community case studies

Johnson County (Iowa City), Iowa

Building a crisis center from the ground up. Johnson County (Iowa City), Iowa Michael Flaum, MD

Iowa City is a university town in eastern Iowa. Home to the largest of Iowa’s three state universities (the University of Iowa), it has a population of roughly 100-175K depending upon how you count (i.e., students account for 30K, surrounding towns another 30-40K). The University is the major employer, including one of largest university hospitals in the country – the University of Iowa Hospitals and Clinics (UIHC).

It is a generally a service-rich area relative to most of the rest of the state, and that is true of behavioral health services. UIHC has over 75 psychiatric inpatient beds across five units (all of which are locked). It also has a relatively large department of psychiatry, accounting for a sizable proportion of the overall number of psychiatrists in the state (Iowa ranks in the bottom five states nationally in terms of psychiatrists per capita).

The most common “front door” for accessing anything other than routine psychiatric services has been the UIHC emergency room. It is a top-tier ER that we are lucky to have in our community. But for behavioral health resources, it leaves much to be desired. Up until recently, it had essentially no dedicated behavioral health staff or resources, other than two “psych rooms” where patients presenting with behavioral health complaints would wait to see the “psych on-call.” That person, as is the case in many academic medical centers, is usually a junior psychiatry resident. And that person’s job has traditionally been one of triage – i.e., to make the following a dichotomous decision: Does this person need admission? If yes, (and if there is a bed available) they are admitted, and then it is up to the inpatient team to determine what is going on and what to do about it. If yes, and there is no bed available, then a system is in place to find a bed somewhere else in the state and transfer them there (with transport provided by law enforcement). If the client does not want to be admitted, there is a very low bar for a 48-hour emergency hold – the resident simply calls the magistrate on call, indicates why they feel the person may be a danger, and that is usually all it takes. Not surprisingly, given the resident’s junior status and relative lack of experience, they tend to err on the side of caution, and so tend to admit if there is any real question. And typically, if the person could go back to where they came from without the situation that brought them in likely to escalate, then they probably wouldn’t have come to the ER in the first place. So, it is not surprising that our rate of admission had been on the order of 60% or more for all those who presented to the ER with a behavioral health chief complaint.

What happens when two out of every three people who present to a busy emergency room with a behavioral complaint get admitted? You run out of inpatient beds pretty quickly, so patients tend to sit waiting for beds in the emergency room, untreated, and mostly unattended to other than by a police or security officer sitting outside the room. The two “psych rooms” are always full, and others with similar complaints end up occupying more and more of the other beds in the ED, typically accounting for the longest lengths of stay in the ED, and slowing down ED throughput dramatically. Some 300 general patients per month were discovered leaving the UIHC ED before being seen because of waiting times (i.e., sitting in the waiting area, as all the ED beds were full) – this is not limited to behavioral health patients, but all patients. And of those behavioral health patients who are admitted to inpatient psychiatric units, many are discharged shortly after arriving. An analysis of all psychiatric admissions (~ 2010) found that the modal length of stay across all psychiatric units was two days, followed next by one day (length of stay is often presented as an average, but in this case, we are looking at the most common lengths of stay). Those data strongly suggest that these were avoidable admissions. Indeed, a common scenario is that a person comes in at night, exhausted after a difficult day – often involving alcohol or substance use, and they have “hit the wall.” After getting some sleep, the next morning things seem different and they are no longer anywhere near the level of crisis they had been in just a day before. But in our system, many of these patients had already been admitted, or worse, they are still sitting in the ER, waiting to be transported to an inpatient facility on the other side of the state – on a temporary legal hold, vowing never again to make the mistake of seeking help. Because this scenario occurs all around the state, i.e., an over-reliance on inpatient hospitalization, it is not at all uncommon for there to be no inpatient beds anywhere in the state, necessitating patients spending longs periods of time, up to several days in the ED. We had essentially no treatment services in the ED, so patients would be waiting for their “treatment” to being once they found their way to an inpatient unit.

Iowa has a long history of local control for its mental health services. Indeed, up until a recent legislative change, each of Iowa’s 99 counties had its own funding streams, and many decisions about services were made at the county level. In the county in which Iowa City is located – Johnson County – a “System of Care” (SOC) group had been established in the early 2000s at the suggestion of a consultant who had been brought in to advise about how our community could improve behavioral health services. This proved to be a critical step in much of what would happen since. The group consisted of representatives from a variety of agencies and entities that share a common population – including each of the hospitals (in addition to UICH, there is a VA hospital and one small private hospital), the primary substance use service provider, the local homeless shelter (which provides a variety of housing services), multiple law enforcement agencies (county sheriff and the police of several surrounding municipalities as well as University police), community support service providers, family and consumer advocacy groups and others. The group met monthly. All those who came did so voluntarily on their own time and expense. The group was organized by a staff member from the county (who later went on to become our jail alternatives person).

Initially, the major benefit of the SOC group was to familiarize ourselves with one another, putting names with faces, getting each other’s’ cell phone numbers, etc., i.e., establishing relationships between people who worked at these various agencies. That proved to be tremendously valuable, especially when working with our highest utilizers of services, many of whom were well known to most of these agencies – and would go from one to the next without any coherent plan or communication between them. Early on, many of us in the SOC group found ourselves doing informal “case conferences” before and after the meeting, and because there was initially some understandable discomfort in discussing personal health information, we soon engaged colleagues from the law school to help us come up with processes to share information in a manner that was consistent with all of the various regulations that govern such disclosure. This specific project alone was extremely useful, within and across the various systems, resulting in a manual and curriculum for information sharing, and a “release of information” form that specifically allowed for sharing across various institutions and agencies.

Through the years, the SOC group continued to meet monthly and pursued a variety of projects, most of which had the common denominator of breaking down silos of services and increasing coordination. Examples included establishing a jail diversion program using a boundary spanner model, behavioral health training for law enforcement and other first responders, and enhanced supported housing for those who were chronically homeless in the context of severe mental illness. We also conducted a detailed financial analysis of all costs associated with a small group of extremely high utilizers of services.

Efforts to enhance our crisis system remained a high priority – both at the local and state levels. One part of this initiative was aimed at improving services within the UIHC emergency department. In 2019, after several years of planning, the UIHC department of psychiatry established a psychiatric arm of the emergency room (which they called a “crisis stabilization unit”), and that has substantially improved some of the problems in the ER described above. However, it is still hospital-based and still results in a relatively high rate of inpatient admission and a predominantly “medical model” response to crisis.

The SOC group wanted an alternative to the two choices that existed for people in behavioral health crisis: hospital or jail. We wanted a third choice. We investigated various models around the county, with subgroups of our SOC team visiting many sites including San Antonio, Arizona (Phoenix and Tucson), Kansas, and Miami. We quickly learned that when you see a crisis center, you’ve seen only one; each of those we visited felt very different from one another.

We liked the idea of having an actual physical space – rather than a “virtual” crisis system – envisioning a campus of sorts, where various health and human service providers would come together, so clients could go to one central place. But we also knew that creating such a campus would be an expensive undertaking, with no obvious sources of funding. We recognized that building grassroots support for this among the community was going to be necessary if it was ever going to happen.

Thus began a multi-year series of meetings, discussions, focus groups, etc., with city councils of all the local municipalities, school boards, university officials, advocacy groups, county boards of supervisors, (not only in our county but in 188 ROADMAP TO THE IDEAL CRISIS SYSTEM surrounding counties), law enforcement leadership, legislators and candidates running for any local office – basically anyone whose agenda we could get on – to talk about why this project was important, and why it might matter to them. That process also allowed us not only to educate and build enthusiasm for the project, but to elicit from the community what kinds of services they perceived to be lacking. We found that most people we spoke with had some personal experience, either themselves, with a family member or a friend who has had some sort of behavioral health crisis and struggled to find timely help. Awareness of the problem was there – they just needed to understand some of the reasons behind it and learn that there were other models that could serve the community better.

Ultimately, through this process we recognized and prioritized the need for several different kinds of services: We wanted: 1) a place for people in behavioral health crisis to be quickly and efficiently evaluated, and in which needed treatment could begin immediately; 2) a safe place for people who were intoxicated to “sleep it off” prior to such an evaluation; 3) ready access to full medical detox if needed (our community was under-capacity for detox, resulting in many admissions to medical inpatient hospital beds); and 4) some capacity to provide a safe place for a subgroup of clients who utilized the services mentioned above, but who might need a few extra days to figure out next steps.

Finally, our community was in need of a low barrier shelter, i.e., a place for people homeless people who were still using alcohol or other substances to be able to safely spend the night, especially during the winter months.

We referred to the model as an “access center”, i.e., a place where people could come to access a variety of services in one place – that would be easy to navigate, welcoming and integrated. We underscored the potential for decreased utilization of costly and inefficient services like hospital emergency rooms, inpatient units and jails (our community had repeatedly voted down a series of bonds to expand our county jail, which was chronically over capacity.).

Eventually, the questions evolved from “What is an access center?,” to questions like “How big does it need to be?,” “How much will it cost, and how will it be financed?,” “Where will it be located?,” “Who will it serve, and when are we going to build it?”

And by 2018, we had financial commitments from the county and each of the municipalities within the county, as well as some funding from surrounding counties that together were nearing the expected capital costs. Somehow, we had cobbled together more than 7 million dollars for these capital expenses.

A staffing model and ongoing cost estimates was developed based on expected third-party payer reimbursement. Much time has been spent with the various payors in an effort to gain their support and optimize funding streams for each of the services. That effort was complicated by the fact that Iowa’s Medicaid system for behavioral health is divided among multiple managed care organizations (MCOs), and those entities change over time. Among the current MCOs, all appear to be supportive, and details for reimbursement of each of the services have been worked out, but it will likely be an ongoing challenge to ensure that reimbursement is reasonably commensurate with services provided.

An ideal site was located and purchase, near to the existing homeless shelter and substance use provider agency, all the hurdles around zoning and NIMBY (Not in My Back Yard) issues were cleared, architects, contractors, etc. were engaged, and the building is set to open in January 2021.

About 75% of the building will be used for the access center with the remainder used for the winter shelter. The access center will have capacity for up to 40 clients at a time, to be used flexibly across four types of services:

  1. Evaluation and treatment: A 23-hour crisis observation, in which a full assessment is done and treatment is initiated; expected length of stay (LOS) of approximately one day.
  2. Sobering: A safe place for those who are acutely intoxicated to be cared for, followed by an evaluation when sober as needed; expected average LOS of up to 12 hours.
  3. Medical detox: For those in need of full detoxification services; expected average LOS of up to several days.
  4. Crisis stabilization: For those who need more time before returning to community; expected LOS of up to five days.

The land and building are owned by the county. The county contracted with a local mental managing entity. The managing entity will subcontract the provision of some services with other provider agencies in the community. Specifically, the substance abuse provider agency will provide the sobering and detox services, and the agency that currently provides mobile crisis services will do the initial triage evaluations. A fairly extensive set of outcomes have been developed, with expected reporting requirements and performance targets specified in the contract. These will be reviewed on a regular basis by a community-based oversight board.

Embedded within the oversight and outcome measures is the expectation that all services are performed in a welcoming and engaging, trauma-informed and culturally-sensitive manner.

It has certainly been a journey, spanning more than a decade of effort thus far (at the time of this writing we are still a few months away from opening our doors). Our goals were and remain ambitious, including a change in the expectations of the community that high quality behavioral health crisis services should be available to every member of our community, just like the fire department or ambulance services.

Finally, while the efforts described above were all directed at improving crisis behavioral health services for one community, the impact of this work has already expanded well beyond our local borders. Apparently, word got around the state about our efforts, and people began talking about “access centers” as if it were a known type of entity and a missing part of the service array. Much to our surprise, in a 2018 “State of the State” address, our governor talked about the “need to expand access centers.” This was followed by legislation that year that mandated a minimum of access centers around the state. As is often the case, this was an unfunded mandate, leaving it to the local regions to figure out how to finance them. Had we been consulted, many of us would probably have suggested working things out in one site before replicating it elsewhere, but before we knew it, criteria for the components of an access center were incorporated into state administrative code. The first of these actually opened in the southern part of the state in late 2019, and at least two others besides the one in Johnson County are expected to open within the next year.

We chose to call our access center the “Guidelink Center,” following suggestions from a host of focus groups conducted by a marketing firm we hired. The Guidelink Center is scheduled to open in spring 2021.

© Copyright 2022 - Crisis Roadmap
Support Provided by