Terminated, Withdrawn, or Suspended Suicide Prevention Studies

Abstract

This cross-sectional study assesses the prevalence and associated characteristics of registered studies related to suicidal thoughts and behaviors that have been discontinued.

Introduction

Suicide is a leading cause of death worldwide, and clinical trials are essential for advancing evidence-based prevention and treatment strategies.¹ However, compared with other health domains, few clinical trials examine suicidal thoughts and behaviors (STBs).² This gap reflects multiple factors, including the frequent exclusion of individuals at elevated suicide risk from trials not designed to address STBs and practical and regulatory complexities associated with studying high-risk populations.² These factors may contribute to perceptions about study feasibility and associated risks, leading to study discontinuation.

Prior work has found disease-agnostic study discontinuation rates ranging from 7.4% to 30.0%,^3,4 with a rate of 20.4% for registered clinical trials for the pharmacologic treatment of posttraumatic stress disorder among veterans.⁵ To our knowledge, no prior investigation has examined the scope and associated features of discontinued suicide-related studies, including trials. Systematic examination of study trajectories is therefore needed to address potentially unfounded barriers to study conduct. Using publicly available data from ClinicalTrials.gov, we assessed the prevalence and associated characteristics of registered STB-related studies that have been discontinued (ie, terminated, withdrawn, or suspended [eAppendix in Supplement 1]).

Methods

For this cross-sectional study, we queried ClinicalTrials.gov on April 22, 2026, to identify studies registered between January 1, 1990, and April 22, 2026, using STB-specific keywords (suicidality, suicide, suicide prevention, suicidal ideation, suicide attempt, self-injurious thoughts and behaviors). Entries were screened by study staff to exclude irrelevant studies (eg, “suicide genes” related to programmed cell death). We extracted study identification and/or title, status (eg, terminated, withdrawn, or suspended), reasons for discontinuation, study design, sponsor, study start period, and whether STBs were a primary or secondary outcome or included for other reasons (eg, safety monitoring) from the ClinicalTrials.gov registry. Investigator-reported qualitative reasons for early termination were extracted and independently coded by 2 investigators (H.L. and J.F.) using qualitative thematic analysis, with discrepancies resolved by consensus with a third coder (I.H.S.). Analyses occurred from December 2, 2025, to April 28, 2026. This study used publicly available data and did not meet the definition of human participants research per the Common Rule (45 CFR §46); therefore, institutional review board approval was not required. We followed the STROBE reporting guideline.

Results

Overall, 1205 registered studies included suicide-related keywords, of which 579 (48.0%) were completed, 398 (33.0%) were active, 137 (11.4%) had unknown status, and 91 (7.6%) were discontinued. Among the 91 discontinued studies, 52 (57.1%) were terminated, 35 (38.5%) were withdrawn, and 4 (4.4%) were suspended (Table 1). Seventy-two studies (79.1%) focused on STBs as a primary or secondary outcome, while 19 (20.9%) were STB related based on keyword screening for other reasons (eg, safety monitoring). In addition, 82 studies (90.1%) were interventional and 9 (9.9%) were observational. Primary coded investigator-reported reasons for discontinuation were categorized as problems with recruitment (26 [28.6%]), funding issues (18 [19.8%]), principal investigator–related considerations (8 [8.8%]), and COVID-19 interference (7 [7.7%]) (Table 2). Although 4 studies (4.4%) were suspended due to site-wide Federalwide Assurance restrictions, no identified registered studies explicitly reported discontinuation due to study-specific participant safety concerns within the trial protocol.

Table 1. Characteristics of Discontinued Suicide-Related Studies.

Characteristic	No (%) of studies (N = 91)
Reason for discontinuation
Terminated	52 (57.1)
Withdrawn	35 (38.5)
Suspended	4 (4.4)
Study design
Interventional	82 (90.1)
Observational	9 (9.9)
Funder type
Federal	13 (14.3)
Industry	6 (6.6)
Other	72 (79.1)
Study start perioda
1990-1995	1 (1.1)
1996-2000	0
2001-2005	3 (3.3)
2006-2010	8 (9.1)
2011-2015	12 (13.2)
2016-2020	28 (30.8)
2021-2025	39 (42.9)

^a Based on the date the study started.

Table 2. Qualitatively Coded Themes for Discontinuation Reason.

Discontinuation theme	No. (%) of studies (N = 91)	Examples of listed discontinuation reason
Recruitment	26 (28.5)	“Terminated early due to concerns from the NIH DSMB regarding the rate of recruitment and the resulting inability to accomplish the initially approved study aims.”
		“Lack of eligible participants during study time period.”
		“Study terminated early due to slow accrual or other non-safety related issues.”
Funding	18 (19.8)	“Study was never funded nor started, no participants were enrolled.”
		“No funding.”
		“Trial never received funding.”
COVID-19	7 (7.7)	“Face-to-face appointments, which were required by the IRB for suicide risk evaluation, were suspended by the IRB during the COVID-19 pandemic. During this pause, the PI left the institution. Thus, all study activities were terminated.”
		“Due to COVID-19 pandemic.”
		“This study has [sic] withdrawn due to recruitment issues at our pediatric primary care setting. Due to the COVID-19 pandemic and other unforeseen medical absences, the study was unable to recruit participants.”
PI issue	8 (8.8)	“PI leaving institution.”
		“The study was stopped due to the PI leaving the institution. The grant was ended early.”
		“PI”
Oversight committee stop	3 (3.3)	“On March 21, 2025, we received a notice that this study was terminated as it no longer met the priorities of the administration.”
		“This study was determined by IRB not to be humans subjects research.”
		“Terminated early due to non–safety-related DMC recommendations.”
Did not start	3 (3.3)	“Never initiated, did different study.”
		“Study was not initiated due to lack of resources.”
		“We never started the study.”
Reason not listed or vague	5 (5.5)	“The follow-up of the last patient has been done.”
Safety stop	4 (4.4)	“U.S. Department of Health and Human Services OHRP issued an FWA restriction on NYSPI research that included a pause of human subjects research as of June 23, 2023. This study will not resume recruitment after the resumption of research.”
		“The U.S. Department of Health and Human Services (HHS) Office of Human Research Protections (OHRP) issued an FWA restriction on NYSPI research that also included a pause of human subjects research as of June 23, 2023.”
		“In response to QA/compliance concerns about specific studies, NYSPI is undertaking an Institute-wide safety review of research protocols. Studies will resume recruitment after the review has confirmed they are deploying adequate safety protections.”
Scientific reason	2 (2.2)	“This study protocol was revised, which required closing it and submitting a new application. Since recruitment launched for this protocol, no participants were enrolled and run in this study protocol.”
		“Failure to achieve primary objective.”
		“Interim analysis showed differential treatment effects.”
Study should not have been registered	5 (5.5)	“Study should not have been listed on ClinicalTrials.gov.”
		“Should not have been listed on ClinicalTrials.”
Business reasons	10 (11.0)	“Study stopped due to business reasons.”
		“Staffing and space deficiencies prevented completion of the study.”
		“The study was stopped due to drug unavailability. The drug is unavailable because production was discontinued in the US.”
		“Technical reasons with the online platform for students.”
		“Transferred.”

Abbreviations: DMC, data monitoring committee; DSMB, data safety monitoring board; FWA, Federalwide Assurance; IRB, institutional review board; NIH, National Institutes of Health; NYSPI, New York State Psychiatric Institute; OHRP, Office of Human Research Protections; PI, principal investigator; QA, quality assurance.

Discussion

Approximately 7.6% of suicide-related studies registered on ClinicalTrials.gov have been discontinued, a rate at the lower end of previously reported disease-agnostic discontinuation estimates.^3,4 Discontinuation was primarily attributed to recruitment, funding, and administrative challenges. Despite concerns that suicide-related clinical trials may pose heightened safety risks,² we did not identify any studies in our sample that explicitly reported discontinuation due to study-specific participant safety events.

Four studies in our analysis were suspended due to site-wide Federalwide Assurance restrictions at the New York State Psychiatric Institute (NYSPI) following an individual’s death by suicide while participating in a non–suicide-focused trial and the attendant discovery by regulators of “potential systemic and serious noncompliance with the HHS [Department of Health and Human Services] regulations.”⁶ The NYSPI subsequently paused all human participant research,⁶ which led to trial suspensions, ostensibly including those identified in our ClinicalTrials.gov analysis. This highlights the importance of distinguishing between protocol-related risks and broader oversight or implementation issues.

Limitations of this study include the reliance on registry-reported data, which may be incomplete, inconsistently coded, or lacking detailed context regarding reasons for discontinuation. In addition, studies varied widely in design and measurement approaches. A nontrivial proportion of studies (11.4%) were listed as having an unknown status, potentially underestimating the scope of discontinued studies. Suicide prevention also extends beyond examining suicidal ideation and attempts as outcomes; thus, broader research efforts may not have been captured.

Expanding the suicide prevention evidence base will require strengthening recruitment infrastructure and addressing practical, ethical, and regulatory challenges associated with high-risk study populations. With proper risk mitigation strategies implemented and appropriate oversight,⁷ suicide prevention research is feasible and essential.

Supplement 1.

eAppendix. ClinicalTrials.gov Definitions

Supplement 2.

Data Sharing Statement