A new analysis highlights Australian research positives and areas with room for improvement.
A comprehensive report on recent Australian clinical trial activity has revealed that research on some conditions falls short relative to burden of disease.
And while Australian clinical trial activity compares well with other OECD countries, concerns regarding a disappointingly low rate of data sharing and issues related to inadequate collection of sociodemographic data were also highlighted.
Conducted by researchers from the University of Sydney’s NHMRC Clinical Trials Centre, the investigation examined Australian trials registered on the ANZCTR or clinicaltrials.gov from 2006-2020. The authors published an overview in a perspective for the MJA, with the full report available online.
Australian Institute of Health and Welfare data on burden of disease (disability-adjusted life years, or DALY) was also used for the analysis.
Cancer, which has the highest burden of disease at around 18% of total burden of disease, had the highest number of trials (20% of total) and second highest number of participants (around 16% of total), outnumbered only by cardiovascular disease.
The biggest losers were musculoskeletal conditions. The analysis found that the number of musculoskeletal trials accounted for around 9% of all trials, while the number of participants in these trials accounted for around 5% of total trial participants.
Meanwhile, the burden of disease for musculoskeletal conditions is 13% of total burden, with the authors suggesting that compared with burden of disease, musculoskeletal trial activity is lower than would be expected.
Other conditions potentially underrepresented in clinical trials relative to burden of disease are mental health and neurological, at least in terms number of participants. In terms of number of trials, the pattern is consistent with burden of disease for those conditions.
Cardiovascular disease, which comprises around 14% of total burden of disease, had only 12% of all trials. However, the number of total participants was higher and accounted for the highest proportion of trial participants among all conditions (20%).
The researchers also considered types of trials conducted. Most (75%) looked at treatment strategies, of which drug trials were the majority (45% of all trials; 60% of treatment trials). Other treatments were devices (11% of all trials), surgery (4%) and ‘other’ (20%) including exercise, physiotherapy, cognitive therapy, special diets, herbal medicines and web-based treatments.
There were relatively few trials looking at diagnosis/prognosis and early detection/screening, with each comprising about 2% of all trials.
In all, there were almost 20,000 trials and almost 9 million participants (which includes overseas participants in some trials) over the study period.
The authors noted that clinical trial activity in Australia rates quite well compared with other OECD countries, with higher per capita involvement than the US, UK, France and Germany, but lower than Belgium, Denmark and the Netherlands.
The authors highlighted several areas for improvement, particularly data sharing and the opportunity to contribute to the open science movement.
“Since October 2018, the International Committee of Medical Journal Editors and the World Health Organization (WHO) have required trialists to state whether they plan to share de-identified data at the time of registration,” the authors wrote in the MJA.
“Since then, only 485 (23%) of 2143 Australian trials have indicated the intention to share data. Internationally, rates are similar, with 23% of trials on the WHO trials platform intending to share data.”
The authors also called for improved reporting on the inclusion of priority and underrepresented groups, such as Aboriginal and Torres Strait Islander peoples, older adults, and people with co-morbid illness or rare diseases.
They outlined measures in development which aim to address this with standardised and consistent data capture for sex, gender, cultural, ethnic and linguistic diversity information.
