Page 3 - The Use of Cannabis and Cannabinoids in Treating Symptoms of Multiple Sclerosis: a Systematic Review of Reviews

Curr Neurol Neurosci Rep (2018) 18:8 Page 3 of 12 8

Identification Records identified through database Additional records identified through other
sources
searching
(n = 0)
(n =263)

Records excluded
(n = 117)
Records after duplicates
Screening (n = 252) 76 removed after reviewing title
41 at title stage (clearly
irrelevant titles)
and abstract

Full-text articles assessed Full-text articles excluded
for eligibility 24 = overview/commentary articles
(n =124)
Eligibility studies of MS and/or cannabinoids
(n =135)
15 = review didn’t cover clinical
12 = Irrelevant
2 = Unable to access full text
9 = Reviews of cannabinoid
mechanisms or endocannabinoids
61 = Did not meet AMSTAR criteria
3 & 6 for quality
1 = Protocol only

Reviews included in
Included quantitative synthesis
(n = 11)

Fig. 1 PRISMA flow diagram for selection of reviews

Two reviewers independently examined titles and abstracts AMSTAR tool documents assessed risk of bias at the review
using Covidence software. Relevant review articles were ob- level.
tained in full, and independently assessed for inclusion in the To be eligible for inclusion, a review needed to meet
review by two reviewers. Reasons for exclusion were docu- criteria 3 and 6 of the AMSTAR tool. These criteria required
mented in Covidence. Inter-reviewer disagreement was re- that a review described a comprehensive search and described
solved by consensus in all cases. the characteristics of the studies included in the review. Those
studies that did not meet criteria 3 and 6 are listed in
Supplementary Table 1 with the other excluded studies.
Assessment of Methodological Quality Details of reported potential conflicts of interests of review
authors were extracted (see Supplementary Table 3). Details
The full text reviews deemed eligible by two reviewers were of AMSTAR scores for individual items are also reported in
assessed for quality by one reviewer and these quality ratings Supplementary Table 3.
were checked by a second reviewer. Methodological quality
ratings described the methodological quality across 11 pre- Grading of Evidence
defined domains for each included review using the
AMSTAR measurement tool to assess the methodological An evidence grade was given to each review using the
quality of systematic reviews [20] (Appendix 2). The Scottish Intercollegiate Guidelines Network (SIGN) grading