Funding characteristics of randomised clinical trials supported by the Swiss National Science Foundation: a retrospective cohort study.

AIMS OF THE STUDY: Failure to publish publicly funded research represents a waste of scarce research resources across medical disciplines and countries. In Switzerland, about 40% of randomised clinical trials (RCTs) supported by the Swiss National Science Foundation (SNSF) were not published. We aimed to describe funding characteristics of published and unpublished RCTs supported by the SNSF, to quantify the amount of money spent for unpublished studies, and to compare our results to a similar study performed in the UK. METHODS: We established a retrospective cohort of RCTs funded by the SNSF up to 2015. For each RCT proposal, two investigators independently identified corresponding publications in electronic databases and trial registries. Teams of two investigators independently extracted details from the original SNSF proposal and, if available, from trial registries or publications. In addition, we surveyed principal investigators about trial costs and additional sources of funding. RESULTS: We included 101 RCTs supported by the SNSF between 1986 and 2015. Most were single-centre RCTs with a median of 138 participants (interquartile range [IQR] 76-400). Overall, 67 (67%) principal investigators responded to our main survey questions. Median total costs per RCT were CHF 428 000 (IQR 282 000-900 000) of which the SNSF provided a median CHF 222 000 (67% of total costs, IQR 40-80%). Most investigators (70%) mentioned additional funding, mainly from their own institution or private foundations. A total of CHF 6.7 million was granted to RCTs that remained unpublished. Funding characteristics were similar to publicly funded trials in the UK. CONCLUSIONS: A third of the total SNSF grant sum spent on healthcare RCTs between 1986 and 2015 did not result in peer-reviewed scientific publications. New SNSF grant schemes might improve publication outcomes but their effectiveness needs to be evaluated.

Systematic review on costs and resource use of randomized clinical trials shows a lack of transparent and comprehensive data.

OBJECTIVES: Randomized clinical trials (RCTs) are costly. We aimed to provide a systematic overview of the available evidence on resource use and costs for RCTs to support budget planning. STUDY DESIGN AND SETTING: We systematically searched MEDLINE, EMBASE, and HealthSTAR from inception until November 30, 2016 without language restrictions. We included any publication reporting empirical data on resource use and costs of RCTs and categorized them depending on whether they reported (i) resource and costs of all aspects at all study stages of an RCT (including conception, planning, preparation, conduct, and all tasks after the last patient has completed the RCT); (ii) on several aspects, (iii) on a single aspect (e.g., recruitment); or (iv) on overall costs for RCTs. Median costs of different recruitment strategies were calculated. Other results (e.g., overall costs) were listed descriptively. All cost data were converted into USD 2017. RESULTS: A total of 56 articles that reported on cost or resource use of RCTs were included. None of the articles provided empirical resource use and cost data for all aspects of an entire RCT. Eight articles presented resource use and cost data on several aspects (e.g., aggregated cost data of different drug development phases, site-specific costs, selected cost components). Thirty-five articles assessed costs of one specific aspect of an RCT (i.e., 30 on recruitment; five others). The median costs per recruited patient were USD 409 (range: USD 41-6,990). Overall costs of an RCT, as provided in 16 articles, ranged from USD 43-103,254 per patient, and USD 0.2-611.5 Mio per RCT but the methodology of gathering these overall estimates remained unclear in 12 out of 16 articles (75%). CONCLUSION: The usefulness of the available empirical evidence on resource use and costs of RCTs is limited. Transparent and comprehensive resource use and cost data are urgently needed to support budget planning for RCTs and help improve sustainability.

Discontinuation and non-publication of randomised clinical trials supported by the main public funding body in Switzerland: a retrospective cohort study.

OBJECTIVE: The Swiss National Science Foundation (SNSF) promotes academic excellence through competitive selection of study proposals and rigorous evaluation of feasibility, but completion status and publication history of SNSF-supported randomised clinical trials (RCTs) remain unclear. The main objectives were to review all healthcare RCTs supported by the SNSF for trial discontinuation and non-publication, to investigate potential risk factors for trial discontinuation due to poor recruitment and non-publication, and to compare findings to other Swiss RCTs not supported by the SNSF. DESIGN: We established a retrospective cohort of all SNSF-supported RCTs for which recruitment and funding had ended in 2015 or earlier. For each RCT, two investigators independently searched corresponding publications in electronic databases. In addition, we approached all principal investigators to ask for additional publications and information about trial discontinuation. Teams of two investigators independently extracted details about study design, recruitment of participants, outcomes, analysis and sample size from the original proposal and, if available, from trial registries and publications. We used multivariable regression analysis to explore potential risk factors associated with discontinuation due to poor recruitment and with non-publication, and to compare our results with data from a previous cohort of Swiss RCTs not supported by the SNSF. RESULTS: We included 101 RCTs supported by the SNSF between 1986 and 2015. Eighty-seven (86%) principal investigators responded to our survey. Overall, 69 (68%) RCTs were completed, 26 (26%) RCTs were prematurely discontinued (all due to slow recruitment) and the completion status remained unclear for 6 (6%) RCTs. For analysing publication status, we excluded 4 RCTs for which follow-up was still ongoing and 9 for which manuscripts were still in preparation. Of the remaining 88 RCTs, 53 (60%) were published as full articles in peer-reviewed journals. Multivariable regression models suggested that discontinued trials were at higher risk for non-publication than completed trials (adjusted OR 7.61; 95% CI 2.44 to 27.09). Compared with other Swiss RCTs, the risk of discontinuation for SNSF-supported RCTs was higher than in industry-initiated RCTs (adjusted OR 3.84; 95% CI 1.68 to 8.74), but not significantly different from investigator-initiated RCTs not supported by the SNSF (adjusted OR 1.05; 95% CI 0.51 to 2.11). We found no evidence that the proportion of discontinued or unpublished RCTs decreased over the last 20 years. CONCLUSIONS: One out of four SNSF-supported RCTs were prematurely discontinued due to slow recruitment, 40% of all included RCTs and 70% of all discontinued RCTs were not published in peer-reviewed journals. There is a case to reconsider how public funding bodies such as the SNSF could improve their feasibility assessment and promote publication of RCTs irrespective of completion status.

Premature trial discontinuation often not accurately reflected in registries: comparison of registry records with publications.

BACKGROUND: One quarter of randomized clinical trials (RCTs) are prematurely discontinued and frequently remain unpublished. Trial registries can document whether a trial is ongoing, suspended, discontinued, or completed and therefore represent an important source for trial status information. The accuracy of this information is unclear. OBJECTIVE: To examine the accuracy of completion status and reasons for discontinuation documented in trial registries as compared to corresponding publications of discontinued RCTs and to investigate potential predictors for accurate trial status information in registries. METHODS: We conducted a cross-sectional study comparing information provided in publications (reference standard) to corresponding registry entries. First, we reviewed publications of RCTs providing information on both discontinuation and registration. We identified eligible publications through systematic searches of MEDLINE and EMBASE (2010-2014) and an international cohort of 1,017 RCTs initiated between 2000 and 2003. Second, pairs of investigators independently and in duplicate extracted data from publications and corresponding registry records. Third, for each discontinued RCT, we compared publication information to registry information. We used multivariable regression to examine whether accurate labeling of trials as discontinued (vs. other status) in the registry was associated with recent initiation of RCT, industry sponsorship, multicenter design, or larger sample size. RESULTS: We identified 173 publications of RCTs that were discontinued due to slow recruitment (55%), harm (16%), futility (11%), benefit (5%), other reasons (3%), or multiple reasons (9%). Trials were registered with clinicaltrials.gov (77%), isrctn.com (14%), or other registries (8%). Of the 173 corresponding registry records, 77 (45%) trials were labeled as discontinued and 57 (33%) provided a reason for discontinuation (of which 53, 93%, provided the same reason as in the publication). Labeling of discontinued trials as discontinued (vs. other label) in corresponding trial registry records improved over time (adjusted odds ratio 1.16 per year, confidence interval 1.04-1.30) and was possibly associated with industry sponsorship (2.01, 0.99-4.07) but unlikely with multicenter status (0.81, 0.32-2.04) or sample size (1.07, 0.89-1.29). CONCLUSIONS: Less than half of published discontinued RCTs were accurately labelled as discontinued in corresponding registry records. One-third of registry records provided a reason for discontinuation. Current trial status information in registries should be viewed with caution.

A systematic review of discontinued trials suggested that most reasons for recruitment failure were preventable.

OBJECTIVE: To collect and classify reported reasons for recruitment failure in discontinued randomized controlled trials (RCTs) and to assess reporting quality. METHODS: We systematically searched MEDLINE and EMBASE (2010-2014) and a previous cohort of RCTs for published RCTs reporting trial discontinuation due to poor recruitment. Teams of two investigators selected eligible RCTs working independently and extracted information using standardized forms. We used an iterative approach to classify reasons for poor recruitment. RESULTS: We included 172 RCTs discontinued due to poor recruitment (including 26 conference abstracts and 63 industry-funded RCTs). Of those, 131 (76%) reported one or more reasons for discontinuation due to poor recruitment. We identified 28 different reasons for recruitment failure; most frequently mentioned were overestimation of prevalence of eligible participants and prejudiced views of recruiters and participants on trial interventions. Few RCTs reported relevant details about the recruitment process such as how eligible participants were identified, the number of patients assessed for eligibility, and who actually recruited participants. CONCLUSION: Our classification could serve as a checklist to assist investigators in the planning of RCTs. Most reasons for recruitment failure seem preventable with a pilot study that applies the planned informed consent procedure.