Preventing research waste
Biomedical research can be wasted at every stage, from choosing a question to sharing usable results. Prevention connects sound research planning, stewardship of data, complete reporting and open science.
Scale and consequences
An estimate discussed in 2022 put wasted biomedical research effort at 85%, with global financial losses exceeding US$200 billion annually. Inefficiencies span strategic planning, design, execution, reporting and publication, and many are preventable. Irrelevant questions and outcomes, failure to use existing evidence or reduce bias, unpublished results, poor reporting and data that cannot be reused all contribute. Failure to inform new studies with systematic reviews has been noted in more than half of studies.
Research waste obstructs evidence-based care and forfeits opportunities to prevent illness and death, treat disease, improve wellbeing and innovate. Redundant studies also expose people and animals unnecessarily to risks and experimental procedures. Reducing wasted money, time and resources can also reduce research projects' carbon footprints.
Non-publication and its causes
Nearly half of clinical studies were estimated to remain unpublished in a 2017 conference account. A 2017 study across medical specialties asked authors why their conference abstracts had not become full journal papers, allowing multiple reasons. Lack of time and resources dominated, placing the main obstacles with authors rather than journals. Publications that are poorly planned may never receive the resources needed to write them. Expert medical writers can help investigators whose lack of writing time prevents reporting.
A study of 4,347 registered interventional trials at 51 US academic centres, with primary completion dates from October 2007 to September 2010, found that 2,458 (56.5%) had been published by July 2014. Among these trials, 1,005 (23%) enrolled more than 100 patients, 1,216 (28%) were double-blind and 2,169 (50%) were phase II–IV. Publication within two years was recorded for 1,245 trials (28.6%), while 952 (21.9%) were published more than 24 months after primary completion. Only 547 of the 4,347 trials (13%) reported results on ClinicalTrials.gov. Dissemination performance was poor and varied noticeably between academic centres.
A 2019 account of 500 large trials concluded that those unreported almost four years or more after completion were unlikely to be published later or post registry results; the evidence lost concerned almost 90,000 participants. Non-reporting undermines participants' expectation that accepting research risks will contribute to science, damages public trust and weakens the published evidence.
Negative results are a strong predictor of non-publication. An academic researcher argued in 2018 that pressure for high-impact papers discourages time spent reporting negative findings likely to reach low-impact journals, and that funders' emphasis on practical applications compounds that disincentive. He saw little practical change despite proposals for journals to guarantee publication of refuting or contradictory perspectives.
Negative, confirmatory and inconclusive results can advance science and prevent redundant work, even when dissemination is difficult. In 2012, proposed routes included dedicated journals, possibly open access, abridged article formats and review focused on scientific validity rather than impact; journals and publishers were still exploring these approaches.
Accountability for reporting
A 2019 editorial proposed treating non-reporting as poor academic practice with institutional responsibility.
- Institutions should suspend investigators who fail to report within a year of trial completion, unless extenuating circumstances prevent reporting.
- Institutions should consider non-reporting in academic promotion decisions.
- Funders should withhold support from researchers who fail to report results.
- Investigators with completed, unreported trials should be barred from new grant applications and have current grants suspended.
- Institutions themselves should be held accountable for reporting, as a stronger incentive.
Funders can also encourage dissemination through payment conditions: a 2022 account described the UK National Institute for Health Research Health Technology Assessment programme withholding the final 10% of a study grant until the full report was available.
Research grounded in existing evidence
Good research practices address relevant, necessary questions, follow appropriate methodological standards such as Good Clinical Practice, and register protocols accurately and promptly. Evidence-based research uses earlier research systematically and transparently to inform studies that answer meaningful questions validly, efficiently and accessibly, including consultation with clinicians and patients about relevant questions and outcomes. An international network promoting that approach was established in 2014.
Fewer than half of phase III randomised trials published in The New England Journal of Medicine, The Lancet and JAMA in 2016–2018 justified the trial through a systematic review. Reported rates in high-impact journals were 33% for orthopaedic trauma in 2015–2018, 54% for urology in 2014–2019 and 22% for ophthalmology and optometry through 2018. A separate estimate put failure to assess new findings against existing evidence at 48.6% of published studies.
Funders and regulators have key roles in implementing evidence-based research. In 2022, funders differed in whether they explicitly required systematic-review justification of new studies; writers preparing grants could advocate the approach and explain it to colleagues and clients.
Writers' actions across the research lifecycle
Four interlinked strategies can reduce waste: good research practices, data stewardship, responsible reporting including transparency and disclosure, and open science. The following actions were recommended for medical writers and communicators.
| Strategy | Recommended actions |
|---|---|
| Good research practices | Advocate scientifically sound, efficient clinical trials; advocate evidence-based research; follow Good Clinical Practice principles; advocate clear, easily implementable protocols; ensure timely, accurate protocol registration; follow good documentation practices. |
| Data stewardship | Advocate data minimisation; educate clients about funding requirements for findable, accessible, interoperable and reusable (FAIR) data management; educate patients and the public about FAIR data management through medical communications. |
| Responsible reporting | Ensure timely public posting of results; write clear, accurate, fit-for-purpose documents; protect personal data through proactive anonymisation to produce redaction-friendly documents; practise lean writing; follow good documentation practice; extend reach to patients and the public through plain-language summaries. |
| Open science | Report scientific information accurately and responsibly; advocate publication of negative results; develop a publication plan; advocate open-access publication; follow reporting guidelines through EQUATOR (Enhancing the QUAlity and Transparency Of health Research); avoid predatory journals; follow good publication practice, including transparency about medical writers' and communicators' involvement. |
Writers may not collect or manage data themselves, but can ensure stewardship is considered in study design and that grant applications address FAIR data-management requirements. In laboratory manuals and protocols, they can advocate collecting only necessary data and samples, while public communications can explain data sharing and address confidentiality and privacy concerns.
Clear, complete, accurate documents facilitate efficient review of manuscripts and regulatory applications. Timely dissemination of favourable and unfavourable results reduces duplicated effort and repeated mistakes, and biomedical publications support evidence-based clinical practice. Ethical, accurate reporting and good publication practice can also promote public trust in science.