Anchorage HMIS report: point‑in‑time count shows rise in unsheltered people; dashboards explain data differences

Jessica Parks, a data analyst with the community HMIS lead agency, briefed the Assembly Housing and Homeless Committee on July 16 about the Anchorage data snapshot, dashboards and the January 27, 2025 point‑in‑time (PIT) count.

The PIT count is a HUD‑mandated once‑a‑year point‑in‑time estimate of people experiencing homelessness on a single night. Parks said the 2025 PIT occurred Jan. 27 and used a three‑day field window and deduplication to avoid double‑counting. “This year, that point in time count happened on January 27, 2025. It was done over the course of three days,” she said.

On the night of the count the HMIS data showed 402 people were unsheltered; emergency shelter (year‑round) housed 748 people, seasonal/cold‑weather shelters housed 365, and transitional housing programs counted 233, for a total of 2,655 people counted across categories on the PIT night, Parks said. Compared with 2024 numbers, the unsheltered PIT count rose from 346 to 402.

Parks explained why the monthly HMIS dashboard and the PIT snapshot can show different totals. The dashboard’s monthly active‑people number reflects anyone with recorded contact in the HMIS system who is considered literally homeless, including people who have recent program interactions and a 90‑day inactivity policy. “If somebody is in a shelter and they leave that shelter and they don't tell someone at that shelter where they're going … our inactive policy is that we don't drop them from that number until it has been 90 days,” Parks said. That policy can produce higher monthly totals than the one‑night PIT snapshot.

Parks highlighted occupancy and outcomes: permanent supportive housing and other permanent housing programs were largely at capacity; overall utilization was about 92% on the PIT night. She emphasized June results tied to recent federal rental assistance deployment: June outflow (people exiting homelessness to housing) exceeded inflow and the ERA funds helped prevent new homelessness and house people.

The PIT survey added a question in 2025 asking people who were unsheltered why they were not in shelter that night. The most common answers were: lack of available beds or being on a waiting list (27%); safety concerns about shelters (14%); personal preference to stay outdoors (13%); and past negative shelter experiences (12%). Barriers such as transportation, privacy needs, health or disability limitations, and pet policies were cited less frequently but were present in responses.

Committee members asked for clarifications about apparent discrepancies between the PIT totals and the dashboard’s active‑homelessness counts; Parks and other staff reiterated that the two tools measure different things for different purposes and that the statewide dashboards default to the entire state unless users filter to Anchorage.

Parks also reported operational items: a new accessible shower trailer was near completion and expected to arrive in Anchorage within weeks; it will include a roll‑in shower and accessibility features for people with mobility issues.

The briefing underlined continuing system capacity limits—shelters and transitional housing were near capacity on the PIT night—and the importance of ongoing outreach, real‑time bed coordination and funding to move people into stable housing.