Did non-Cochrane systematic reviews of nursing follow the protocols registered in the International Prospective Register of Systematic Reviews (PROSPERO) ?




Oral session: Investigating bias (6)


Thursday 24 October 2019 - 16:00 to 17:30


All authors in correct order:

Hu KY1, Zhang JX2, Zhang T3, Zhang WY4, Suo LL5, Ma YX6, Ma B3
1 Evidence-Based Nursing Center, School of Nursing, Lanzhou University, China
2 Gansu Provincial Hospital, China
3 Evidence-Based Medicine Center, School of Basic Medical Sciences, Lanzhou University, China
4 Evidence Based Social Science Research Center, School of Public Health, Lanzhou University, China
5 School of Foreign Languages and Literatures, Lanzhou University, China
6 Evidence-based Nursing Center, School of Nursing, Lanzhou University, China
Presenting author and contact person

Presenting author:

Kaiyan Hu

Contact person:

Abstract text
Background: one of the most effective means to ensure that reviews are as transparent and bias-free as possible is to explicitly state the study hypotheses and methodology before beginning to conduct the research, thereby ensuring that the results do not influence the hypotheses.

Objectives: this study aimed to compare published non-Cochrane systematic reviews and their associated published protocols from the nursing group of the International Prospective Register of Systematic Reviews (PROSPERO), with the goal of determining what changes were made and how likely these changes were to have resulted in the introduction of bias.

Methods: this study consisted of the retrospective comparison of non-Cochrane reviews and their associated published methodology available from the PROSPERO’s nursing group. We electronically searched for protocols and their corresponding reviews that were 'completed and published' in PROSPERO’s nursing group from inception to 30 September 2018. Two review authors independently identified changes to the different sections of the protocol compared with the completed systematic review and classified the changes as none, partial, or complete.

Results: we identified previously published protocols for 16 reviews, of which all exhibited changes compared with their protocols. Only one review exhibited only a partial change in the search strategy section and the title. The remaining 15 reviews contained at least one section that had undergone a complete change. The greatest variation between reviews and their protocols was in the search strategy, in which all 16 reviews had undergone alteration. Changes made in other sections included expanding/narrowing types of participants included and the addition/deletion of outcome and outcome measures, which all could have contributed to bias in the review.

Conclusions: non-Cochrane reviews that are registered in the nursing group of PROSPERO are less consistent with their previously published protocols, and such changes may introduce bias. PROSPERO should rigorously examine submitted protocols, as this would improve the quality of registration and thus the transparency of the non-Cochrane systematic review process.

Patient or healthcare consumer involvement: by examining the accumulated body of evidence, systematic reviews can provide more reliable results for a range of health care enquiries (e.g. what are the benefits and harms of therapeutic interventions, what is the accuracy of diagnostic tests). To ensure the safety and effectiveness of medical decisions, it is important to promote the reviews as being as transparent and bias-free as possible.