Where can I download the ICE “List of Deaths in ICE Custody” FOIA PDF and how do I parse it into a spreadsheet?

1. Where to download the official FOIA PDFs, quickly and reliably

ICE has posted titled PDFs of detainee deaths in its FOIA reading room—examples include “List of Deaths in ICE Custody · October 2003 - December 2, 2013” and an updated file covering 2003–2017, both hosted on ice.gov ^{[1] [2]}. Civil liberties and watchdog groups have also published their FOIA-obtained compilations—for instance, the ACLU/partner report includes a FOIA-derived PDF about deaths in ICE custody and links to the source FOIA material ^[3]. The ICE FOIA Library is the canonical place to search for these documents and related records; ICE publishes FOIA-released materials there and provides guidance about submitting FOIA requests ^{[5] [6]}.

2. What those PDFs usually contain, and why parsing can be tricky

The FOIA PDFs are typically compilations of death reports, administrative reviews, and ODO/medical reviewer materials that vary in structure, redaction and format across time periods, meaning columns and fields are not standardized across files ^{[3] [7]}. ICE’s public-facing detainee death reporting policy and its FOIA outputs reflect statutory and appropriations-mandated disclosure timelines, but individual reports often contain redactions and narrative prose rather than machine-friendly tables ^{[8] [9]}. Watchdog groups and investigative projects have noted that to get full contextual records—medical reviews, transport logs, local records—researchers often must pursue additional FOIA requests or litigation because the PDFs alone can omit detail ^{[7] [10]}.

3. An alternative: obtain the raw data ICE provided to third parties or FOIA requesters

Some researchers and data projects redistribute the raw tabular data ICE produced in response to FOIA requests; the Deportation Data Project and similar archives note that they post raw ICE-provided datasets ^[4]. Where a raw dataset exists, using that avoids the messiness of PDF extraction and preserves ICE’s original fields and metadata; locating such datasets can be faster than manual parsing and is consistent with how several academic and advocacy reports assembled their analyses ^{[4] [10]}. If the desired dataset is not publicly reposted, ICE’s FOIA office and FOIA Library provide the procedural path to request underlying electronic records ^{[6] [5]}.

4. High-level strategy to convert a FOIA PDF into a spreadsheet (limitations noted)

Given the variable formats and redactions, a practical approach is to first check for an available machine-readable dataset ^[4], then download the PDF from ICE or an archival site ^{[1] [2]}, inspect whether the file contains consistent tables or only narrative reports ^{[3] [9]}, and—if no raw table exists—use a combination of automated extraction and careful manual validation to build a clean table; sources provided here do not include step-by-step technical parsing instructions or endorse specific extraction tools, and researchers have historically used FOIA follow-ups to obtain cleaner records when PDFs proved inadequate ^{[5] [7]}. When preparing final spreadsheets for analysis or publication, document provenance (which FOIA PDF or dataset was used and relevant release dates) and note any redactions or inferred fields so downstream users understand limitations ^{[9] [10]}.

5. Closing — the transparency angle and next practical steps

For immediate access, download the PDFs directly from ICE’s FOIA URLs and cross-check with advocacy-group compilations ^{[1] [2] [3]}; for cleaner analytics, search repositories that republish ICE raw data or file a FOIA request via ICE’s FOIA office to ask specifically for electronic tables and supporting records ^{[4] [6]}. The available reporting makes clear that full context often requires follow-up FOIA work because the public PDFs can be fragmented, redacted, and inconsistent in structure ^{[7] [9]}.