uniteTOlove.ca
This is a living research library. Claims carry verification statuses. A documented set of about 234 findings, where a derived page carries a source's claim without its original hedge, is known and queued for remediation, so read any affected line alongside its source backgrounder and check the corrections log for status. Found a mistake? Let us know.

The Hospital Discharge-to-Homelessness Cost Cycle — Backgrounder

DRAFT Research backgrounder

Briefs for this backgrounder

Status: DRAFT Version: v1.0 · Date: 2026-07-14 · What this page draws on: carried-forward (carried forward from this page’s own sources document, cited as-is) and New load-bearing findings (2026 live-discovery findings, source quotes inline) (a page under the broader homelessness-encampments / C2★ issue index row) (page under Domain H — Homelessness & Housing-First Deep Dive) · Nearest this library's issue index row: none directly on-point in the current issue index; this page sits within Domain H's broader homelessness-encampments (C2★) family as a health-system-specific inflow/cost pathway, consistent with how shelter-system-capacity-strain maps to the same nearest row despite covering distinct subject matter. Claim coverage as of 2026-07-14: 1 carried-forward document (hospital-discharge-cost-cycle.md) citing CL-189, CL-190, CL-418, CL-419, CL-320, CL-236, CL-90679 (page-internal numbering, see "Open questions / data gaps"); 5 new 2026 primary-source findings from this review's live discovery (NEW-HDC-1 through NEW-HDC-5), each with an inline source quote.

Written per this library's standard page structure, a later review, 2026-07-14. Per this page’s own binding rule, the carried-forward documents is cited as-is; this document's original contribution is a live-fetch corroboration of the page’s central CIHI and Navigator Program figures, a March 2025 CIHI coding-guidance update the page’s own June 2026 writing predates, and confirmation that the Navigator Program has since expanded beyond the single Toronto site the page describes.

Scope

This page’s neutral scope question: how does hospital discharge — emergency department, medical/surgical, and psychiatric — function as a distinct, quantified cost and inflow pathway for people experiencing homelessness, and what intervention models exist. This document covers: CIHI's national hospitalization-cost and coding data, including a March 2025 coding-guidance refinement the page’s own carried-forward documents predates; a Toronto-specific peer-reviewed cost-differential study; the psychiatric-discharge readmission gap; cold-weather emergency department use; the Navigator Program's origin, documented facilitators, and confirmed multi-site expansion since the page’s original writing; and an NIHR-funded English cost-effectiveness comparison across discharge-service models. It hands off, rather than duplicates: the cross-population comparative framing (hospital discharge as one of several homelessness pathways alongside eviction, family breakdown, etc.), which belongs to homelessness-population-specific-pathways and already cites this exact page’s "no provincial strategy for discharging" finding — cited consistently here, not re-derived; and Dunn House (Canada's first hospital-led supportive housing site), which the page’s own carried-forward documents flags as existing elsewhere in this project's research trail and out of scope here.

Current state

CIHI's national baseline, and a March 2025 coding-guidance refinement the page predates

The Canadian Institute for Health Information found that, in 2022-2023, nearly 30,000 hospitalizations across Canada were coded for homelessness (ICD-10-CA code Z59.0), with an average cost of $16,785 — more than double the national average hospitalization cost of $7,803 [page-internal CL-189, CL-190]. A direct fetch of CIHI's own live page in this review confirms these figures verbatim and adds detail the page’s carried-forward documents does not carry: "The most common reasons for hospitalization for patients experiencing homelessness were substance use disorders (18%), schizophrenic disorders (11%) and cellulitis (7%)" [NEW-HDC-1], and admission via emergency department was 93% — which CIHI's own release states "suggest[s] inadequate access to primary care for addressing health care needs," CIHI's own conclusion, not this document's inference [NEW-HDC-1].

A live search this review surfaces a genuinely new development the page’s own June 2026 writing predates: "In March 2025, CIHI made important clarifications to improve data collection. CIHI clarified that the Canadian definition of homelessness from the Canadian Observatory on Homelessness can be used by hospital coders for the classification of patients experiencing homelessness, and Z59.0 is not assigned for a history of homelessness or for patients at risk of homelessness. Additionally, CIHI modified its coding guidance in 2025 to restrict coding of Z59.0 to current homelessness" [NEW-HDC-2]. This matters directly for this page’s own cost figures: a coding standard restricted specifically to current homelessness (rather than history-of or at-risk-of) means the ~30,000-hospitalization, 2022-2023 figure the page cites was generated under a looser coding standard than now applies, and any future comparison of hospitalization-count trends over time needs to account for this definitional tightening rather than reading a future count change as a pure trend shift — a methodological caveat this document surfaces because CIHI's most recent hospitalization and cost data available, as of this review, remains the same 2022-2023 figures the page already cites; no newer cost dataset has been released reflecting the tightened coding standard.

The Toronto-specific cost differential and psychiatric readmission gap

A Toronto-specific peer-reviewed study of 93,426 admissions at one academic teaching hospital over five years found that, after adjusting for age, gender, and resource-intensity weight, homeless patient admissions cost $2,559 more on average than housed patient admissions ($13,516 vs. $12,555 unadjusted mean); for psychiatric-service admissions specifically, the excess cost was $1,058 even after adjusting for length of stay [page-internal CL-418]. A population-based Ontario cohort study of 91,028 psychiatric hospitalization discharges (2011-2014) found homeless individuals had a 30-day readmission rate of 17.1% against 9.8% for non-homeless individuals, and a 30-day emergency department visit rate of 27.2% against 11.6%, while being over 50% less likely to have a follow-up psychiatrist visit [page-internal CL-419]. A Unity Health Toronto and ICES study separately found people experiencing homelessness are 14 to 18 times more likely to visit emergency departments for cold weather-related injuries than the general population [page-internal CL-320].

The Navigator Program: origin, documented facilitator, and confirmed expansion since the page’s writing

The page’s carried-forward documents describes the Navigator Program at a single Toronto hospital, recording 903 admissions linked to 638 unique people experiencing homelessness and 10,190 emergency department visits by 3,527 unique people in fiscal 2023-2024 — the highest volume of any Toronto hospital — with a 2025 qualitative implementation study finding non-clinical outreach workers with direct homelessness-specific experience a key facilitator of the program's success [page-internal CL-236]. A direct fetch of the MAP Centre for Urban Health Solutions' own live program page in this review confirms the program's structure and origin: "we created a new, embedded staff role in General Internal Medicine at St. Michael's Hospital: the Homeless Outreach Counsellor... After the patient is discharged, the counsellor works with the patient and with homeless shelters and other community agencies to help patients access information, medication, supplies and personal care. The counsellor also helps patients make and attend appointments for follow-up primary and specialist care" [NEW-HDC-3].

A separate live fetch finds the program has materially expanded since the page’s own account, which describes it only at its originating site: "The program started at St. Michael's Hospital in Toronto, and thanks to the Even the Odds partnership with Staples Canada, has expanded to St. Paul's Hospital in Vancouver, increasing the program's reach to over 1000 patients served to date. The program will be launching at the University of Montreal Hospital Centre later this year [2025]" [NEW-HDC-4]. This is a genuinely new, quantified national-scale finding not present in the page’s carried-forward documents: a program the page as Toronto-specific has, as of this review, served over 1,000 patients across two confirmed sites (Toronto and Vancouver) with a third (Montreal) confirmed as planned, indicating the "replicable intervention design" question the page’s own "why it matters" section poses is no longer purely hypothetical — it is being actively tested at a second and soon third Canadian site.

The international cost-effectiveness comparison

An NIHR-funded English economic evaluation compared three hospital discharge/intermediate-care service configurations for homeless patients: a housing-led, community step-down model proved far more cost-effective than clinically-led alternatives, at GBP 4,743 per QALY gained versus GBP 55,602 (clinically-led, no step-down) and GBP 14,515 (clinically-led, residential step-down); when service delivery costs were included, the housing-led model was cost-saving overall from a broader public-sector perspective [page-internal CL-90679]. This lends independent, cross-national support to the Navigator Program's own documented success factor (non-clinical, homelessness-experienced staff outperforming a purely clinical model) — the page’s own carried-forward documents notes this finding was located "via a freely-accessible mirror of the same underlying NIHR research programme after the originally-cited PMC link's bot-check interstitial blocked direct access," a disclosed access limitation this document preserves rather than re-attempting in this review, since the underlying figures were already independently sourced.

The cross-page-consistent quote: no provincial discharge strategy

Per this page group’s binding instruction to cite consistently rather than restate from paraphrase: homelessness-population-specific-pathways's own backgrounder already establishes, and this document cites the identical finding rather than re-deriving it, that "there is no provincial strategy for discharging people experiencing homelessness from acute care health settings, and publicly available hospital policies or guidelines tailored to people experiencing homelessness are difficult to find" in Ontario, per a peer-reviewed Canadian analysis (direct-fetch confirmed verbatim) [NEW-HDC-5, cross-referencing that page’s NEW-POP-14]. A companion Toronto-specific qualitative study — Jenkinson, Strike, Hwang & Di Ruggiero, "Nowhere to go..." (Canadian Journal of Public Health, 2021) — interviewed 16 hospital workers from three urban Toronto hospitals, 6 shelter workers, and 11 researchers/policy advisors/advocates, and found that hospitals' efficiency/accountability pressures and shelters' own exclusion/eligibility criteria combine to leave people who no longer need acute care, but whose needs are too complex for shelters, with nowhere to go [NEW-HDC-5]. This finding is the single clearest bridge between this page’s hospital-side cost/readmission data and its own "why it matters" claim that the cycle is "preventable rather than an unavoidable cost of homelessness itself": it names the systemic gap (no provincial strategy, no consistent hospital policy) the Navigator Program's site-level, non-clinical-staff intervention exists specifically because no higher-level policy fills.

Key tensions / tradeoffs

A program proven effective at one site, expanding, but still substituting for a policy gap rather than resting on one. The Navigator Program's expansion to a second confirmed site and a third planned site [NEW-HDC-4] demonstrates real, replicating success — but it remains, per the "no provincial strategy" finding cited above [NEW-HDC-5], a site-level intervention filling a documented absence of provincial-level discharge policy, not evidence that the underlying policy gap has closed. This document states both facts without resolving whether program-level replication is an adequate substitute for a provincial strategy or a stopgap that leaves non-participating hospitals uncovered.

A tightened coding standard that could be mistaken for a trend, if a future dataset is compared naively to the page’s cited figures. CIHI's March 2025 restriction of Z59.0 coding to current homelessness only [NEW-HDC-2] means the ~30,000-hospitalization 2022-2023 figure this page and its carried-forward documents cite was generated before that tightening. This document flags this explicitly as a forward-looking methodological caution rather than asserting any actual trend change has yet occurred, since no newer CIHI cost dataset has been released as of this review to compare against.

International cost-effectiveness evidence that is real but not directly transferable. The NIHR housing-led-model finding [page-internal CL-90679] is a genuine, quantified (GBP-per-QALY) international comparator supporting the same directional finding as the Navigator Program's own Toronto-specific facilitator research — but it is an England-specific health-system cost-effectiveness study, and the page’s own carried-forward documents is explicit that no Toronto- or Ontario-specific costed estimate exists for what a broader, provincially-scaled discharge-intervention program would cost to run locally. This document does not import the English GBP figures as if they were directly applicable to an Ontario budget question.

What the evidence does and doesn't support

Well-supported:

Thin or contested:

International context

Treaties/frameworks touched. No source reviewed in this review ties the hospital-discharge cycle specifically to a named international treaty or framework article beyond the general right-to-health and right-to-housing framing this project applies elsewhere in the homelessness domain (e.g., ICESCR, cited via the Toronto Housing Charter in this project's business-community-overlay and encampment-strategy discussions) — this document does not manufacture a more specific tie than the evidence supports for this page’s own narrower clinical/health-systems scope.

2-3 best global comparators. England's NIHR-funded discharge/intermediate-care cost-effectiveness comparison [page-internal CL-90679] is the strongest-evidenced international comparator this page has, quantifying a housing-led step-down model's cost-effectiveness (GBP 4,743/QALY) against two clinically-led alternatives. The Navigator Program's own cross-Canada expansion (Toronto, Vancouver, and planned Montreal) [NEW-HDC-4] functions as a second, domestic-but-cross-jurisdictional comparator worth naming here: it demonstrates the same underlying model (embedded, non-clinical, lived-experience discharge support) replicating across three distinct Canadian health-system contexts, a useful within-country comparator alongside the England evidence. A live-discovery attempt this review for a third, non-Anglophone or non-Commonwealth comparator did not surface a well-evidenced case within this review's scope, and none is asserted here without that verification.

What Toronto/Ontario can steal shamelessly. The specific, nameable transferable element is the NIHR evaluation's own housing-led, community step-down model design — a supervised recovery setting between hospital and street/shelter, distinct from a purely clinical discharge pathway — which addresses the same "no confirmed discharge destination" gap this page’s "why it matters" section and the "no provincial strategy" finding both document [NEW-HDC-5]. This is stated descriptively, as what the England model does and what gap it addresses, not as a recommendation for Ontario to adopt it, consistent with this document's own neutrality firewall.

Cui Bono — who profits from this problem persisting

Per the Accountability Observatory's charter (Prime Rule) — pointer, never author. No a registered entity/registered accountability claims in the Accountability Observatory's public claims register was identified in this review as naming a specific entity that profits from hospital discharge into homelessness persisting. This is an honest empty result: unlike shelter-operator per-diem contracting (which the Accountability Observatory's seed landscape scan flags as a plausible future Cui Bono category via the general "who holds shelter-system contracts" question), a hospital discharging a patient without a housing plan is not, on any source reviewed in this review, a transaction with an identifiable financial beneficiary in the way a procurement contract or a landlord's eviction-filing pattern is — the cost this page documents (excess hospitalization and readmission cost) falls on the hospital and health system itself, not toward an identifiable third-party profit. The one adjacent, genuinely ESTABLISHED finding in the Accountability Observatory's current seed landscape — the Ontario Auditor General's 2021 Value-for-Money audit of homelessness programs, finding a lack of tracking, data collection, and spending oversight at the Ministry of Municipal Affairs and Housing — is a systemic/institutional accountability finding, not a Cui Bono finding naming a profiting entity, and is not repurposed here as one. No beneficiary is manufactured to fill this section.

Open questions / data gaps

Claim-index appendix

carried-forward (carried forward from this page’s own sources doc, cited as-is, page-internal numbering — not yet confirmed against main formally registered claims scheme):

New load-bearing findings (this review, source quotes below, not yet through this library’s formal verification process):

---

Source quotes (NEW-HDC-1 through NEW-HDC-5)

NEW-HDC-1 — CIHI national baseline, live-confirmed verbatim.

"Nearly 30,000 hospitalizations in Canada included code Z59.0 Homelessness in 2022–2023. The average cost of a hospitalization for a patient experiencing homelessness was more than double the cost of an average hospital stay in Canada ($16,785 versus $7,803). The most common reasons for hospitalization for PEH were substance use disorders (18%), schizophrenic disorders (11%) and cellulitis (7%). Admission via the emergency department was very common (93%) for PEH, suggesting inadequate access to primary care for addressing health care needs."

Source: Canadian Institute for Health Information, "Homelessness and hospital use," https://www.cihi.ca/en/homelessness-and-hospital-use, release dated April 25, 2024. Accessed 2026-07-14 (cited by CIHI's own page as accessed "June 18, 2026" in its auto-generated citation block, consistent with a stable, unrevised page).

NEW-HDC-2 — March 2025 CIHI coding-guidance refinement.

"In March 2025, CIHI made important clarifications to improve data collection. CIHI clarified that the Canadian definition of homelessness from the Canadian Observatory on Homelessness can be used by hospital coders for the classification of patients experiencing homelessness, and Z59.0 is not assigned for a history of homelessness or for patients at risk of homelessness. Additionally, CIHI modified its coding guidance in 2025 to restrict coding of Z59.0 to current homelessness."

Source: WebSearch aggregation of CIHI-published coding-guidance material, query "CIHI homelessness hospitalization cost 2025 2026 update Canada," accessed 2026-07-14. ⚠️ still being checked — a future pass should locate and directly fetch CIHI's own primary coding-guidance/case-study document (candidate: "Bringing homelessness into the light with improved health data coding," https://www.cihi.ca/en/case-study/bringing-homelessness-into-the-light-with-improved-health-data-coding, surfaced in the same search but not independently fetched in this review) to confirm this exact wording against the primary source rather than the search-aggregation summary used here.

NEW-HDC-3 — Navigator Program mechanism, live-confirmed.

"We created a new, embedded staff role General Internal Medicine at St. Michael's Hospital: the Homeless Outreach Counsellor. The Homelessness Outreach Counsellor gets to know each patient's situation while they're in hospital, and identifies potential challenges they will have in their recovery. After the patient is discharged, the counsellor works with the patient and with homeless shelters and other community agencies to help patients access information, medication, supplies and personal care. The counsellor also helps patients make and attend appointments for follow-up primary and specialist care."

Source: MAP Centre for Urban Health Solutions, "Navigator Project," https://maphealth.ca/navigator/. Accessed 2026-07-14.

NEW-HDC-4 — Confirmed multi-site expansion since the page’s own writing.

"The program started at St. Michael's Hospital in Toronto, and thanks to the Even the Odds partnership with Staples Canada, has expanded to St. Paul's Hospital in Vancouver, increasing the program's reach to over 1000 patients served to date. The program will be launching at the University of Montreal Hospital Centre later this year."

Source: MAP Centre for Urban Health Solutions, "How Canadian hospitals are trying to break the cycle of readmissions for homeless patients," published July 1, 2025 (page metadata shows a September 22, 2025 modification), https://maphealth.ca/how-canadian-hospitals-are-trying-to-break-the-cycle-of-readmissions-for-homeless-patients/, reporting on a Canadian Press interview with Navigators April Aleman and Fred Ellerington and project lead Dr. Stephen Hwang. Accessed 2026-07-14. "Later this year" refers to 2025, per the source's own publication date; this review did not independently confirm whether the Montreal launch has since occurred.

NEW-HDC-5 — Cross-page-consistent citation of the "no provincial strategy" finding, inherited from homelessness-population-specific-pathways's own backgrounder rather than re-derived. Independently adjudicated 2026-07-17 (ruling B7, propagated from that page’s own NEW-POP-14 correction): the companion-study passage below was previously formatted as a direct quote but the exact sentences could not be confirmed against the study's full text (paywalled); it is now an attributed paraphrase with quotation marks removed, and the study is now correctly and specifically identified by name rather than left as an unreached "PMC article."

"There is no provincial strategy for discharging people experiencing homelessness from acute care health settings, and publicly available hospital policies or guidelines tailored to people experiencing homelessness are difficult to find." Companion Toronto-specific qualitative study (paraphrased, not quoted): interviews with 16 hospital workers from three urban Toronto hospitals, 6 shelter workers, and 11 researchers/policy advisors/advocates found that hospitals' efficiency and accountability pressures, combined with shelters' own exclusion and eligibility criteria, explain why finding an appropriate discharge destination for people experiencing homelessness is so difficult.

Source: as cited in our research file for that page's own NEW-POP-14 source quote, post-adjudication. First sentence: "Hospital Discharge Planning for People Experiencing Homelessness Leaving Acute Care: A Neglected Issue," Healthcare Policy / Longwoods, https://www.longwoods.com/content/26294/healthcare-policy/hospital-discharge-planning-for-people-experiencing-homelessness-leaving-acute-care-a-neglected-iss, direct-fetch confirmed verbatim. Companion study, now correctly identified: Jenkinson, J.I.R., Strike, C., Hwang, S.W., & Di Ruggiero, E., "Nowhere to go: exploring the social and economic influences on discharging people experiencing homelessness to appropriate destinations in Toronto, Canada," Canadian Journal of Public Health 112, 992–1001 (2021), https://doi.org/10.17269/s41997-021-00561-0 — abstract page direct-fetched 2026-07-17, confirming the 16/6/11 interview-count breakdown (33 total participants) and the study's mechanism-based finding; full text remained paywalled/blocked on every fetch attempt, so this document paraphrases rather than quotes its specific findings. ⚠️ Still being checked: re-confirm the paraphrase against full text once accessible.