988 Crisis Line and Mobile Dispatch Software: How Does a Counselor Send a Team and Find a Bed Without Leaving the Call?

Why the counselor is doing three jobs with two hands

A counselor is nineteen minutes into a call with a woman in her car in a supermarket park. The risk assessment is done. The plan is a mobile team out and, if that does not settle it, a crisis stabilisation chair rather than an emergency department. To make that happen while staying on the line, the counselor has to know which of the three mobile teams is free and where they are, and whether the stabilisation unit across the county has a chair open tonight. The team roster is a shared calendar. The team's location is a text message to a supervisor. Bed availability is a phone call to a unit that may not answer for ten minutes. She puts the caller on a soft hold, or she talks and types and hopes, and if it takes long enough the practical outcome becomes a police welfare check, which is the exact outcome the whole system was built to avoid.

That is the shape of the problem. Since 988 launched in 2022 the front door has worked far better than the two doors behind it. The Crisis Now framing of someone to talk to, someone to respond, somewhere to go describes three services, and most regions bought software for the first one, radios and phones for the second, and nothing at all for the third. The counselor is the integration layer, in real time, while holding a life on the line.

The cost is not abstract. State contracts increasingly pay on answer rate, in-state answer rate and disposition mix, meaning the proportion of contacts resolved without law enforcement or an emergency department. Every minute a counselor spends hunting a bed is a minute she is not answering the next call, so the coordination failure shows up directly in the metric the contract pays on.

Problem one: contact records and dispatch records are different systems and one caller

iCarol is the established platform for crisis lines and it does the contact record properly: intake, risk assessment, disposition, follow up, the reporting a Lifeline centre needs. It was built when a crisis line was a phone room. Bamboo Health OpenBeds solves a different and real problem, which is bed and treatment capacity visibility across a region, and it is the right answer for many states. Julota comes at it from the community and cross-agency direction. Each is competent in its lane. The counselor lives at the intersection of all three lanes, and the intersection is where nothing is built.

What a custom build does: one contact object that carries the person from the first hello through dispatch, arrival, on-scene disposition, bed placement and follow up, with each stage adding to the same record rather than starting a new one in a different tool. That sounds obvious and it is exactly what the current tooling does not give you, because each product owns a stage. The practical consequence of one record is that when the same person calls again in four days, the counselor sees what happened last time, including what the mobile team found in the house, which is often the most useful clinical information anyone has.

Problem two: dispatch is a map problem and it is being solved by memory

Mobile crisis teams are a small fleet with hard constraints. A team is two people, one of whom may need to be a licensed clinician for the assessment to count. Teams have shift boundaries, geographic zones, and a wide variance in how long a call takes because a good on-scene resolution can run two hours. Coverage gaps are the norm, not the exception, and the supervisor holds the picture in their head.

What a custom build does: a dispatch board with live team state, sourced from the teams' own phones rather than from a supervisor's guess. Available, en route, on scene, transporting, in documentation, out of service. Assignment respects licensure requirements and zone, shows estimated arrival honestly, and surfaces the gap before it becomes a problem, for example that the north zone has no clinician-staffed team after 22:00 tonight. Response time from request to on-scene arrival becomes a measured number rather than an estimate the contract manager provides at renewal. This is the single feature most regional systems tell us they lack and it is not technically hard, it is simply nobody's product.

Problem three: bed availability is a phone call, and phone calls decay

Somewhere to go is where the model breaks most often. A crisis stabilisation chair, a detox bed, a youth crisis bed and an inpatient psychiatric bed are four different resources with different admission criteria, and criteria matter enormously: age, acuity, voluntary status, insurance, current intoxication, aggression history and sex-segregated unit rules will all disqualify a bed that appears available. So a counselor calls, asks, and is told no for a reason the board would never have shown.

What a custom build does: model beds as capacity with criteria attached, not as a count. Each facility publishes not just open beds but what they can accept right now, and the request goes to facilities whose criteria the patient actually meets. Give facilities a genuinely fast way to update, meaning one screen on a phone that a charge nurse can do in fifteen seconds between admissions, because any system that asks a busy unit for more than that will go stale within a week and a stale bed board is worse than no bed board. Then measure staleness openly and show it, so a counselor knows whether the green figure is four minutes old or nine hours old. Where OpenBeds already covers your region, integrate with it rather than duplicating it, and spend your build on the referral and acceptance workflow around it.

Problem four: confidentiality boundaries that cannot be handled by good intentions

This system sits between behavioural health, 911, law enforcement, hospitals and community providers, and each boundary has different rules. Substance use treatment records carry stricter federal protection than general health information. What a 911 dispatcher may be told to keep a responder safe is not the same as what a hospital may receive for a warm handoff. A caller may want a mobile team and specifically not want police to know their name, and that request has to be technically enforceable rather than a matter of who is on shift.

What a custom build does: role and purpose based access, with the record segmented so a 911 interface receives only the safety-relevant minimum and never the clinical assessment. Consent captured as a scoped record covering which agencies and which categories, revocable, with every disclosure logged as an event. Anonymous contacts stay anonymous by design rather than by discipline. This matters commercially as well as ethically, because the co-response and 911 diversion arrangements that make your programme fundable depend on both sides trusting that the boundary is real.

What this costs and how long it takes

Across the 2,000-plus projects Digital Heroes has delivered, this category prices as follows. A first release covering the unified contact record with risk assessment and disposition, the mobile team dispatch board with a team mobile app, and the contract metrics with drill-down runs $70,000 to $140,000 in 12 to 18 weeks. A full platform adding bed and capacity coordination with criteria matching, facility referral and acceptance workflow, 911 and law enforcement interfaces, follow up caseloads, chat and text channels and telephony integration runs $180,000 to $400,000 across 6 to 12 months.

What drives cost up: telephony integration, since screen pop, recording linkage and queue data from your phone platform is real work and the effort depends entirely on which platform you run. Text and chat channels, which are a growing share of 988 volume and are a separate concurrency model from voice because one counselor holds several conversations at once. Number of partner facilities for bed coordination, each of which is an adoption problem more than a technical one. 911 computer aided dispatch interfaces, which are slow because the other side is a public safety system with its own governance. And multi-county or statewide scope, where routing and reporting multiply.

What keeps cost down: dispatch and the unified record first, bed coordination in phase two or via an existing regional platform, and text channels once voice is stable.

Build versus buy, and when buying is the right call

Buy if you are a call centre that answers and refers, with no mobile teams of your own. iCarol will handle your contact record and your Lifeline reporting for a small fraction of a build, and your real constraint is counselors on shift. Buy also if your state has funded a regional platform that your contract expects you to use, because running a parallel system against a state deployment is a losing position regardless of how much better your version is.

Build when two or more of these are true. You dispatch your own mobile teams and the dispatch picture lives in a supervisor's head. You operate across several counties with different partner agencies and different reporting. Your state contract pays against metrics you assemble manually each month. You are trying to prove law enforcement diversion, which requires linking a contact to a disposition to an outcome across agencies, and no single product spans that. Or you are the regional coordinating entity rather than a single centre, in which case the coordination between the three legs of the model is your actual job and buying a product for one leg does not do it.

Our position: the counselor should never be the integration layer. Anything that returns minutes to the person on the line pays for itself in answer rate, and answer rate is what the contract pays on.

How to choose a developer for crisis system software

Ask them to sit with a counselor for a full shift before designing anything. Everything that matters in this build is about seconds and cognitive load during a live call, and that cannot be gathered in an interview. A developer who declines the shift will build you a case management system with a dispatch tab.

Ask how they model consent and disclosure. The right answer is scoped, revocable consent, record segmentation so a 911 interface receives only safety-relevant fields, and disclosure logging as events. Vague reassurance about compliance is not an answer.

Ask what they will do to keep the bed board fresh. If the plan requires facilities to log in and update regularly with no incentive, the board will be stale inside a fortnight. You want to hear a fifteen second update path on a phone, visible staleness indicators, and a plan for integrating with whatever regional platform already exists.

Ask who owns the code and settle it before kickoff. You should own the repository, the cloud accounts and the right to hire anyone else. At Digital Heroes the client owns the code from the first commit. State funded programmes should be especially firm on this, because contracts get recompeted and the system has to be transferable to whoever runs the region next.