69,377 of 92,016 completed studies
have no summary results posted on ClinicalTrials.gov in the downloaded snapshot (75.4%).
Every clinical trial begins with a question. The registry can tell us when a study was completed; it does not necessarily tell us what the researchers found. That gap is the subject of this investigation.
Using ClinicalTrials.gov records for completed interventional studies with completion dates from 2018 through 2023, this report examines whether completed clinical trials have publicly posted results, how results availability varies across funding categories and trial phases, and how much time passes before publicly posted results appear.
The figures describe results posted in this registry—not all publication channels, and not a finding of regulatory noncompliance.
More than three-quarters of the 92,016 completed trials in this extract have no publicly posted registry summary results.
Is that concentrated in one completion year, or visible throughout the six-year cohort?
Hover over either colour to view counts and percentages. Click the legend to focus on one reporting status.
Across the six completion-year groups, the proportion with posted registry results ranges from 22.5% to 26.4% in the October 2026 export.
This pattern appears in every completion-year group, but it must not be interpreted as a year-on-year trend in reporting behaviour: older studies have had more time to post results. An overall percentage also hides variation between types of studies.
Different funding categories show very different rates of posted results.
The important question is whether that difference persists across completion years, rather than appearing in only one year.
Hover to see the number of studies behind each percentage. Click or double-click legend items to compare specific funder categories.
The contrast is particularly visible in 2023: the NIH category has 88% with posted results, versus 17.4% for Other.
These are descriptive categories, not comparable measures of compliance or causal effects.
Studies can differ in design, eligibility for reporting requirements, and other characteristics; the completion-year groups also have unequal time to report results.
A closer examination of the funding categories is needed
Drug development studies can be classified into phases; device, behavioural and other interventions frequently do not use those phases.
Comparing every study as though it had the same design would be misleading.
What happens when funders are compared within trial phases?
Hover for sample sizes. Only phase–funder groups with at least 100 trials are coloured; blank cells are suppressed comparisons, not zeros.
Even within Phase 3, Industry records 74.3% with results posted, while Other records 25.8%.
The funding gap therefore cannot be described simply as one group running more early-phase studies.
But trial phase is still only one difference: country, intervention, sponsor responsibilities and regulatory coverage may also matter.
Results visibility has two dimensions: whether results appear and when they appear.
A graph showing only eventual reporters misses the trials whose results remain absent.
Here, the denominator includes all eligible completed studies, whether or not their results were posted.
For a four-year comparison, the analysis uses trials completed between 2018 and 2021.
Each has more than 48 months of observation by the 28 October 2026 data snapshot.
Records with negative date intervals are set aside for this timing comparison, rather than being treated as zero-month reporting.
Select a funder with the dropdown. Hover to see the number and proportion of all eligible studies that had posted results by each month. The dashed 12-month line is a reference, not a universal legal deadline.
At 48 months, the proportion with posted registry results ranges from 3.2% to 68.2% across these five groups.
33 records with a first results-posting date preceding the stated primary completion date were excluded from this timing calculation.
Crucially, studies with no posted results remain in the denominator. This is a more informative picture of results availability than comparing waiting times among reporters alone.
Percentages can feel abstract.
Each point below is a registered study with its own title, intervention and participant count.
Explore individual records to see what is—and is not—visible in the registry.
Use the filters to explore individual cases. Hover to see an NCT identifier; look it up at ClinicalTrials.gov to read the full record.
This plot uses a reproducible simple random sample of 2,400 eligible studies, rather than all 92,016 trials, to keep the online page responsive.
Do not use the colour balance in this scatterplot to estimate registry posting rates; the full-cohort comparisons appear in the previous charts.
Vertical jitter is only a display aid. Enrolment values come from the current registry record and may be estimated or actual.
The strongest finding is not simply that a large number of trials have no posted registry results.
It’s that results availability differs markedly by study characteristics, and the differences remain apparent when comparisons are made between more similar groups.
The revised fourth chart also shows how long the results remain unavailable when all eligible studies are counted, instead of looking only at those that eventually posted.
These are substantial gaps in registry visibility—not proof that research findings were suppressed.
To establish compliance, it would be necessary to determine which specific trials were subject to legal or funder requirements, whether extensions were granted, and whether results were submitted but not yet publicly posted.
This dataset does not contain enough information to make those judgements.
The question the data can answer is simpler and still important:
After a clinical trial ends, how readily can someone find its summary results in the public registry?
Dataset: ClinicalTrials.gov CSV export, downloaded 15 October 2026; filters: Study Status = Completed; Study Type = Interventional; Completion Date = 1 January 2018 to 31 December 2023; all conditions, interventions and countries. 92,016 unique NCT IDs were retained. The records reflect the registry at export, not a frozen historical snapshot of each completion year.
Outcome: “Results posted” uses the exported Study Results flag; first posting date uses Results First Posted. These are not the same as results first submitted, and lack of registry results does not prove lack of journal publication (ClinicalTrials.gov, n.d.-a, n.d.-b).
Grouping: Funding categories are the registry’s classifications; they are not audited statements of who paid for each study. Trial phases can be not applicable to certain interventions.
Chart 3: Groups with fewer than 100 records are suppressed to avoid visually emphasizing volatile percentages. That threshold is a display rule, not a statistical significance test.
Chart 4: Uses studies completed in 2018–2021 with exact primary completion dates and at least 48 months of observation as of export. Month-only primary completion dates and negative recorded delays are excluded from timing analysis. All remaining no-results records are included in the denominator. The 12-month marker is contextual and is not used to judge individual legal compliance.
Chart 5: Simple random sample restricted to the five displayed funder groups with reported positive enrolment. It illustrates individual records rather than representing every study; sampling creates uncertainty not captured by the other charts.
Interpretation: Study registration is international and not all registered trials are governed by the same US statutory requirements. The Final Rule applies to certain trials, while NIH also has a broader dissemination policy (National Institutes of Health, n.d.). Because the export does not establish legal coverage, this report does not identify violations or deliberate non-reporting.
ClinicalTrials.gov. (2026, October 28). ClinicalTrials.gov study search: Completed interventional studies with completion dates 2018–2023 [Data set]. National Library of Medicine. https://clinicaltrials.gov/
ClinicalTrials.gov. (n.d.-a). CSV download. https://clinicaltrials.gov/data-about-studies/csv-download
ClinicalTrials.gov. (n.d.-b). Study basics. https://clinicaltrials.gov/study-basics
National Institutes of Health. (n.d.). Policy and regulation on ClinicalTrials.gov registration and reporting. https://grants.nih.gov/policy-and-compliance/policy-topics/clinical-trials/reporting/understanding