19832), recorded out of relapse and confirmed by a follow-up assessment at six months. However, other definitions of progression given in the original paper could be accepted, including a persistent half-point increase starting from EDSS score > 5.5, as it is often reported in the literature".
Authors should only specify a small number of primary outcomes (maximum three) to be investigated in their review; the more outcomes the greater the chance of spurious results.

Assessment of methodological quality

The review authors who selected studies for inclusion in a Cochrane review should also assess the methodological quality of the studies.
Based on the recommendations reported in the Cochrane Reviewers' Handbook 2009 section 8, we suggest that review authors use the following definitions in the assessment of risk of bias:

Sequence generation 
- Adequate, sequence generation was achieved using a random number table or computer random number generation. Drawing lots, tossing a coin, shuffling cards, and throwing dice are also adequate if performed by an independent adjudicator.
- Inadequate, the sequence generation method is not, or may not be, random. Quasi-randomised studies, those using dates, names, or admittance numbers in order to allocate patients are inadequate and will be excluded for the assessment of benefits but not for the assessment of harms.
- Unclear, the trial is described as randomised but the method of sequence generation was not specified.

Allocation concealment
- Adequate, allocation was controlled by a central and independent randomisation unit, serially numbered, opaque and sealed envelopes, or similar, so that intervention allocations could not have been foreseen in advance of, or during, enrolment.
- Inadequate, if the allocation sequence was known to the investigators who assigned participants or if the study was quasi-randomised.
- Unclear, the trial was described as randomised but the method used to conceal the allocation was not described, so that intervention allocations may have been foreseen in advance of, or during, enrolment.

Blinding
- Adequate, the trial was described as blinded, the parties that were blinded, and the method of blinding was described, so that knowledge of allocation was adequately prevented during the trial. 
- Not performed, the trial was not blinded, so that the allocation was known during the trial.
- Unclear, the trial was described as double blind, but the method of blinding was not described, so that knowledge of allocation was possible during the trial.

Incomplete outcome data
- Adequate, the numbers and reasons for dropouts and withdrawals in all intervention groups were described or if it was specified that there were no dropouts or withdrawals.
 - Inadequate, the number or reasons for dropouts and withdrawals were not described.
- Unclear, the report gave the impression that there had been no dropouts or withdrawals, but this was not specifically stated

Selective outcome reporting
- Adequate, pre-defined, or clinically relevant and reasonably expected outcomes are reported on.
- Inadequate, one or more clinically relevant and reasonably expected outcomes were not reported on; data on these outcomes were likely to have been recorded.
- Unclear, not all pre-defined, or clinically relevant and reasonably expected outcomes are reported on or are not reported fully, or it is unclear whether data on these outcomes were recorded or not.

Other Potential threats to validity
Was the study apparently free of other problems that could put it at a risk of bias?
- Adequate,  the study appears to be free of other sources of bias
- Inadequate, if the study stopped early due to some data-dependent process or had extreme baseline imbalance or has been claimed to have been fraudulent or had some other problem.
- Unclear, insufficient information to asses whether an important risk of bias exists or insufficient rationale or evidence that and identified problem will introduce  bias.

Disagreement between review authors regarding these aspects should be resolved by discussion. The contact Editor is available for assistance regarding this issue.

Data collection

The Editorial Base provides a data collection form together with instructions for correctly identifying study identifiers and citation identifiers. It is recognised that the data collection form may need to be modified to suit the requirements of a particular review. The authors who select and assess the methodological quality of studies, are those who should extract the data from the studies. The data must be extracted independently by each author. Disagreements about extracted data are resolved by subsequent discussion.
In order to allow an "intention-to-treat" analysis, whenever possible the total number of patients assigned to each group should be identified and noted, irrespective of compliance, whether or not they received treatment, or were otherwise excluded from the originally published trial analysis. If complete patient data is not available in publications reporting the trial, Cochrane authors are encouraged to contact the principal investigator of the trial requesting that the information be supplied and other clarifications as appropriate. Involved pharmaceutical companies should also be contacted if necessary. The Editorial Base is available to offer advice on how to extract data from study publications. When the results of a trial have not been published, but the author is confident about quality of the available data, then the data should be extracted and included in the review. Unpublished studies should be identified as such in the review.
Information on adverse events including rare events and long-term complications should always be extracted.

'Characteristics of included studies' table
This should include the following information:
Methods:
Randomisation and allocation concealment,
Design (if not parallel group trial)
Details of blinding and placebo control (if used)
Number of participants who withdrew
Number lost to follow-up
Whether single or multicentre trial; if the latter, number of participating centres
Country/countries where trial was conducted 
Participants:
Number of participants in each treatment group,
Clinical characteristics of participants, including diagnostic criteria, age, sex, disability score, etc.
Interventions:
Description of interventions received by all trial groups
How interventions were delivered, dose, route of administration, duration of treatment
Outcomes:
List of primary and secondary outcomes reported by the trial
Times when outcomes were measured.
Number of adverse events
Notes:
Any other pertinent information, such as trial funding.+

Analysis

Authors are encouraged to find local statistical support, but guidance can given by the statistical editor of the Cochrane MS Group, if necessary. Statistical analyses should in any event be conducted in compliance with the revised Section 9 of the Cochrane Handbook for Systematic Reviews of Interventions, which gives details of how to apply methods and covers a range of other important topics. The guidance provided by the Handbook is intended to encourage quality and consistency across Reviews.


Summary statistics: continuous outcomes (Section 9.2.3 of Cochrane Handbook for Systematic Reviews of Interventions).
The weighted mean difference should be used if they outcomes are measured in the same way in all trials considered. The standardised mean difference (SMD) can be used to combine trials that measure the same outcome using different methods (e.g. two different scoring systems to measure disability).
All continuous outcome data require a mean and standard deviation for each allocation group from each trial. If standard deviations are not reported, it may be possible to calculate them from data given in the paper: the Cochrane Handbook for Systematic Reviews of Interventions (section 9.4.5.1) explains how this can be done.


Other types of data
The ways that other types of data (such as dichotomous, ordinal, counts and rates, time to event outcome) should be analysed are described in section 9.2 of the Cochrane Handbook for Systematic Reviews of Interventions.


Intention-to-treat analysis (Section 16.2 of Cochrane Handbook for Systematic Reviews of Interventions)
Strict intention-to-treat (ITT) analysis means:
1) All participants are included in the analysis
2) All participants are analysed in the group to which they were allocated, regardless of whether or not they received the allocated intervention.
Authors should attempt to fulfil both criteria for ITT analysis for all trials. If some participants were not analysed in the group to which they were randomised, there may be sufficient information in the trial report to restore them to the correct group. The authors of the original study may be able to provide the missing information.
Where participants are lost to follow up or withdraw from the study, and their data are not available, then the primary analysis should be an "available case" analysis (i.e. analysis of patients who completed the trial for the outcome in question and who form the denominator for the analysis). However authors must always keep in mind the potential for bias arising from non-ITT analyses  seek to assess the effects of the missing data. The effects of the missing data can be explored by sensitivity analyses (for example, assuming the worst case scenario in which all missing participants have a certain outcome). For  other scenarios  possible scenarios see the Cochrane Handbook for Systematic Reviews of Interventions, sections 9.7).
A fixed effects analysis may be used if all of the trials are sufficiently similar, in which case it is reasonable to assume that the underlying effect size is the same for all trials. If this assumption is not reasonable, a random effects analysis may be used to obtain an overall indication of the results of the studies. If the trials are too dissimilar it may not be possible to meaningfully combine their results.

Measuring heterogeneity
The I2 statistic should be used to measure heterogeneity (Section 9.5.2 of Cochrane Handbook for Systematic Reviews of Interventions)
As a general rule, if I2 exceeds 50%, the heterogeneity is substantial and the authors should draw attention to it and discuss possible causes (Higgins 20023).
The implication of substantial heterogeneity is that the results of the considered trials differ, for the outcome being considered, more than would be expected by chance. Thus some factor other than the trial intervention per se is influencing the result. An example of such a factor could be drug dose; if trials use different doses, their results may be different. It is therefore important to investigate factors that may be responsible for heterogeneity.
The possibility of major heterogeneity may be considered in the Protocol which should specify which subgroup analyses to perform in the event of major heterogeneity being found.


If pre-specified subgroup analyses do not explain the heterogeneity, an overall summary may be obtained from a random effects analysis (see Section 9.4.2 Cochrane Handbook for Systematic Reviews of Interventions). It is also possible that a meta analysis of any kind is not possible Subgroup analyses
Only a small number of the most important clinical outcomes should be subject to subgroup analyses, and these must always be pre-specified in the study protocol. Only pre-specified subgroup analyses should be conducted. Subgroup analyses may produce misleading results which must therefore be interpreted with caution. Such results should be used principally for hypothesis generation (see Section 9.6.3 Cochrane Handbook for Systematic Reviews of Interventions).
If there is evidence that the intervention has different effects in different subgroups, results for subgroups rather should be reported rather than the results of the overall analysis. It is essential to keep subgroup analyses to a minimum to avoid generating spurious results. Each subgroup analysis must have a clear rationale, which should be explained in the protocol.

References

(1) Higgins JPT, Green S (editors). Cochrane Handbook for Systematic Reviews of Interventions Version 5.0.2 [updated September 2009]. The Cochrane Collaboration, 2009. Available from www.cochrane-handbook.org.

(2) Kurtzke JF. Rating neurologic impairment in multiple sclerosis: an expanded disability status scale (EDSS). Neurology 1983;33(11):1444-52. [ MEDLINE: 84040263]

(3) Higgins JPT, Thompson, SG. Quantifying heterogeneity in a meta-analysis. Statistics in Medicine 2002; 21:1539-58.

Reporting of reviews

For the format and presentation of reviews, authors are referred to the Cochrane Handbook for Systematic Reviews of Interventions. The Cochrane MS Group's Editorial Team have not produced any additional material. With regard to interpretation of results this is left to the review authors, who are free to comment on the relevance of the results to clinical practice, and to suggest the need for further research. 
Authors are encouraged to cite other reviews performed by Cochrane MS Group authors where these are relevant.

Editorial Process

Titles

Proposals for review titles are submitted to the Cochrane MS Group on a form provided by the MS Managing Editor. The Editorial Team recommends that at least two authors be involved in writing a review. Authors cannot submit more than two review titles at the same time. When proposing a review title, authors must also indicate a realistic date for submission of the review protocol, which must be not  more than six months after registration of the title of the review.
Review titles are examined by the Editorial Team which evaluates three main aspects:
(1) Research questions and available randomized controlled trial.

2) The relevance of the topic to people with MS, their relatives and caregivers, to clinicians and professionals involved in MS.
(3) Whether a similar review has been prepared or is in preparation.
Titles which have been approved by the MS Editorial Team are circulated to all other Cochrane Review Groups to avoid duplication.  Decisions on review title acceptance/rejection will be communicated to the contact author by e-mail.
A Contact Editor (chosen among the seven Cochrane MS Editors) will be appointed to support the review team during the  preparation of  their protocol and  review. 

Protocols

After having registered the review title, authors are strongly invited to visit the Cochrane Collaboration website where the following can be downloaded:
- General information on the Cochrane Collaboration and the Cochrane MS Group
- Hard copy of the Cochrane Handbook for Systematic Reviews of Interventions
- Open learning material for reviewers
- Installation instructions for RevMan 5
- RevMan User Guide
- Cochrane Style Guidelines
- Document setting out the MS Group's editorial process
- An example of an Cochrane MS Group review published in the Cochrane Library

Authors are also encouraged to attend a protocol training day and a RevMan 5 training day at their local Cochrane Centre before beginning their review protocol.
Authors are expected to submit their review protocol within six a month period. However, if they experience problems further time will be conceded. After expiry of this period, the title will be de-registered and made available to other authors.


The review protocol must be prepared in RevMan 5 and preferably submitted trough Archie, the Cochrane Collaboration's central server for managing documents and contact details. Authors obtain access to Archie from the time the title is registered to check in or out their working versions of the protocol or the review. Authors are advised to carefully proof-read their protocol before submitting it.  It is the responsibility of the contact author to ensure that the protocol adheres to the formatting rules specified in Cochrane Style Guidelines. The Managing Editor checks that the draft protocol is ready to go through the editorial process. The Trials Search Co-ordinator checks the search strategy and reference sections. If the protocol is not ready to be sent to peer referees, the contact author of the review is invited to work with the Editorial base team until the protocol is ready for editorial comments.


The Managing Editor will send the draft protocol to the peer referees who are selected in turn from our pool of referees unless the review topic falls within the expertise of a particular referee. The peer referees are sent a checklist to help structure their evaluation of the protocol; they are expected to send back their evaluation within 20 days of receipt of the protocol. If they are unable to send in their evaluation within 20 days, other referees will be appointed.
The Managing Editor will send to the Contact Editor (who is responsible for ensuring that the authors respond appropriately to the referee's comments) and the contact author of the review a letter in which the comments from referees are listed. The refereeing process is not anonymous. Comments and criticisms by external peer referees and the MS Editorial Team are sent to the authors with the names of their originators attached.

Review authors are asked to resubmit the revised version of the protocol to the Editorial base within a month, with a cover letter providing detailed responses to each of the referees' comments and specifying whether and how the text of the protocol has been changed in response to each comment. The cover letter and the revised protocol are sent  to the referees for approval. The entire refereeing process is expected to be completed within three months.
Authors are welcome to visit the editorial base at any stage if they wish to discuss problems with other experienced authors.
Once the protocol is approved it is published in the MS Module of the Cochrane Database of Systematic Reviews of The Cochrane Library.

In the event that the review topic is concerned with an area being covered by another Cochrane Review Group, that Group will be contacted and asked if they would be willing to collaborate with the authors/editorial base with a view to providing support.

Reviews

The Trials Search Co-ordinator is available to assist authors throughout the preparation of their review, including help with developing search strategies and providing hard copies of full articles if they have difficulties in obtaining them locally.

The review refereeing process is the same as that for the protocol: the peer referees and Contact Editor who refereed the protocol are also asked to examine the review. If any of these persons is unavailable to referee the review, the Managing Editor will appoint substitutes.
The peer referees are sent a checklist to help structure their evaluation of the review and  are expected to send their comments/suggestions within 30 days of receipt of the review. If they are unable to send their comments within that time, other peer referees will be appointed by the Managing Editor.

The Consumer Co-ordinator (who is a member of the Editorial base) will review/prepare the plain language summary if the authors have not prepared one, or will comment on/revise the plain language summary prepared by the authors. The Consumer Co-ordinator will also send the review to two consumers of the MS-CRG and will collect and summarise their feedback for sending on to Managing Editor. The final decision on plain language summary content rests with the authors (although it must adhere to the Cochrane Handbook for Systematic Reviews of Interventions).


The Managing Editor will forward to the contact author and to the Contact Editor: all review feedback, the drafted/altered plain language summary, and a list of the main points that must be addressed before the review can be published. The contact author is responsible for ensuring that all this material is sent on to all review co-authors.
The contact author will respond to the feedback via the editorial base within 20 days. This response must include replies to each of the issues raised by the peer referees.
In the event of  disagreement between the contact author and co-authors during the preparation of the review, the problem will be referred to the Contact Editor. If the problem cannot be resolved, it will than be referred to the Co-ordinating Editor. If the Co-ordinating Editor is the Contact Editor for the review, the problem will be referred to the other editors.


Once a review has been approved, the Managing Editor will inform the contact author who will be asked to approve the proof held in Archie for publication and arrange for the Licence for publication and the Declaration of interest forms to be signed by all co-authors. Copy editing of the review will be carried out by the primary author and editorial base before publication in the Cochrane Database of Systematic Reviews.
Since the refereeing process generally takes not less than six months, review authors are urged to submit through Archie a draft of their review within 12-18 months from the date of the published protocol. Protocols that have not been converted into full review within two years are withdrawn, except  under extraordinary circumstances sanctioned by the Editorial Team.

Updating

The Trials Search Co-ordinator runs the Group's search strategy twice a year. When new trial reports are identified, they are included in the Group's Specialised Register of Controlled Trials and notified to the Contact author. Authors are strongly encouraged to update their reviews and to take into account pertinent new data, if available, starting from the date the search was last run. Where no new evidence has been identified, authors have to update the date of  their search strategy and state that no new evidence has been found.
An updated review is submitted for editorial approval in the same way as a new review.
A copy of the updated review is also sent to the MS Consumer Co-ordinator so that an updated draft plain language summary can be prepared, or the plain language summary prepared by the authors can be examined. Once approved by the authors the new plain language summary will be included in the updated review.
Copy editing of the updated review is carried out by the contact author of the review and the editorial base  before publication in the Cochrane Database of Systematic Reviews. If the update
requires a new citation, all the review authors must submit to the Managing Editor an updated, signed Licence for publication and Declaration of Interest forms. If a review is not updated after two years, three additional months are given to authors to complete the updating. After this extension, editors will decide whether the review should be given to a new author team for updating.

Criticism

Comments and criticisms received on published Cochrane MS Group reviews are dealt with according to the standard procedure defined centrally by The Cochrane Collaboration. The Group's Criticism Editor is responsible for organising and summarising all post-registration criticisms. These are forwarded to the Contact author, via the editorial base, for examination and response.

Plain Language summary

The plain language summary aims to summarise the results of the review in a style understandable by consumers and laypersons.
Review authors may draft the plain language summary themselves (ideally with consumer input) or ask the Cochrane MS Group to prepare it.

 

Page last updated: Mon 18th Mar 2013 12:42:16 CET