Realigning a fragmented behavioral health system
The ask
A 20+ hospital regional health system needed a systemwide behavioral health strategy to improve care coordination, reduce avoidable care escalations, better optimize resource utilization across sites, and address an addiction recovery center that was posting seven-figure monthly losses.
This was not just a struggling behavioral health service line. It was a system, like others, designed to generate the results it was currently experiencing. Behavioral health was split across dozens of underutilized ambulatory sites and four inpatient campuses. Hospital CEOs were incentivized on campus-level performance, and service-line leaders were measured on their own parts of the workflow. Behavioral health was being treated as a medical service line inside an incentive structure that made system-level thinking difficult and, at times, in direct conflict with what the service line actually needed. All of this was exacerbated by a culture where a single dissenting opinion from compliance, legal, finance, or a similar function could grind progress to an immediate halt.
What was really going on
Where the tension was
Leaders knew difficult decisions were needed, but many of those decisions ran directly against how individuals were measured, rewarded, and protected. The organization wanted behavioral health to prove itself before investing further, yet the very structure of the system made that proof hard to generate. The addiction recovery center became the focal point of that tension: should the system keep filling beds with the wrong patients to reduce visible losses, or recommit to the original model and absorb the discomfort of short-term tradeoffs?
What changed the trajectory
I reframed behavioral health as a system problem, not a campus-level consideration. That shifted the conversation from “how do we deal with this individual part of the larger whole?” to “what does the system need behavioral health to do, and what decisions are required to make that possible?” From there, the work was not about offering generic recommendations. It was about forcing explicit alignment around what success meant, what tradeoffs leadership was actually willing to make, and which decisions belonged to advisors versus accountable executives. That distinction mattered. In a culture where any one function could stall action, part of the work was helping leadership stop outsourcing decisions to advisory groups and start using those groups properly: to inform decisions, not make them.
What changed in the end
The work resulted in a system-level behavioral health strategy that redefined how services were organized, delivered, and governed across the enterprise.
Inpatient and ambulatory assets were repositioned to function as a coordinated system rather than a collection of sites. Care pathways were clarified to reduce unnecessary escalations and improve patient flow across settings. The role of each asset: crisis, stabilization, and longitudinal management was explicitly defined, allowing resources to be aligned with actual patient need rather than historical footprint.
At the same time, leadership aligned on a clear set of strategic priorities, decision rights, and performance expectations for behavioral health as a system service, not a collection of local programs. That alignment enabled decisions that had previously stalled to move forward with clarity and accountability.
The addiction recovery center was stabilized as part of this broader realignment, but more importantly, the system regained the ability to manage behavioral health intentionally, balancing access, experience, workforce sustainability, and financial performance as part of a coherent strategy rather than a series of reactive adjustments.
What this case reveals
Behavioral health does not fail for lack of ideas. It fails when it is forced into operating and incentive models it was never designed to fit.
Applying traditional medical-surgical service line logic to behavioral health introduces structural friction, fragmenting care, distorting incentives, and limiting performance.
Alignment is not enough. Progress requires clarity about who advises, who is consulted, and who decides. Without that, organizations default to consensus, and stall.
Care models must be grounded in operational reality. You cannot promise convenience the workforce cannot sustain. Every access strategy is a workforce strategy, whether leaders acknowledge it or not.
Incentives are not neutral. Compensation models produce exactly what they are designed to produce, even when those results run counter to the system’s stated goals.