Unique ID issued by UMIN | UMIN000018189 |
---|---|
Receipt number | R000021062 |
Scientific Title | Adherence to protocol registration of systematic reviews published in high-impact-factor journals: meta-epidemiological study |
Date of disclosure of the study information | 2015/07/07 |
Last modified on | 2017/02/28 23:38:06 |
Adherence to protocol registration of systematic reviews published in high-impact-factor journals: meta-epidemiological study
Adherence to protocol registration of systematic reviews published in high-impact-factor journals: meta-epidemiological study
Adherence to protocol registration of systematic reviews published in high-impact-factor journals: meta-epidemiological study
Adherence to protocol registration of systematic reviews published in high-impact-factor journals: meta-epidemiological study
Japan |
[Criteria for considering studies for this review]
We will include any systematic reviews or meta-analysis for interventions. Studies will be excluded if the reviews are for diagnostic test accuracy, non-randomized studies or meta-epidemiology. Eligibility will be assessed by two authors (H.T and Y.T.) independently. Differences regarding eligibility will be resolved by consensus and with the help of the senior author (Y.K.).
[Data collection and analysis]
[Selection of studies]
We will start with the review of titles of the articles identified in the initial research and proceed to review of abstracts, whose titles might be of relevance, and finally identify eligible studies on the basis of their full text.
Not applicable |
Malignancy
NO
[Review question]
This review will be performed to evaluate the quality of systematic reviews in high-impact-factor journals. We will investigate adherence to protocol registration of the reviews.
Others
[Rationale]
Thanks to dissemination of method of systematic reviews, anyone can tackle systematic reviews. There are abundant of systematic reviews for healthcare intervention, and there are duplication or overlap of reviews. Duplication of reviews may lead alpha error.
In addition, drug company come to use systematic reviews of the effects of their product as marketing tools because clinicians put much values on systematic reviews. However, there are concerns of publication bias or selective outcome reporting of systematic reviews.
Protocol registration was disseminated as the PRISMSA statement in 2009 to avoid duplication and publication bias. In order to detect potential alpha error or selective outcome reporting, we will investigate adherence to protocol registration.
[Primary Outcome]
Adherence to protocol registration
[Secondary Outcome]
1. Number of systematic reviews that refer directly to PRISMA statement
2. Adherence to protocol registration among systematic reviews that refer directly to the PRISMA statement
3. Publication bias in each registered and unregistered systematic reviews
Others,meta-analysis etc
Not applicable |
Not applicable |
Male and Female
[inclusion criteria]
the 10 journals classified by the Journal Citation Reports 2013 as general internal medicine journal associated with the highest impact factors
[Data extraction and management]
Two authors (H.T. and Y.T.) will review titles and abstracts identified by the searches independently. They will independently detect protocol registration information from abstract or full text if needed. In addition, each author will check reference of the PRISMA statement.. Disagreements will be resolved by consensus and with the help of the senior investigator (Y.K.).
[Searches]
[Search method]
We will select the 10 journals classified by the Journal Citation Reports 2013 as general internal medicine journal associated with the highest impact factors; we will exclude the Cochrane Database of Systematic Reviews because protocol registration is strictly performed in articles published in the database. We will search MEDLINE (source, PubMed, between August 1st 2009 and June 30th 2015) using the journal names and "Search Strategy Used to Create the Systematic Reviews Subset on PubMed" last modified February 2015; the search will be not subject to language restrictions. We will select systematic reviews for any randomized intervention. We will exclude systematic reviews for diagnostic test accuracy and systematic reviews for non-randomized observational studies
the Cochrane Database of Systematic Reviews
1
1st name | |
Middle name | |
Last name | TSUJIMOTO Hiraku |
Hyogo Prefectural Amagasaki General Medical Center
Hospital Care Research Unit
Higashinaniwa 2-17-77, Amagasaki, Hyogo, Japan
+816-6480-7000
hira.to.ber3598@gmail.com
1st name | |
Middle name | |
Last name | TSUJIMOTO Hiraku |
Hyogo Prefectural Amagasaki General Medical Center
Hospital Care Research Unit
Higashinaniwa 2-17-77, Amagasaki, Hyogo, Japan
+816-6480-7000
hira.to.ber3598@gmail.com
Hyogo Prefectural Amagasaki General Medical Center
Hyogo Prefectural Amagasaki General Medical Center
Local Government
NO
2015 | Year | 07 | Month | 07 | Day |
Published
http://www.jclinepi.com/article/S0895-4356(17)30181-6/pdf
Completed
2015 | Year | 07 | Month | 06 | Day |
2015 | Year | 07 | Month | 07 | Day |
[Analysis]
[Strategy for data synthesis]
First, we examined rate of systematic reviews that refer directly to protocol registration among all the identified reviews.
Second, we examined adherence rate of protocol registration among systematic reviews that refer directly to the PRISMA statement.
Finally, we plot estimates of study findings (the log odds ratio) against their standard error in order to investigate publication bias in each registered or unregistered systematic reviews.
In addition, the Egger test will be used to examine publication bias.
[Analysis of subgroups or subsets]
To identify factors associated with the adherences to protocol registration, we will examine the following a priori defined possible predictors using univariate analysis: year of publication, sources of trial funding, journal, journal impact factor, region in which the study was led, sample size and citation of the PRISMA statement. The important factors strongly associated with protocol registration (P< 0.2, by univariate analysis) will be further analyzed by logistic regression.
2015 | Year | 07 | Month | 04 | Day |
2017 | Year | 02 | Month | 28 | Day |
Value
https://center6.umin.ac.jp/cgi-open-bin/ctr_e/ctr_view.cgi?recptno=R000021062