Keep Factually independent
Whether you agree or disagree with our analysis, these conversations matter for democracy. We don't take money from political groups - even a $5 donation helps us keep it that way.
Fact check: How does the 2025 migrant death toll in ICE detention compare to the average annual toll from 2015 to 2020?
Executive Summary
The available reporting shows the 2025 death toll in ICE custody is substantially higher than the average annual toll for 2015–2020, with contemporaneous counts ranging from 12 to 16 deaths reported at different points in 2025 and repeated statements that this exceeds the earlier multi-year average [1] [2] [3]. Differences between outlets reflect timing and scope — early summer tallies (12–13) were later updated to at least 16 by late September 2025 — but every source cited frames 2025 as an unusually deadly year for immigration detention [4] [3].
1. How advocates and reporters frame 2025 as a sharp departure from previous years
Reporting in mid-2025 and updated into September describes 2025 as the deadliest year in immigration detention in recent memory, with multiple outlets asserting that the year’s toll already exceeds the average from 2015–2020. The Marshall Project and related coverage emphasize that by July the count had reached 12 and matched 2024’s toll, and later coverage lists 13 and then 16 confirmed deaths as authorities and advocates continued to document cases [1] [2] [4]. This framing is consistent across articles published June through late September 2025, indicating a steady upward revision of the fatality count as new cases were confirmed [3] [2].
2. What the raw counts reported actually are and how they changed over 2025
Sources show a clear progression: a July 2025 report cites 12 deaths since October 2024, noting the total had matched the previous year and was likely to rise [1]. Subsequent reporting in late September 2025 lists 13 deaths in one account and at least 16 deaths in others, with those higher totals identified as covering deaths “since January” or “this year” [2] [3]. These increases reflect ongoing confirmations and retroactive classification of detainee deaths, underscoring that early tallies were provisional and that independent reporting continued to add cases to the 2025 total through September [4] [3].
3. What “average annual toll from 2015–2020” means in these accounts
The pieces repeatedly compare 2025’s toll to an unstated but lower multi-year average for 2015–2020, framed as the baseline that 2025 surpasses. Reporters do not uniformly publish the computed numeric average in the excerpts provided; instead, they rely on contextual claims that the 2025 totals exceed the earlier period’s normal annual deaths [1] [4]. This means the conclusion — 2025 is higher than the 2015–2020 average — rests on two factual pillars in the coverage: rising confirmed 2025 deaths (12–16) and historical totals for 2015–2020 that were lower, though the exact average number is not given in these summaries [3].
4. Where sources agree and where they diverge — timing, classification, and scope
All sources agree that 2025 is unusually deadly in ICE custody and that conditions and treatment are contributors cited by advocates and some journalists [4]. They diverge on the precise count at any given date: mid-summer accounts show 12–13, later September dispatches report 16, and the phrasing sometimes differs between “since October 2024,” “since January,” or “this year,” affecting comparability [1] [2] [3]. These discrepancies arise from reporting windows, retrospective case confirmations, and differing inclusion criteria (for example, deaths in ICE custody vs broader immigration detention systems) noted across the pieces [3].
5. Context offered by reporters linking higher deaths to detention practices and population shifts
Multiple articles connect the rising 2025 death toll to larger detention populations and worsened conditions, pointing to overcrowding, medical neglect, and recent increases in detentions that exceed 60,000 in some accounts [3]. The reporting cites individual cases — such as alleged untreated illnesses and suicides — to illustrate mechanisms behind fatalities, and frames the trend as part of systemic problems rather than isolated incidents [3] [4]. Those contextual claims are presented alongside official counts and activist statements, providing explanatory context even where causation is complex and under investigation [1].
6. Potential agendas, reporting limits, and what is not yet resolved
Coverage consistently amplifies concerns from advocates and lawmakers about conditions; this amplifying role should be recognized as an advocacy-laden angle even as journalists document deaths and official responses [4] [3]. Conversely, the summaries do not include ICE’s internal data or full medical-review outcomes, leaving open questions about cause-of-death adjudications, classification rules, and whether all deaths reported by journalists will be reflected in agency tallies. The evolving counts and different time frames mean precise comparisons to the 2015–2020 average are subject to revision as investigations and official statistics are finalized [2].
7. Bottom line: comparison and remaining uncertainty for readers
Based on the contemporaneous reporting provided, 2025’s death toll in ICE custody — reported as 12, 13, and later at least 16 across June–September 2025 pieces — is described consistently as higher than the 2015–2020 annual average, though the exact numeric baseline from 2015–2020 is not specified in these excerpts [1] [3]. The main uncertainties are the evolving case count, differing reporting windows, and formal cause-of-death adjudications; readers should treat the 2025 totals as current best counts that surpass the earlier multi-year baseline but remain subject to official confirmation and dataset harmonization [3] [4].