Assessing adverse effects of interventions in Cochrane reviews: what types of study designs are used?

Article type
Authors
Parra MT1, PorfĂ­rio GJM1, Silva AA1, Martimbianco ALC1, Freitas CG1, Torres MFS1, Reiter M1, Pesavento T1, Riera R1, Torloni MR1, Atallah AN1
1Brazilian Cochrane Centre, Brazil
Abstract
Background:
According to the Cochrane Handbook for Systematic Reviews of Interventions (Loke 2011), every healthcare intervention comes with the risk, great or small, of harmful or adverse effects. To minimize bias in reviews, it is important to assess not only favourable outcomes but also the adverse outcomes of interventions. Although adverse effects are most reliably assessed through randomized trials, it is difficult to observe uncommon, long-term or previously unrecognized effects within planned trials. Adding other study designs can be a strategy to address adverse effects. Depending on the research question, the reviewer may choose to take a narrow or a broad focus to identify those.

Objective:
To assess the types of study designs most frequently chosen by Cochrane authors to evaluate adverse effects of interventions.

Methods:
We identified and read the Methods sections of all systematic reviews and protocols from the Multiple Sclerosis and Rare Diseases of the Central Nervous System Group. We selected those that proposed to assess adverse events and/or safety of interventions. We identified and present the types of study designs that were included by the authors.

Results:
The vast majority (90%) of the 51 studies (38 reviews and 13 protocols) proposed to assess adverse effects in their Methods sections. Almost three-quarters (74%) of these 46 studies included only RCTs to assess adverse outcomes.

Conclusion:
According to this sample of Cochrane reviews, RCTs are the main type of study design used to assess adverse effects of interventions.