Abstract
This cross-sectional study evaluates clinical trials’ rates of compliance with the legal requirements of the US Food and Drug Administration’s Amendments Act of 2007 for timely registration of clinical trials, data verification, delayed reporting, and trial document submission.
The US Food and Drug Administration (FDA) Amendments Act of 2007 (FDAAA 2007), as further clarified by regulations effective in 2017 (the Final Rule), requires sponsors of clinical trials covered by the legal requirements to register and report clinical trials to ClinicalTrials.gov.1 In a previous report,2 we assessed compliance with the legal requirement to report clinical trial results; 1722 of 4209 trials (40.9%) due to report results did so within the 1-year deadline. In this report, we evaluate compliance with additional FDAAA requirements.
Methods
This cross-sectional study followed the Strengthening the Reporting of Observational Studies in Epidemiology (STROBE) reporting guideline and was approved by the University of Oxford, which exempted it from ethics review owing to the use of publicly available data. On January 18, 2021, 4 years after the implementation of the Final Rule, we downloaded the ClinicalTrials.gov database. Using our prior methods2 and public data, we identified all registered trials with data elements consistent with coverage by the Final Rule and a primary completion date after January 17, 2017. ClinicalTrials.gov designates covered trials as “probable applicable clinical trials” or “applicable clinical trials” depending on whether they first enrolled participants before or after implementation of the Final Rule, respectively.
We assessed the proportion of trials in compliance with US regulations for results reporting and in 4 additional areas: trial registration within 21 days of first enrollment of a participant (timely registration), verification of the accuracy of registration data within the last year (annual data verification), requesting delays for results reporting prior to 1 year from the primary completion date (requests for certificates for delayed reporting), and the submission of protocols and statistical analysis plans with results (document submission). Additional methods are in the eTable in the Supplement. For the 4 additional areas, we descriptively assessed factors associated with compliance in each area using univariable and multivariable logistic regression. We set a conservative Bonferroni-corrected significance threshold of P < .001 and report 99.5% CIs. P values were 2-sided. This study used Python 3.8.1 (Python Software Foundation) for all data collection, handling, and analysis.
Results
We identified 27 645 covered trials. Of the 8863 trials due to report results, 3499 (39.5%) did so within the 1-year deadline, and 6099 (68.8%) reported at any time. After excluding the trials not relevant to a given requirement (eFigure in the Supplement), we derived appropriate cohorts for each analysis. The proportion of trials in compliance ranged from 66.0% for on-time requests for delayed reporting to 99.1% for document submission with results (Table 1). Trials registered late were a median of 111 (interquartile range, 28-354) days late, trials with delayed annual data verification were a median of 275 (interquartile range, 122-550) days late, and trial with dilatory requests for delayed reporting were a median of 72 (interquartile range, 22-174) days late. Of the 5449 due trials with full results posted to the registry (ie, completed quality control checks), 5401 (99.1%) accounted for all required documents; only 107 (3.1%) of the 3414 due trials that had not submitted results had any documents available.
Table 1. FDAAA Compliance and Characteristics in Assessment Areas.
| Detailed compliance data | No. (%) | |||||||||
|---|---|---|---|---|---|---|---|---|---|---|
| Compliant results reporting | Timely registrationa | Annual data verification | Certificate of delay requests | Document submissionb | ||||||
| Total (n = 8863) | Compliant (n = 3499; 39.5%) | Total (n = 27 645) | Compliant (n = 24 429; 88.4%) | Total (n = 16 709) | Compliant (n = 12 632; 75.6%) | Total (n = 1354) | Compliant (n = 893; 66.0%) | Total (n = 5449) | Compliant (n = 5401; 99.1%) | |
| Covered status | ||||||||||
| Is a probable applicable clinical trial | 5484 (61.9) | 2249 (41.0) | 9282 (33.6) | 7818 (84.2) | 5180 (31.0) | 3586 (69.2) | 770 (56.9) | 505 (65.6) | 3812 (70.0) | 3764 (98.7) |
| Is an applicable clinical trialc | 3379 (38.1) | 1250 (37.0) | 18 363 (66.4) | 16 611 (90.5) | 11 529 (69.0) | 9046 (78.5) | 584 (43.1) | 388 (66.4) | 1637 (30.0) | 1637 (100) |
| Industry sponsor | 3951 (44.6) | 1784 (45.2) | 11 444 (41.4) | 10 516 (91.9) | 6603 (39.5) | 5001 (75.7) | 1204 (88.9) | 814 (67.6) | 2522 (46.3) | 2509 (99.5) |
| Trial contains a drug | 6096 (68.8) | 2632 (43.2) | 19 074 (69.0) | 17 358 (91.0) | 11 492 (68.8) | 9004 (78.4) | 1061 (78.4) | 713 (67.2) | 4008 (73.6) | 3969 (99.0) |
| Phased | ||||||||||
| Early phase | 3487 (39.3) | 1456 (41.8) | 12 415 (44.9) | 11 533 (92.9) | 7802 (46.7) | 6297 (80.7) | 649 (47.9) | 418 (64.4) | 2276 (41.8) | 2254 (99.0) |
| Late phase | 1770 (20.0) | 869 (49.1) | 5513 (19.9) | 5102 (92.5) | 3247 (19.4) | 2611 (80.4) | 510 (37.7) | 357 (70.0) | 1266 (23.2) | 1257 (99.3) |
| NA | 3606 (40.7) | 1174 (32.6) | 9717 (35.1) | 7794 (80.2) | 5660 (33.9) | 3724 (65.8) | 195 (14.4) | 118 (60.5) | 1907 (35.0) | 1890 (99.1) |
| No. of trials registerede | ||||||||||
| Quartile 1 (1-11) | 2357 (26.6) | 458 (19.4) | 7128 (25.8) | 6135 (86.1) | 4526 (27.1) | 2788 (61.6) | 315 (23.3) | 162 (51.4) | 830 (15.2) | 821 (98.9) |
| Quartile 2 (12-220) | 2390 (27.0) | 709 (29.7) | 6710 (24.3) | 5759 (85.8) | 4095 (24.5) | 2747 (67.1) | 613 (45.3) | 374 (61.0) | 1278 (23.5) | 1268 (99.2) |
| Quartile 3 (221-987) | 2102 (23.7) | 993 (47.2) | 6955 (25.2) | 6153 (88.5) | 4128 (24.7) | 3464 (83.9) | 196 (14.5) | 157 (80.1) | 1577 (28.9) | 1563 (99.1) |
| Quartile 4 (988-3341) | 2014 (22.7) | 1339 (66.5) | 6852 (24.8) | 6382 (93.1) | 3960 (23.7) | 3633 (91.7) | 230 (17.0) | 200 (87.0) | 1764 (32.4) | 1749 (99.2) |
Abbreviation: NA, not applicable.
All trials in the cohort were included in this analysis.
Includes trials that proactively declared having no statistical analysis plan.
Applicable clinical trials are covered trials that began after January 17, 2017; probable applicable clinical trials began before but ended after January 17, 2017.
Early phase = phase 1/2 and 2, late phase = phase 2/3, 3, and 4; NA typically indicates a device trial.
The range of the number of registered trials on ClinicalTrials.gov in each quartile is provided after each quartile name.
Table 2 shows the unadjusted and adjusted odds ratios (ORs) for compliance. In the adjusted models, industry sponsors were more likely to register trials on time (OR, 2.03; 99.5% CI, 1.74-2.37; P < .001), verify data annually (OR, 1.52; 99.5% CI, 1.31-1.76; P < .001), and request reporting delays (OR, 4.71; 99.5% CI, 2.19-10.15; P < .001) within the required time frames, consistent with our prior report.2 Likewise, sponsors with more registered trials were more likely to register on time (quartile 3: OR, 1.76; 99.5% CI, 1.47-2.11; P < .001; quartile 4: OR, 2.74; 99.5% CI, 2.23-3.36; P < .001), verify data annually (quartile 3: OR, 4.35; 99.5% CI, 3.60-5.25; P < .001; quartile 4: OR, 8.70; 99.5% CI, 6.92-10.93; P < .001), and request reporting delays on time (quartile 3: 6.07; 99.5% CI, 2.77-13.32; P < .001; quartile 4: OR, 10.58; 99.5% CI, 4.53-24.7; P < .001). Applicable clinical trials were also more likely to meet registration (OR, 2.14; 99.5% CI, 1.87-2.44; P < .001) and verification (OR, 1.99; 99.5% CI, 1.74-2.28; P < .001) requirements than probable applicable clinical trials.
Table 2. Unadjusted and Adjusted Odds Ratios (ORs) for Factors Associated With Compliancea.
| Factor | Timely registration | Annual data verification | Certificate of delay requests | Document submission | ||||||||||||
|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|
| Crude | Adjusted | Crude | Adjusted | Crude | Adjusted | Crude | Adjusted | |||||||||
| OR (99.5% CI) | P value | OR (99.5% CI) | P value | OR (99.5% CI) | P value | OR (99.5% CI) | P value | OR (99.5% CI) | P value | OR (99.5% CI) | P value | OR (99.5% CI) | P value | OR (99.5% CI) | P value | |
| Is an applicable clinical trialb | 1.78 (1.57-2.01) | <.001 | 2.14 (1.87-2.44) | <.001 | 1.62 (1.43-1.83) | <.001 | 1.99 (1.74-2.28) | <.001 | 1.04 (0.71-1.52) | .74 | 1.17 (0.78-1.76) | .20 | DNCc | NA | NA | NA |
| Industry sponsored | 1.86 (1.63-2.13) | <.001 | 2.03 (1.74-2.37) | <.001 | 1.01 (0.90-1.14) | .74 | 1.52 (1.31-1.76) | <.001 | 1.88 (1.06-3.33) | <.001 | 4.71 (2.19-10.15) | <.001 | 2.34 (0.80-6.83) | .009 | 2.54 (0.76-8.54) | .01 |
| Trial contains a drug | 2.15 (1.89-2.43) | <.001 | 1.35 (1.16-1.57) | <.001 | 1.59 (1.40-1.79) | <.001 | 0.94 (0.80-1.10) | .21 | 1.29 (0.82-2.02) | .07 | 1.08 (0.60-1.94) | .67 | 0.64 (0.19-2.17) | .23 | 0.62 (0.16-2.44) | .25 |
| Phaseb | ||||||||||||||||
| Early phase | 1 [Reference] | 1 [Reference] | 1 [Reference] | 1 [Reference] | 1 [Reference] | 1 [Reference] | 1 [Reference] | 1 [Reference] | ||||||||
| Late phase | 0.95 (0.77-1.16) | .40 | 0.80 (0.65-0.99) | <.001 | 0.98 (0.82-1.17) | .72 | 0.92 (0.77-1.11) | .16 | 1.29 (0.85-1.96) | .05 | 1.14 (0.73-1.79) | .32 | 1.36 (0.37-5.04) | .44 | 0.96 (0.24-3.83) | .92 |
| NA | 0.31 (0.27-0.36) | <.001 | 0.38 (0.32-0.44) | <.001 | 0.46 (0.40-0.52) | <.001 | 0.46 (0.39-0.54) | <.001 | 0.85 (0.49-1.47) | .32 | 0.97 (0.47-2.00) | .90 | 1.09 (0.37-3.16) | .80 | 0.95 (0.29-3.09) | .88 |
| No. of trials registeredd | ||||||||||||||||
| Quartile 1 (1-11) | 1 [Reference] | 1 [Reference] | 1 [Reference] | 1 [Reference] | 1 [Reference] | 1 [Reference] | 1 [Reference] | 1 [Reference] | ||||||||
| Quartile 2 (12-220) | 0.98 (0.83-1.15) | .68 | 1.04 (0.88-1.23) | .48 | 1.27 (1.09-1.47) | <.001 | 1.36 (1.17-1.59) | <.001 | 1.48 (0.93-2.34) | .005 | 1.42 (0.89-2.28) | .01 | 1.39 (0.30-6.35) | .48 | 1.19 (0.26-5.57) | .71 |
| Quartile 3 (221-987) | 1.24 (1.05-1.47) | <.001 | 1.76 (1.47-2.11) | <.001 | 3.25 (2.74-3.86) | <.001 | 4.35 (3.60-5.25) | <.001 | 3.8 (1.90-7.62) | <.001 | 6.07 (2.77-13.32) | <.001 | 1.22 (0.30-5.03) | .64 | 1.57 (0.37-6.58) | .30 |
| Quartile 4 (988-3341) | 2.20 (1.81-2.67) | <.001 | 2.74 (2.23-3.36) | <.001 | 6.93 (5.59-8.59) | <.001 | 8.70 (6.92-10.93) | <.001 | 6.30 (2.99-13.24) | <.001 | 10.58 (4.53-24.7) | <.001 | 1.28 (0.32-5.15) | .56 | 1.58 (0.39-6.44) | .29 |
Abbreviation: NA, not applicable.
Each adjusted model included all listed covariates.
Early phase = phase 1/2 and 2; late phase = phase 2/3, 3, and 4; NA typically indicates a device trial.
Applicable clinical trial status perfectly predicted compliance with the document reporting rule, and therefore the regression did not converge (DNC). This covariate was therefore dropped from the adjusted regression.
The range of the number of registered trials on ClinicalTrials.gov in each quartile is provided below each quartile name.
Discussion
In a cohort of 27 645 covered clinical trials, 4 years after the effective date of the Final Rule, we found that compliance with legal requirements for timely registration of clinical trials, data verification, delayed reporting, and trial document submission ranged from 66.0% to 99.1%. There are some limitations; our study is not comprehensive of all FDAAA requirements, and the publicly available data vary in quality. Some covered trials may not have been identified because of the exclusion of certain fields from the public data, although this effect is likely small and limited to probable applicable clinical trials. Our findings suggest that the FDA should proactively implement additional measures to improve compliance with the FDAAA,3 such as audits that are made public, notices of noncompliance, and the potential imposition of civil monetary penalties. In January 2021, ClinicalTrials.gov reminded users about the deadline for requesting delays to results reporting.4 Open dialogue and sharing of best practices for improvement among sponsors may also aid compliance.5,6
eTable. Supplemental Methods Detail
eFigure. Analysis Population Flowchart
References
- 1.Zarin DA, Tse T, Williams RJ, Carr S. Trial reporting in ClinicalTrials.gov—the final rule. N Engl J Med. 2016;375(20):1998-2004. doi: 10.1056/NEJMsr1611785 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 2.DeVito NJ, Bacon S, Goldacre B. Compliance with legal requirement to report clinical trial results on ClinicalTrials.gov: a cohort study. Lancet. 2020;395(10221):361-369. doi: 10.1016/S0140-6736(19)33220-9 [DOI] [PubMed] [Google Scholar]
- 3.Silverman E. FDA finalizes penalties for trial sponsors, but critics say they fall short. STAT News. Published August 18, 2020. Accessed October 18, 2020. https://www.statnews.com/pharmalot/2020/08/18/fda-clinical-trials-transparency-nih/
- 4.National Library of Medicine . What’s New. ClinicalTrials.gov. Published January 25, 2021. Accessed February 12, 2021. https://clinicaltrials.gov/ct2/about-site/new
- 5.O’Reilly EK, Hassell NJ, Snyder DC, et al. ClinicalTrials.gov reporting: strategies for success at an academic health center. Clin Transl Sci. 2015;8(1):48-51. doi: 10.1111/cts.12235 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 6.Snider SH, Flume PA, Gentilin SL, Lesch WA, Sampson RR, Sonne SC. Overcoming non-compliance with clinical trial registration and results reporting: One Institution’s approach. Contemp Clin Trials Commun. 2020;18:100557. doi: 10.1016/j.conctc.2020.100557 [DOI] [PMC free article] [PubMed] [Google Scholar]
Associated Data
This section collects any data citations, data availability statements, or supplementary materials included in this article.
Supplementary Materials
eTable. Supplemental Methods Detail
eFigure. Analysis Population Flowchart
