ICE deaths since inception by year
Executive summary
The public record on deaths in U.S. Immigration and Customs Enforcement (ICE) custody is fragmented: authoritative agency reporting covers Fiscal Year (FY) 2018 forward, independent researchers and advocates have compiled earlier totals, and media investigations document recent spikes — notably 2025 as the deadliest year since 2004 (according to ICE-era tallies) [1] [2] [3]. Exact year-by-year counts "since inception" cannot be reconstructed from a single authoritative public dataset in the provided reporting, so this analysis synthesizes available tallies, key studies, and the limits of the sources [1] [4] [5].
1. What “since inception” actually means and why it matters
“As inception” can mean the birth of modern ICE in 2003–2004 or earlier federal immigration custody systems; public reporting most consistently tracks deaths from 2004 onward, and ICE’s own public death-reporting portal provides complete reports only beginning in FY2018 — a statutory requirement enacted in the DHS Appropriations Bill [1] [2]. Researchers and watchdogs therefore typically treat 2004–present as the practical window for compiled tallies, while noting gaps, differing inclusion rules (for example, deaths shortly after release or in Border Patrol custody are often excluded), and uneven transparency across administrations [2] [1] [5].
2. Aggregated counts before the FY2018 reporting era
Independent analyses cited in the sources place hundreds of deaths in ICE custody across the 2004–2019 period: one conservative-leaning policy outlet reported 193 deaths from 2004 through 2019 [4]. Wikipedia’s compendium of reported ICE detention deaths also treats 2004 as its starting point and notes that yearly totals are compiled from agency notices and media reporting, but the entry cautions about variations in how deaths are described in agency communications [2]. These aggregated figures provide a baseline but are compilations of disparate public notices and news reports rather than a single audited dataset [2] [4].
3. FY2018–2023: ICE reporting and independent studies
Since the FY2018 congressional mandate, ICE publishes individual detainee death reports within 90 days, which makes FY2018+ the most reliably documented stretch for researchers [1]. Peer-reviewed and public-interest research has used these disclosures to study causes and patterns — for example, an academic update covering FY2021–2023 followed earlier work on FY2018–2020 and flagged persistent preventable causes, including COVID-19 and suicide, even as vaccine availability and declining daily populations changed the context [5]. Advocacy research (ACLU) examining 2017–2021 concluded that a large share of those deaths could have been preventable with adequate care and reviewed 52 deaths in that period — a finding that contrasts with ICE’s emphasis on adherence to standards while highlighting investigative shortfalls [6] [1].
4. The 2025 spike and what the sources report
Multiple news organizations and ICE notices documented a sharp rise in deaths in 2025: Reuters and other outlets reported at least 30 deaths in ICE custody by mid-December 2025, calling 2025 the highest total since 2004 [3]. The Guardian compiled case-level reporting on the 32 people who died in custody in 2025 and traced circumstances including suicides, medical crises, and violent incidents during a year of expanded detention levels under immigration enforcement policies [7]. Wikipedia’s ongoing compendium also marks 2025 as the deadliest year since 2004 and identifies December 2025 as a particularly deadly month, while emphasizing reporting and labeling choices made by ICE in public notices [2].
5. Why year-by-year totals diverge and the accountability implications
Counts diverge because sources use different inclusion criteria (e.g., excluding deaths after release or in Border Patrol custody), because ICE’s historical public reporting is uneven prior to FY2018, and because advocacy groups, journalists, and academic researchers apply variant methods to verify and classify deaths [1] [5] [2]. That methodological patchwork matters: it shapes perceptions of whether deaths are preventable and whether agency investigations are thorough — conclusions underscored by watchdog reports claiming a high proportion of preventable deaths and by critiques of ICE’s investigative practices [6] [5].
6. Bottom line — the best available year-by-year path forward
A single, authoritative year-by-year table “since inception” is not present in the cited reporting; the most reliable official route for precise annual counts is ICE’s detainee death reporting portal for FY2018 onward, and researchers must triangulate earlier years from compiled datasets like academic reviews, watchdog tallies, and media databases [1] [5] [4]. Policymakers and the public should treat pre-2018 aggregated figures as useful but imperfect baselines, rely on ICE’s FY2018+ disclosures for verified case reports, and weigh independent analyses that document causes and systemic patterns — especially the surge documented in 2025 [1] [5] [3].