Sacramento, California – Sacramento has no shortage of plans, committees, dashboards, agreements or good intentions. In fact, that may be part of what makes the problem so hard to see.
On paper, the region has built a sizable machine to respond to homelessness, mental health crises and emergency calls. Sacramento Steps Forward manages the Homeless Management Information System, or HMIS, for the local Continuum of Care. The Coordinated Access System is supposed to help match people to shelter and housing.
The City of Sacramento has outreach teams, 311 intake and encampment response tools. Sacramento County oversees behavioral health, crisis programs, 5150-related processes, substance use services and teams such as HEART. Hospitals have their own electronic health records. Dispatchers and first responders work through another set of systems.
Each piece has a purpose. Each one sees part of the person.
Too often, no one sees the whole person.
That is the black hole at the center of Sacramento’s response: critical information does not consistently travel with people as they move between the street, a shelter referral, a hospital emergency room, a mental health crisis call, a county service program or a city outreach encounter. Housing status, prior assessments, vulnerability scores, behavioral health history, shelter waitlist position, case management notes and recent contacts may sit in different places, protected by different rules and operated by different agencies.
The public sees the outcome: people cycling through crisis again and again. Frontline workers see the grind: repeated assessments, incomplete histories, missed referrals and systems that still depend too much on phone calls, relationships and manual workarounds. The people living through it feel something more personal. They are asked to retell their story as if nobody was listening the first time.
The latest homelessness numbers show why this matters. Sacramento County’s 2026 Point-in-Time Count estimated 7,458 people experiencing homelessness, up from 6,615 in 2024. That is a 12.7% increase. The sheltered number improved, rising 21.8% to 3,253 people, which suggests that shelter expansion and coordinated access are doing real work. But the unsheltered count also rose, reaching 4,205 people.
The geography changed, too. Unsheltered homelessness declined 19% in the City of Sacramento, while it increased 103% in unincorporated county areas. That shift is more than a map detail. It raises a serious operational question: can Sacramento’s systems keep up when homelessness is moving across jurisdictional lines faster than the data follows?
Sacramento Steps Forward CEO Lisa Bates put a careful frame around the 2026 numbers.
“This year’s data shows that investments in shelter and coordinated access are making a meaningful difference. To build on this progress, our community must continue expanding pathways from homelessness to permanent housing through sustained investments in housing, prevention, and coordinated access.”
That statement is important because both things can be true. Sacramento can be making progress, and the system can still be badly fragmented. More people may be getting into shelter, while too many others remain outside the reach of the region’s real-time information network.
The city and county have already acknowledged the need for better coordination. Their Homeless Services Partnership Agreement, executed in 2022 and 2023, calls for shared work on the Local Homeless Action Plan, integration of emergency shelter beds into coordinated access, data sharing to the extent allowed by law, use of HMIS and CAS, releases of information, joint reporting and coordinated planning.
Those commitments are not small. They show that the region understands the assignment. But signed agreements do not automatically change what happens during a crisis call at midnight, or during a hospital discharge, or when an outreach worker meets someone under a freeway overpass.
That is where the gap lives: between policy language and daily operations.
The clearest example of both promise and limitation is the Healthcare x Homelessness Pilot. Four major Sacramento health systems, CommonSpirit Dignity Health, Kaiser Permanente, UC Davis Health and Sutter Health, entered data-sharing agreements with Sacramento Steps Forward and HMIS. The pilot allows hospital navigators to use HMIS for pre-discharge referrals into coordinated access. Hospitals can also identify patients experiencing homelessness who were not previously known to the homelessness response system.
The goal is simple and humane: stop discharging people back to the street when there may be a better path.
But the pilot also reveals how difficult even limited data sharing can be. Legal and compliance reviews took time. HIPAA concerns had to be addressed. Hospitals had their own risk calculations. Records had to be matched between different health systems and HMIS without a universal client identifier. Consent language, indemnifications, privacy protections and quality-control processes all had to be worked through.
That is not a criticism of the pilot. It is the lesson of the pilot. If connecting hospitals to homelessness data requires this much effort, then connecting behavioral health, dispatch, city outreach, county crisis teams and housing providers into a more seamless network will require real leadership, money, staffing and public accountability.
Right now, the system still leaves too much to chance.
A hospital may know a patient has acute behavioral health needs and nowhere stable to go, but may not have immediate clarity about coordinated access status or an active case manager.
A county crisis team may respond to someone in psychiatric distress but lack real-time shelter availability or recent outreach notes. A city outreach worker may know the person’s location and encampment history but not the clinical context that would help shape the next step.
A dispatcher may know the call type and safety risk but not the housing and service history that might support a diversion response instead of another trip to jail or the emergency department.
None of this means data should be thrown open recklessly. Privacy matters. Consent matters. Medical and behavioral health information must be protected. But privacy cannot become a convenient excuse for a fragmented system that repeatedly loses people in the spaces between agencies.
Good data sharing is not about exposing everything to everyone. It is about giving the right people the right information at the right moment, with strong safeguards, clear rules and real oversight.
The annual Point-in-Time Count itself shows the limits of the current system. In 2026, Sacramento relied on 779 trained volunteers, a 30% increase from 2024, to cover 200 census tracts and conduct about 1,200 surveys across a county of roughly 1,000 square miles. That is a major civic effort. It is also a reminder that Sacramento still does not have complete, daily, by-name visibility into its unsheltered population.
The PIT Count is useful. It gives the region a benchmark. But it is still a snapshot. People move. Crises happen between counts. New people become homeless. Others enter hospitals, jails, shelters, treatment programs or temporary housing. A system that depends heavily on periodic counts will always be looking partly backward.
The consequences are not abstract.
Fragmented data slows housing prioritization. It increases the risk that people repeat assessments. It weakens continuity after shelter or housing placement, especially for people with serious mental illness or substance use challenges. It can cause people to cycle through emergency rooms, crisis teams, law enforcement contacts and temporary beds without a stable plan following them through the system.
The 2026 data showed encouraging shelter gains among high-vulnerability groups, including a 71.4% increase in sheltered adults with serious mental illness. But the broader picture remains troubling. Overall homelessness went up. Unsheltered homelessness went up. People with complex needs still remain outside.
That tells us Sacramento’s system may be getting better at helping some people once they are fully inside the coordinated response. The harder question is whether the region is catching people early enough, linking them quickly enough and keeping them connected after the first intervention.
For emergency and mental health response, that question becomes urgent. Historical regional data cited in the report showed more than 1,600 psychiatric emergency department visits per month in earlier periods, along with prolonged boarding times and related costs for staffing, security, law enforcement holds and testing. When information is split across systems, the default response can become expensive and repetitive: emergency room, street, crisis call, transport, discharge, repeat.
That is bad policy. It is also a bad human experience.
Sacramento does publish important data. Sacramento Steps Forward provides PIT reports, HMIS dashboards, coordinated access metrics and HUD system performance measures. The city-county agreement calls for semiannual reporting to elected officials. These tools help the public see broad movement in the system.
But the public still cannot easily see whether the systems are actually connecting.
How many hospital-identified homeless patients are successfully matched to HMIS and referred into coordinated access? How long does it take from a crisis encounter to a housing or treatment plan? How often are assessments duplicated? How many people touched by county behavioral health are also active in the homelessness system? Which providers are outside full HMIS participation? Where do referrals break down?
These are not side questions. They are the accountability questions.
Sacramento needs a public integration scorecard. Not a dashboard that reveals private client details, but one that tracks whether the city, county, hospitals, behavioral health providers, dispatch systems and homelessness agencies are closing the gaps they have already identified. The scorecard should measure referral completion rates, successful HMIS matching from hospitals, time from crisis contact to stable pathway, duplicated assessment reduction, behavioral health-to-housing coordination, non-HMIS provider participation and outcomes after discharge.
Without those measures, the region is asking the public to trust that coordination is improving. Trust is not enough. Not when the stakes are this high and the costs are this visible.
There are models to build from. The health pilot offers one. Built for Zero communities have shown the value of real-time, by-name data across providers. California’s broader health and whole-person care efforts point toward stronger links between housing, health care and behavioral health. Sacramento does not have to invent the idea from scratch.
But it does have to execute.
That means technical work: better record matching, privacy-preserving data links, shared workflows and perhaps regional middleware that can connect systems without flattening all privacy protections. It means legal work: clear agreements that allow useful information to move while protecting sensitive records. It means staffing: people who can enter, clean, verify and act on the data. It means culture change: agencies agreeing that their own database is not the center of the universe.
Most of all, it means designing the system around the person in crisis, not the agency receiving the call.
Sacramento has pieces of a better system already in place. It has HMIS. It has coordinated access. It has a city-county partnership. It has shelter investments producing measurable gains. It has a hospital pilot that proves cross-sector data sharing is possible, even if difficult.
But the black hole remains.
Until critical information can follow people across homelessness services, behavioral health, hospitals, outreach and emergency response, Sacramento will continue paying for fragmentation. It will pay through repeated crisis calls. It will pay through emergency room use. It will pay through staff time wasted on duplicated work. It will pay through public frustration.
And the people experiencing homelessness will pay the highest price.
The next phase of Sacramento’s homelessness response should not be judged only by how many programs exist or how many meetings are held. It should be judged by whether a person in crisis is recognized, connected and helped without having to start from zero every time.
That is the standard. Sacramento has the tools to reach it. Now the region has to prove it can connect them.