Preprints.org logo
Preprint
Short Note

This version is not peer-reviewed.

Failure to Post Clinical Trial Results: A Serious Threat to Public Trust

Submitted:

21 October 2026

Posted:

22 October 2026

You are already at the latest version

Abstract
A large proportion of registered clinical trials fail to post their data even after a year of completion. No causal factor appears to explain this failure other than deliberate selective reporting in commercial interest. Since regulators, journal editors and academics appear to have failed to arrest this, citizen science needs to take active steps and challenge the concerned researchers, regulators and journal editors too.
Keywords: 
clinical trials
;  
data transparency
;  
drug approval
;  
citizen science
Subject: 
Medicine and Pharmacology  -   Clinical Medicine
A vast majority of clinical trials do not post their data even after a year or more from the date of completion. Our exploration of clinicaltrials.gov for six diseases revealed that between two third to three forth of interventional trials completed before March 2025, (so that they have more than one year time to post their data) had not posted their results (table 1). It is likely that clinical researchers think that for drugs that are not ultimately approved, the trial results may not be important, although the regulations do not explicitly say so. Similarly for trials with smaller enrolment sizes, the results might be perceived unimportant. But such perception does not seem to adequately explain the failure to post data. The drugs approved and aggressively marketed had only slightly better record of posting data. The enrolment size explains only a small part of the failure (table 1). It is therefore difficult to understand the causes of large-scale failure to post results without considering the possibility of deliberate selective reporting.
The situation might be even worse for clinical trial registration in some other countries1. Even more concerning is the fact that papers often get published in high impact journals2 or public declarations are made3 without posting the results. So, the readers do not have the option to quickly check the claims with original data. Despite early concerns raised4, the proportion of reported trials seems to have declined instead of improving. The multiple players in the game include (i) the regulators (ii) the pharma companies (iii) journal editors (iv) the academics without a conflict of interest and (v) the ultimate consumer. We argue here that the failure to post data does not serve the interests of any of them, unless hiding the results is intentional and misleading people is the purpose. Let us examine the possible roles of all the stake holders.
(i)
The regulators: While at times, the regulators can conditionally exempt or allow late posting of data5, it is expected to be an exception rather than the rule. The large-scale reluctance to post data clearly indicates failure of the regulatory bodies. Although posting the data is a mandate, there is no clarity on what action needs to be taken on failure to post. The fact appears to be that it is possible to get away without making the data transparent and that’s why a large proportion of clinical trials are able to keep their outcomes hidden.
(ii)
The profit makers: The pharma companies are known to have commercial interests and that’s why regulating authorities are needed to ensure fairness of clinical trials. Posting clinical trial data should be in the interest of pharma companies since not posting the data enables and strengthens the accusation of deliberate reporting bias, in order to mislead the reader6-9. The PIs, institutions and funder of clinical trials need to ensure transparency and easy availability of the results in order to avoid such accusations, unless selective reporting and misleading the reader is intentional. It can be understood if commercial interests induce selective reporting of convenient results and hiding the inconvenient ones. The pharma companies need to take extra care to demonstrate that this is not the case. Posting all the data after completion is an essential part of the required care.
(iii)
Research journals: Journals publishing clinical trial-based papers could have ensured that at least for the papers they published, data are posted on the registration site. The fact that papers are published or press notes are released implies that data organization and analysis has progressed adequately. At this stage there is no reason to delay or avoid posting. But journal editors’ role in preventing selective reporting is bound to be limited. Only those trials whose results are convenient are more likely to be used in writing papers. Journal editors have no control on trials whose results are not published.
(iv)
Academics in basic science: There is always a community of researchers without direct conflict of interests. It is interesting that hardly anyone from this community raises a concern about non-transparency of clinical trial data. This presumably reflects the altered motivation of researchers, brought about by the research evaluation systems. While publishing in high impact journals is valued, raising concerns on published work or scientific sleuthing has little value in career evaluation. Furthermore, the pharma sector is a potential research funder for them. Therefore, career minded academics are not expected to waste their time in raising such concerns. It is not surprising that most scientific sleuths work outside academia. But currently most of the toolkits of the sleuths address image manipulation and plagiarism. Data transparency has not taken central stage in this community as yet.
(v)
If regulators, journal editors, researchers in academia are unable to ensure data transparency citizens interested in their health rights need to do this. Citizens need to question the validity of drugs whose 100 % trial results are not posted. Research is not a monopoly of academia and industrial R and D sector. Citizens can be good researchers too. We suggest that citizens undertake systematic reviews and meta-analyses with slightly more inclusive norms. Systematic reviews need to mention the number of data-unposted trials and their enrolment sizes for a given drug. Assuming that the unpublished trials did not show any beneficial effect of the drug, the net enrollment size including such trials should be used for calculating and reporting indices such as absolute risk reduction (ARR) or numbers needed to treat (NNT). Citizens can also file public interest litigations and demand data transparency; they can legally challenge drug approvals without 100 % data posting.
We don’t believe that all academics and journal editors are “sold out” as sometimes politically accused. They can certainly take up the issues in an appropriate way. The rate at which clinical trial results remain hidden is certainly alarming and brings the entire science behind medicine under question. Supporters of sound science need to ensure transparency in science at any cost to avoid losing public trust.
Table 1. The proportion of interventional and completed (before 31st March 2025) clinical trials for 6 diseases with their results posted on clinicaltrials.gov. Note the overall poor rate of data posting, even for drugs approved and commonly used in clinical practice.
Table 1. The proportion of interventional and completed (before 31st March 2025) clinical trials for 6 diseases with their results posted on clinicaltrials.gov. Note the overall poor rate of data posting, even for drugs approved and commonly used in clinical practice.
Disease Percentage of results posted A drug in the market % posted Enrollment size Predictive power (McFadden's R² expressed as percentage)
Type 2 diabetes mellitus 28.0% semaglutide 42.75 37.8%
Alzheimer’s 35.03% donepezil 34.55 6.3%
Parkinson’s 28.02% levodopa 35.47 4.9%
COPD 29.15% tiotropium 60.62 11.0
Arthritis 30.99% adalimumab 61.15 12.4%
Depression 27.17% escitalopram 46.89 1.0%

References

  1. Munshi, R.; Pilliwar, C.; Maurya, M.R. Public disclosure of clinical trial results at Clinical Trial Registry of India- Need for transparency in research! Perspect. Clin. Res. 2023, 14, 81–85. [Google Scholar] [CrossRef] [PubMed]
  2. Bajaj, H.S.; Welch, M.; Shah, P.; et al. Efficacy and safety of retatrutide, a GIP, GLP-1, and glucagon receptor agonist, in people with type 2 diabetes and inadequate glycaemic control with diet and exercise (TRANSCEND-T2D-1): a double-blind, randomised, phase 3 trial. Lancet 2026, 407, 2402–2413. [Google Scholar] [CrossRef] [PubMed]
  3. Mandrola, J.M. Rethinking inflammation and ASCVD: Four lessens from the ZEUS trial. Medscape News and Perspectives Aug 12, 2026. Available online: https://www.medscape.com/viewarticle/rethinking-inflammation-and-ascvd-four-lessons-zeus-trial-2026a1000rlm.
  4. Ross, J.S.; Mulvey, G.K.; Hines, E.M.; Nissen, S.E.; Krumholz, H.M. Trial Publication after Registration in ClinicalTrials.Gov: A Cross-Sectional Analysis. PLoS Med. 2009, 6, e1000144. [Google Scholar] [CrossRef] [PubMed]
  5. Tse, T.; Fain, K.M.; Zarin, D.A. How to avoid common problems when using ClinicalTrials.gov in research: 10 issues to consider. BMJ 2018, 361, k1452. [Google Scholar] [CrossRef] [PubMed]
  6. Mitra-Majumdar, M.; Kesselheim, A.S. Reporting bias in clinical trials: Progress toward transparency and next steps. PLoS Med. 2022, 19, e1003894. [Google Scholar] [CrossRef] [PubMed]
  7. Turner, E. H.; Matthews, A. M.; Linardatos, E.; Tell, R. A.; Rosenthal, R. Selective publication of antidepressant trials and its influence on apparent efficacy. New Eng. J. Med. 2008, 358, 252–260. [Google Scholar] [CrossRef] [PubMed]
  8. Angell Marcia, Drug companies and doctors: a story of corruption. The New York Review. 15 Jan 2009. Available online: https://www.nybooks.com/articles/2009/01/15/drug-companies-doctorsa-story-of-corruption/?lp_txn_id=1625238.
  9. Cohen, J., & Kupferschmidt, K, 'A very, very bad look' for remdesivir. Science 2020, 370, 642–643. [CrossRef] [PubMed]
Disclaimer/Publisher’s Note: The statements, opinions and data contained in all publications are solely those of the individual author(s) and contributor(s) and not of MDPI and/or the editor(s). MDPI and/or the editor(s) disclaim responsibility for any injury to people or property resulting from any ideas, methods, instructions or products referred to in the content.
Copyright: This open access article is published under a Creative Commons CC BY 4.0 license, which permit the free download, distribution, and reuse, provided that the author and preprint are cited in any reuse.