Drug Manufacturers’ Delayed Disclosure of Serious and Unexpected Adverse Events to the US Food and Drug Administration

Federal regulations define adverse drug events as those “associated with the use of a drug in humans whether or not considered drug related”.¹ Health care professionals and consumers can voluntarily report adverse drug events directly to the US Food and Drug Administration (FDA) or the drug manufacturer. Serious adverse events (AEs) are defined by the regulation as those involving “death, a life-threatening adverse drug experience, inpatient hospitalization or prolongation of existing hospitalization, a persistent or significant disability/incapacity, or a congenital anomaly/birth defect.”¹ Unexpected AEs are defined as those involving “any adverse drug experience that is not listed in the current labeling for the drug product.”¹ Serious and unexpected events are classified as expedited, and manufacturers receiving any such reports are mandated to forward them to the FDA “as soon as possible but in no case later than 15 calendar days of the initial receipt of the information.”¹ The regulation also requests that the manufacturer conduct an investigation and forward findings as follow-up reports to the FDA. Previous studies highlighted that reports by manufacturers to the FDA of serious adverse drug events have increased steadily during the past decade.^2,3 Manufacturer compliance with the regulation to report serious and unexpected AEs to the FDA within 15 calendar days is unknown, although some media coverage has offered anecdotal examples of delays.^4,5 As the FDA uses this information to update drug warnings, delays in reporting can have important public health consequences, particularly if manufacturers selectively delay reporting based on relevant patient outcomes. We investigated patient and event characteristics associated with manufacturers’ delayed submission of the expedited reports to the FDA.

We extracted the quarterly FDA Adverse Event Reporting System data files of AE reports received between January 1, 2004, and June 30, 2014. We excluded direct reports to the FDA (about 5%) in which drug manufacturers are not involved. Our final sample included only the initial reports (excluding follow-ups) characterized by the FDA as expedited, and therefore subject to the regulation requiring the reports to be submitted within 15 calendar days. Analysis was conducted from May 2014 to May 2015. The University of Minnesota Institutional Review Board determined that this study does not meet the regulatory definition of human subjects research.

Our categorical outcome variable indicated whether the number of calendar days between the date the manufacturer received the report and the date the FDA received the same report from the manufacturer (“days to FDA”) was: 15 days or fewer, 16 to 90 days, 91 to 180 days, or more than 180 days. We estimated a multivariable ordered logit model to examine the association between the categorical outcome variable and whether the AE involved patient death, adjusting for the number of unique drugs the patient was taking, the source of the report to the manufacturer (ie, consumer, physician, pharmacist, lawyer, or other), whether the report was electronically submitted, and patient age, sex, and weight. We included an indicator for the missing values of patient age, sex, and weight. To account for time trends, we included quarter and year indicators, and to account for systematic differences across manufacturers, we included indicators for manufacturers. We clustered standard errors at the drug level to account for correlation within drugs.

The study included 1 613 079 AE reports. Kaplan-Meier estimates show that 9.94% of reports (N = 160 383; 40 464 with patient death and 119 919 without patient death) were not received by the FDA by the 15-day threshold (Figure). Results of the log-rank test rejected the equality of the survivor functions by patient death (P < .001). In multivariable analyses, patient death was associated with delayed reporting (Table). A larger adjusted rate of events without patient death was reported to the FDA within the 15-day threshold relative to those without patient death: 88.25% (95% CI, 86.49% to 90.02%) for events involving patient death vs 90.71% (95% CI, 89.48% to 91.94%) for events without patient death, representing a difference of –2.46% (95% CI, –4.46% to –0.46%). Adjusted rates of reports taking between 16 and 90 days to reach the FDA were 6.42% (95% CI, 5.38% to 7.46%) for events with patient death and 5.19% (95% CI, 4.72% to 5.65%) for events without patient death, representing a difference of 1.23% (95% CI, 0.18% to 2.27%). Similarly, adjusted rates for 91 to 180 days were higher for reports with patient death (2.53% [95% CI, 0.04% to 5.01%] for events with patient death and 1.98% [95% CI, –0.14% to 4.11%] for events without patient death, representing a difference of 0.55% [95% CI, 0.02% to 1.08%]).

Our analysis provided evidence that drug manufacturers delay reporting of serious AEs to the FDA. Strikingly, AEs with patient death were more likely to be delayed. It is possible that manufacturers spend additional time in verifying reports concerning deaths, but this discretion is outside the scope of the current regulatory regime.

Our findings are likely an underestimate of overall underreporting or misreporting, given the anecdotal evidence of FDA warning letters to manufacturers alleging downward misclassification of serious AEs. While increased enforcement may decrease violations, a simple alternative would be to recommend direct submission of reports to the FDA rather than via the manufacturer. Further research is needed to better understand the mechanisms behind the manufacturers’ delayed reporting and the optimal regulatory policy toward mandatory disclosures of AEs.

Corresponding Author: Pinar Karaca-Mandic, PhD, Division of Health Policy and Management, University of Minnesota School of Public Health, 420 Delaware St SE, Minneapolis, MN 55455 (pkmandic@umn.edu).

Published Online: July 27, 2015. doi:10.1001/jamainternmed.2015.3565.

Author Contributions: Dr Ma had full access to all the data in the study and takes responsibility for the integrity of the data and the accuracy of the data analysis.

Study concept and design: All authors.

Acquisition, analysis, or interpretation of data: All authors.

Drafting of the manuscript: All authors.

Critical revision of the manuscript for important intellectual content: All authors.

Statistical analysis: All authors.

Obtained funding: Karaca-Mandic.

Administrative, technical, or material support: Ma, Karaca-Mandic.

Study supervision: Ma, Karaca-Mandic.

Conflict of Interest Disclosures: None reported.

Funding/Support: This study was supported by research funding from the University of Minnesota Accounting Research Center (Dr. Ma) and grant K01AG036740 from the National Institute of Aging (Dr Karaca-Mandic).

Role of the Funder/Sponsor: The funding sources had no role in the design and conduct of the study; collection, management, analysis, and interpretation of the data; preparation, review, or approval of the manuscript; and decision to submit the manuscript for publication.