Randomised controlled trials in Irish Health Research: a systematic review

Conference:

SAPC ASM 2018 - London

Talk Code:

P2.27

Presenter:

Barbara Clyne

Twitter:

@BarbaraClyne

Co-authors:

Boland F1, Murphy N1, Murphy E2, Moriarty F1, Wallace E1, Smith SM1, Devane D3, Murphy A2, Fahey T1

Author institutions:

1 HRB Centre for Primary Care Research, Department of General Practice, Royal College of Surgeons in Ireland (RCSI), Dublin, Republic of Ireland 2 HRB Primary Care Clinical Trials Network Ireland, Department of General Practice, NUI Galway 3 HRB-Trials Methodology Research Network, School of Nursing & Midwifery, NUI Galway

Problem

Randomised controlled trials (RCTs), when appropriately designed, conducted, and reported, represent the gold standard in evaluating the effectiveness of healthcare interventions. Despite the advent of reporting guidelines, the reporting of some RCTs remains suboptimal. Over the last few decades, there has been significant investment in improving health research in Ireland, highlighting a commitment to developing research capacity, coupled with increased funding of population health and health services research. Little is known about the extent of trial based research or the quality of RCTs conducted in Ireland.

Approach

A systematic review of RCTs conducted in the Republic of Ireland was undertaken to describe their characteristics and assess overall quality of reporting including methodological quality, trial registration and use of standardised reporting guidelines. RCTs evaluating effects of health interventions, which took place in Republic of Ireland and included a majority (≥80%) Irish population were identified using PubMed, Embase, Scopus, CINAHL, PsychINFO and the Cochrane Central Register of Controlled Trials, up to the end of 2015. Methodological quality was assessed using the Cochrane Collaboration’s RCT risk of bias tool.

Findings

From 15,794 abstracts identified, 3,455 full texts were reviewed. Six hundred and forty-two RCTs published between 1968 and 2015 were included. Between 1968 and 1989, 13% of RCTs were published, 26% were published from 1990-1999, 30% were published from 2001 -2009 and 30% from 2010 -2015. Preliminary results indicate that over 80% of studies were parallel designs. The majority of studies (45%) were classified as drug interventions and less than 10% of all identified trials were conducted in primary care. Twenty-one percent of RCTs published since 2005 (year ICMJE introduced mandatory clinical trial registration) provided a trial registration number, although year on year increases were observed. RCTs published since 1996 (year CONSORT statement was developed), showed similar year on year increases in use of standardised reporting guidelines, however, only 11% of RCTs published since 1996 made specific reference to a standardised reporting guideline. Risk of selection bias was high or unclear for over 60% of studies and risk of performance bias was high for 50% of studies.

Consequences

Although standards are in place to improve prospective registration and transparency in reporting of RCTs, low rates of trial registration and use of standardised reporting guidelines were reported in this study. Further promotion and implementation of trial registration and standardised reporting guidelines is needed.

Submitted by:

Barbara Clyne

Funding acknowledgement:

This research was funded by the Health Research Board (HRB) in Ireland through grant no. HRC/2014/1