Are awaiting classification studies properly managed by Cochrane Reviews?

Article type
Authors
Martimbianco AL1, Pacheco RLP2, Latorraca CDOC3, Ferreira RES1, Riera R4
1Centro Universitario São Camilo
2Centro Universitario São Camilo and Universidade Federal de Sao Paulo
3Universidade Federal de São Paulo
4Universidade Federal de Sao Paulo and Centre of Health Technology Assessment at Hospital Sírio-Libanês
Abstract
Background: studies 'awaiting classification' are potentially eligible studies that cannot be assessed for inclusion or exclusion in a systematic review (SR) due to insufficient information. However, these studies may have an impact on the review’s conclusion, and Cochrane authors should be aware about how to deal adequately with them. The Cochrane Handbook for Systematic Reviews of Interventions recommends that all reasonable attempts to obtain information must be made before studies are definitively categorized as 'awaiting classification'. Furthermore, it is appropriate to describe the study details in the 'Characteristics of studies awaiting classification' table, and to mention those that have the potential to influence the results.

Objectives: to describe how studies 'awaiting classification' have been managed by Cochrane Review authors.

Methods: we analyzed all Cochrane SRs of interventions published in the Cochrane Library; 2019, Issue 1, to identify:
- number of studies and reasons for being listed as 'awaiting classification';
- details from corresponding review authors, as missing information requested and date of contact;
- description of the awaiting studies on PRISMA flow diagram;
- description of the potential impact (or not) of awaiting studies in Discussion or Conclusion sections;
- description in the 'Characteristics of studies awaiting classification' table;
- number of studies awaiting classification included after review updating (when applicable), and if their inclusion changes the review’s conclusion.

Results: Issue 1 of 2019 included 45 SRs, 25 (55.5%), including six updated reviews, presented 107 studies awaiting classification (range: 1 to 24). We describe review authors' reasons for classifying the study as 'awaiting' in Table 1. Twelve of 25 SRs (48%) reported attempts to contact the study authors: two SRs described the date of contact and another two reported that it was not possible to find authors' contact details. One study author answered the email. One of the updated SRs included one study previously classified 'awaiting' and there was no conclusion change; another updated SR excluded one study awaiting assessment since 2009, because it was not randomized. Regarding the report of 'awaiting studies' in a SR, 84% mentioned it in PRISMA diagram and 88% presented the 'Characteristics of studies awaiting classification' table with reasons. Only one SR described the potential impact of studies awaiting classification on the conclusions.

Conclusions: we identified an inconsistency in how Cochrane Review authors deal with studies awaiting classification. A checklist of the key points to manage with these studies will increase the reliability of SRs, especially if the review authors detect potential influence on the review conclusion.

Patient or healthcare consumer involvement: these findings reinforce the need to improve the transparency and reliability of a SR, leading to better evidence-based practice that affects patients and healthcare consumers.