CORRESPONDENCE TO "CHARACTERISTICS OF REVIEWS PUBLISHED IN NURSING LITERATURE: A METHODOLOGICAL REVIEW" BY TORONTO AND COLLEAGUES
Dear Editor,
We would like to thank Dr Toronto and her colleagues1 who conducted a very important study in examining the methodological quality of integrative and systematic reviews identified from Cumulative Index to Nursing and Allied Health Literature (CINAHL) database. They evaluated the reviews according to the selected items in the Preferred Reporting Items for Systematic Reviews and Meta-Analyses (PRISMA) statement and found that the compliance rates of different items in PRISMA statement ranged from 14% to 77%, with a median of 36%.
We are particularly interested in item 7 in the PRISMA statement on whether the last searched dates were specified. Although the trends of reporting the last searched dates had increased from 5% in 2013 to 28% in 2015, only 14% of the included reviews in Toronto and colleagues1 did so.
Therefore, we would like to take this opportunity to share some recently published work on the topic among systematic reviews that were published in nursing journals identified from the Nursing category in the Journal Citation Reports (Science Edition).2 Of the 202 systematic reviews published in 2014, 172 (85.1%) and 155 (74.8%) provided the last search dates up to the year and month, respectively. Direct comparisons between the Toronto and colleagues study and these results may not be appropriate, as the authors included only systematic reviews and were limited to nursing journals within the Journal Citation Reports whereas Toronto and colleagues included interdisciplinary journals.
Furthermore, PRISMA is often specified as the guideline for systematic reviews but not in integrative or other type of reviews. For example, the Enhancing the QUAlity and Transparency Of health Research (EQUATOR) network recommends only systematic reviews to follow PRISMA guideline, but none of the 376 guidelines were related to integrative reviews (http://www.equator-network.org/reporting-guidelines/).
Toronto and colleagues1 also mentioned in their discussion that several studies used the word "to present" in specifying the time frame. They explained that such wording may be misleading, as it could be years before the actual publication. In fact, it was reported that the median time difference between the last searched and full publication date was 669 days (~1.8 years).2 In addition, Palese et al3 reported that the time taken to publish a research article in nursing journals took an average of 981 days from the end of data collection to either online or hardcopy publication.
By providing this information, we hope readers will get a broader perspective on the currentness of the published systematic reviews and improve the quality of such articles published in nursing journals in the longer run.
-Wilson Tam, PhD
Assistant Professor
-Shawn Y. S. Goh, PhD
Senior Lecturer
Alice Lee Centre for Nursing Studies
National University of Singapore
Singapore
REFERENCES