Are most shelter offers from Seattle's outreach team turned down?

1. We defined the claim as most of the city outreach team's shelter offers being turned down, and measured it with the share of offers not accepted. The claim usually heard is about people; the city counts offers, so we rate offers and report what is unknown about people.
2. We read every city report we could find with offer counts: quarterly reports to the City Council for January to September 2023 and October 2025 to June 2026, and the 2023 and 2024 annual reports. We checked each number and quote against the PDF text. Every report behind the quarterly figures is listed in the linked dataset.
3. We computed the share of offers turned down, the share leading to a confirmed stay, and the share of acceptances confirmed, for each year and quarter; the tables and calculations are in the linked dataset.
4. We checked the main caveats against the city's own statements on bed supply in 2025 and 2026, its tables of reasons for declining, and the regional homelessness authority's bed count.
5. We reviewed treatment evidence separately: the county's 2025 syringe-program survey and the City Attorney's 2023 report on repeat offenders.
6. All refusal, reason and stay counts come from one department's reports on its own team. We looked for an independent count, such as provider records or surveys of people living outside, and found none.
7. We recorded each limit beside the evidence it affects.

Editorial note: This check covers shelter offers made by Seattle's Unified Care Team from January 2023 through June 2026. It rates the share of offers turned down, not the share of people, which the city does not publish. All counts come from the Human Services Department's own reports; we found no independent count. It does not rate any shelter provider or outreach group. Quarterly reports for October 2023 through December 2024 were not found. Treatment offers are discussed as context but not rated. No external specialist review has been recorded.

Sources are labeled by kind, with what this fact check takes from each, so readers can check the evidence independently.

1. FactHub record FactHub dataset. Seattle shelter offers: accepted, declined and confirmed stays by year

   Saved analysis of the city's yearly counts of shelter offers, acceptances and confirmed stays, 2023 to June 2026.

   • Offers turned down: 61% (3,489 of 5,693) in 2023; 75% (5,681 of 7,565) in 2024; 74% (5,244 of 7,090) in 2025; 83% (4,863 of 5,886) January to June 2026. Confirmed stays: 17%, not published, 14% and 11% of offers; 44% of accepted offers in 2023.

2. FactHub record FactHub dataset. Seattle shelter offer outcomes by quarter

   Saved analysis of quarterly counts, January to September 2023 and January 2025 to June 2026.

   • Most offers were turned down in every published quarter, from 54% (January to March 2023) to 84% (April to June 2026); offers rose to 2,788 and 3,098 a quarter in 2026, from about 1,770 a quarter in 2025, while acceptances stayed near 500.

3. FactHub record FactHub dataset. Reasons given for turning down Seattle shelter offers, April to June 2026

   Saved analysis of the city's table of reasons for the 2,612 offers declined from April to June 2026.

   • Qualifies: Of 2,612 refusals from April to June 2026: permanent housing 23%, no shared room 20%, stay with partner, family or friends 15%, unknown 11%, would not engage 6%, already has housing or shelter 6%, does not want the shelter offered 5%, pets 1%.

4. Government City of Seattle Human Services Department. UCT Quarterly Reporting (Response to Statement of Legislative Intent MO-004S-A), 2025 end-of-year report, March 24, 2026

   The city's report to the City Council on its encampment outreach team for 2025, with offers, acceptances, confirmed stays and reasons for declining.

   Reliability: Prepared by the department that runs the outreach team, reporting its own results to the City Council; it says confirmed stays are undercounted.

   • Qualifies: The city counts specific offers of an available shelter bed, not people.

     What the source says

     Specific offers of available shelter extended to unsheltered individuals.

     Figure 11 caption, p. 10

   • Qualifies: In 2025 the outreach team had 100 fewer set-aside shelter beds on average than in 2024, which the city says hurt its ability to make offers.

     What the source says

     A lack of shelter to offer: In 2025, UCT had 100 fewer set-aside shelter beds on average to refer people into than in 2024, significantly impacting the ability of the outreach team to both make meaningful offers and move people inside.

     p. 3

   • Qualifies: With fewer beds, the team made shelter offers at only 41% of hazard and obstruction removals late in 2025.

     What the source says

     the reduction in shelter availability has resulted in offers at obstructions and hazard removals being made 41% of the time.

     p. 7

   • Qualifies: Not wanting a shared room was the most common reason for declining in 2025, and the city says single rooms get much higher acceptance.

     What the source says

     In 2025, the most common reason for declining a shelter offer was, "Does Not Want Communal Space." We continue to have much higher acceptance rates to shelters that offer single occupancy units, rather than congregate spaces.

     p. 10

   • Qualifies: In 2025, 55% of people who accepted shelter moved into shelter.

     What the source says

     In 2025, 55% of people who accepted shelter moved into shelter (Figure 11), 7% higher than in 2024.

     p. 10
5. Government City of Seattle Human Services Department. UCT Quarterly Reporting (Response to Statement of Legislative Intent MO-004S-A), Q2 2026, August 14, 2026

   The city's newest report to the City Council on its encampment outreach team, covering January to June 2026.

   Reliability: Prepared by the department that runs the outreach team; it reports its own results and records a reason for every declined offer.

   • Qualifies: In April to June 2026, about 8 set-aside shelter beds were available on a typical weekday and about 3 on weekends.

     What the source says

     On average, there were 8.2 set-aside shelter beds available on weekdays, and 2.8 on weekends in the second quarter of 2026.

     p. 12

   • Qualifies: The city says the two most common reasons for turning down shelter are wanting permanent housing and wanting a private unit.

     What the source says

     The two most common reasons people turned down shelter are wanting permanent housing and wanting an individual unit (rather than a communal space).

     p. 12

   • Qualifies: From April to June 2026, 61% of referrals to shelter or housing led to a confirmed stay in that shelter or housing.

     What the source says

     Sixty-one percent of referrals to shelter or housing in the second quarter of 2026 resulted in confirmed enrollments in those shelters

     p. 12

   • Qualifies: About a fifth of shelter stays can't be confirmed in the data, so confirmed stays are undercounted.

     What the source says

     Approximately 21% of enrollments cannot be confirmed because entries in HMIS do not include any personally identifiable information

     Figure 13 footnote, p. 14

   • Qualifies: In April to June 2025, about 4 set-aside shelter beds were available on an average day, about half the 2026 weekday figure.

     What the source says

     In Q2 2025 there were only 4.2 shelter beds available on average.

     p. 12
6. Government City of Seattle Human Services Department. Unified Care Team report to City Council, Q3 2023 (Statement of Legislative Intent HSD-301-A-001)

   The city's 2023 quarterly report on its encampment outreach team, with offers and reasons for declining.

   Reliability: Prepared by the department that runs the outreach team, reporting its own results to the City Council.

   • Qualifies: In mid-2023, the most common reason for declining was wanting a tiny house.

     What the source says

     The most common reason provided was "Wants Tiny House" (23%).

     Table 3, p. 6
7. Government City of Seattle Human Services Department. Unified Care Team report to City Council, Q1 2023 (Statement of Legislative Intent HSD-301-A-001)

   The city's first 2023 quarterly report with counts of shelter offers and acceptances.

   Reliability: Prepared by the department that runs the outreach team, reporting its own results to the City Council.

   • In January to March 2023, 616 of 1,352 offers were accepted, the highest acceptance share in the reports found (46%).

     What the source says

     In Q1, UCT extended a total of 1,352 offers of shelter to unsheltered individuals from which 616 offers were accepted and resulted in a referral to shelter.

     Table 2, p. 5
8. Government City of Seattle Human Services Department. 2023 Annual Report

   The department's annual report for 2023, the source found for full-year 2023 offer and acceptance totals.

   Reliability: The department's own annual summary; the counts match its quarterly reports.

   • In 2023 the city made 5,693 shelter offers and 2,204 were accepted.

     What the source says

     5,693 offers of shelter extended; 2,204 were accepted

     p. 7
9. Government City of Seattle Human Services Department. 2024 Annual Report: Addressing Homelessness

   The homelessness section of the department's 2024 annual report, the only source found for 2024 offer and acceptance totals.

   Reliability: The department's own annual summary; it gives no count of confirmed stays for 2024.

   • In 2024 the city made 7,565 shelter offers and 1,884 were accepted.

     What the source says

     In 2024, 7,565 offers of shelter were extended, and 1,884 were accepted.

     Addressing Homelessness section, first page
10. Government King County Regional Homelessness Authority. 2026 Point-in-Time Count FAQ (initial release)

    The regional homelessness authority's questions and answers on its 2026 count of homeless people in King County, cited for its estimate of shelter beds against need.

    Reliability: The agency that runs much of the region's shelter system; it argues the system is overburdened, which supports its funding requests.

    • Qualifies: King County has one shelter bed for every three people who need one.

      What the source says

      There's one shelter bed for every three individuals that need one in King County, meaning the system is overburdened.

      p. 4
11. Government Public Health - Seattle & King County. 2025 King County Syringe Services Program Survey

    A survey of people using King County's syringe service programs about drug use, housing and treatment, cited for what opioid users said they want from treatment.

    Reliability: Run by the county health department, which also runs the syringe programs surveyed; respondents are program clients, not a random sample of homeless people.

    • Qualifies: About half of the survey's participants were homeless.

      What the source says

      49% reporting they were homeless

      p. 3

    • Qualifies: 70% of participants who used opioids wanted to reduce or stop their use.

      What the source says

      Among participants who reported any opioid use in the past three months, 70% reported that they were interested.

      p. 8

    • Qualifies: If treatment were free and easy, 43% of opioid users would choose methadone, 25% detox and 21% inpatient treatment; 11% wanted none.

      What the source says

      methadone (43%) was the most common response, followed by detox (25%), outpatient treatment (21%), injectable buprenorphine (21%), inpatient treatment (21%), 12-step program (14%), and oral buprenorphine (14%). Eleven percent indicated they did not want any of these options

      Figure 16 text, p. 10

    • Qualifies: 29% had tried to get help to cut back in the past year and could not get it.

      What the source says

      29% of participants who used drugs reported that there was a time in the past 12 months when they tried to get help to reduce their drug use but could not get help.

      p. 13
12. Government Seattle City Attorney's Office. High Utilizer Initiative: One Year in Review (2023)

    The City Attorney's review of the first year of its program for people with many misdemeanor referrals, cited for what happened when repeat offenders were sent to inpatient treatment.

    Reliability: Has a stake in this finding. The prosecutor's own review of a program it runs; it argues for treatment while in custody, which this finding supports.

    • Qualifies: Repeat offenders released from jail to inpatient addiction treatment left within 24 hours in all but a few cases.

      What the source says

      In all but a few of those cases, the defendant absconded from the program within the first 24 hours.

      Executive summary, pp. 2-3
13. FactHub dataset Seattle shelter offers to people living outside: accepted, declined and confirmed stays, 2023-2026

    • Tables: shelter-offers

      Saved analysis annual-summary

    • Tables: decline-reasons-q2-2026

      Saved analysis decline-reasons-q2-2026

14. FactHub chart seattle-shelter-offers-turned-down
15. FactHub chart seattle-shelter-offer-decline-reasons
16. FactHub chart seattle-shelter-offer-flow

1. November 8, 2026 Put the outreach team back into the title and answer (First Look 2516080994509723114-6d50130f702013dbc49679d82cec5eea). Added a common question on whether too few beds keep acceptances near 500, with 2025 and 2026 bed counts, and said the city doesn't report whether one bed is offered to several people (Skeptic 2516080989209030786-b07b8d18db045043aa126c8d50953ddb, 2516080989211522166-ffa7fb21e627d9967d7d1816133a3205). Reworded the refusal-reasons question around the top recorded reason so the evidence settles it, added the 6% who said they already had housing or shelter, and put only whether people who say yes show up under the verdict (2516080989206011497-b14ff615582687f37f7c03815d3f74da). Said near the verdict that the claim usually heard is about people, and moved the treatment figures out of the conclusion (2516080989214022106-15ea1fa3f5cd8c5e19b5d71761902508; Standards 2516080988470820440-7893a527dae0511ab12a9e24b5d84492). Split the summary and explained held beds and confirmed stays (2516080988461931752-1f59dfdd53de435c1dac6405fca1e666), used one date style (2516080988464964547-95b43ef0c28cd65de1c53551ddbbf23c), and said why three sources were used (2516080988467892493-68862ec289c1f14071e4cbfb76ee2a30).
2. November 8, 2026 Added a flow chart of where April to June 2026 offers went, shown first; redrew the reasons chart as horizontal bars; cites dataset version 2, whose counts are unchanged.
3. November 8, 2026 The flow chart now colors each flow by where it comes from, without highlighting.
4. November 7, 2026 First version.
5. November 7, 2026 Asked the title about offers, the unit the city counts, and rated that claim True; the share of people who refuse is now stated as unknown (First Look 2516081972678439915-125901fb0b5f3631c64a33a5e702533e; Skeptic 2516081901523498684-3ba426346ecbd7e7f78e116cc95ab242, 2516081901526395574-fb1b4d23b5da9d71af50257a12cdd055, 2516081901528924436-da5684ba3d5e095a6c3222424633abfb; Standards 2516081900076041772-64b47bb8268a4c686142b7e9f258d921). Added methadone (43%) and the 11% who wanted no treatment (Source Checker 2516081900782626871-01a02bedf2bbb5d0c81e34a919491e68). Replaced 'nearly doubled' with about two-thirds, cited the 2025 baseline and said what the city reports cover (2516081900779946762-c5498da1999dad226e52ff4153a1acdc). Fixed periods and citations for the 44%, the 2023 tiny-house reason and the 41% (2516081900785291656-3303df88a7fd113af5f082cee583f675; Standards 2516081900068432006-8e011915596a06029cddefad8134e4c1). Stated that all counts come from one department and that people may reach shelter other ways (Skeptic 2516081901517690747-9e42572c1bdeec18aa5d264bdf97ad77, 2516081901520605216-f5a71975758371b00057a2745b2de8d1). Made the summary one sentence naming the evidence and relabeled the survey source (Standards 2516081900073495777-aac38fefe7bc705888b2c0add395f306, 2516081900070981330-46f1042d1015149a0445be3291ae2f86).

Feedback these versions applied

• Version 5 First Look, 2 findings Skeptic, 9 findings Source Checker, 3 findings Standards, 8 findings