Risk of bias and reporting quality of randomised controlled trials in paediatric pain: a cross-sectional study

Date & Time
Wednesday, September 6, 2023, 11:05 AM - 11:15 AM
Location Name
Victoria
Session Type
Oral presentation
Category
Bias
Oral session
Bias
Speakers
Aidan Tan
Ava Curtin, University of New South Wales
Authors
Curtin A1, Eichholz A2, Champion D1, Jaaniste T1, Ren Z3, Tan A1
1School of Clinical Medicine, Faculty of Medicine and Health, University of New South Wales, Australia
2Pain Research Unit, Department of Pain, Sydney Children’s Hospital, Australia
3NHMRC Clinical Trials Centre, University of Sydney, Australia
Description

Background: There are an increasing number of randomised controlled trials (RCTs) in paediatric pain, but the quality of these studies is unknown. It is imperative that RCTs in paediatric pain are of high quality.
Objectives: To determine the risk of bias and reporting quality of RCTs in paediatric pain.
Methods: This was a cross-sectional study of RCTs in paediatric pain which were published in full text in 2021. Studies were identified by searching MEDLINE, Embase, Cochrane Library, CINAHL, and Scopus. Risk of bias was assessed with the RoB 2 tool and reporting quality was assessed with the CONSORT 2010 statement. Risk of bias was assessed as either high, some concerns or low, and reporting quality was assessed as adequate if ≥75% of items in the reporting guideline were reported. All studies were screened and assessed in duplicate by two independent investigators.
Results: We identified 212 RCTs in paediatric pain published in full text in 2021. Almost three quarters of RCTs (n=154, 73%) were at high risk of bias, and almost all other RCTs (n=57, 27%) had some concerns (see Figure 1). Only one RCT (0.4%) was at low risk of bias. Few RCTs (n=27,13%) were adequately reported and none reported ≥50% of items in the abstract. There were no author characteristics associated with risk of bias. RCTs that were blinded were more likely to be at a lower risk of bias and adequately reported X2 ([1, N=212] =17.7, 0.001; [1, N=212] =6, p=0.014). RCTs that reported a conflict of interest were prospectively registered or reported the use of reporting guidelines were more likely to be adequately reported ([1, N=212] =7.02, p=0.008; [2, N=212] =17.1, 0.001; [2, N=212] =7.56, p=0.023). The reporting of some PedIMMPACT outcomes, specifically symptoms and adverse events and the global judgement of satisfaction with treatment, were associated with reduced risk of bias and improved reporting quality ([1, N=212] =4.26, p=0.039; [1, N=212] =4.59, p=0.032; [1, N=212] =5.88, p=0.015; [1, N=212] =5.78, p=0.016).
Conclusions: Recently published RCTs in paediatric pain are at the high risk of bias and have poor reporting quality. Patient, public, and/or healthcare consumer involvement: None.

Figure 1. Risk of bias of randomised controlled trials.png