Like many hospitals, Springhill Medical Center faced mounting challenges in managing patient flow—from ED backups to prolonged discharge times. But with a bold shift toward predictive analytics and AI-driven planning, we turned those roadblocks into rapid wins. As a 270-bed acute-care hospital in Mobile, Ala., our leadership team saw an opportunity to improve operations by reducing our average length of stay, cutting discharge processing times, increasing compliance in entering estimated dates of discharge (EDD), and freeing up inpatient beds occupied by outpatients. These goals led us to embrace a modern, tech-enabled approach to patient flow.
Setting the stage for patient flow improvement
Despite our best efforts—daily huddles, physician-case management meetings, and emailed metrics—we struggled to coordinate care and discharge planning across teams. Communication remained siloed, and teams operated reactively instead of proactively. Without real-time visibility into each patient’s care progression and length of stay, it was difficult to proactively manage discharge planning and patient throughput. The resulting confusion led to backups in the emergency department (ED), long delays in discharging patients, and a generalized capacity crunch.
These well-intentioned efforts—while designed to improve flow—often added to the noise. Separate meetings fragmented communication, and emailed reports rarely translated into timely interventions. Compounding the issue, inaccurate predictions of expected date of discharge (EDD) led to even more reactive planning.
The consequence of inefficient patient flow was felt across the organization. It affected both our financial performance and our ability to deliver timely, effective care. Recognizing the need for a more streamlined and proactive approach, our team set out to find a solution that could automate and standardize discharge planning across the hospital. That search led us to implement LeanTaaS’ iQueue for Inpatient Flow. In a short amount of time, we saw striking results that stemmed from a few key changes:
Reliable discharge predictions at the patient level
With an AI-powered operations platform in our hands, we gained access to accurate, continuously updated estimated dates of discharge (EDDs), helping clinical teams plan each patient’s discharge proactively—not reactively. For the first time, ancillary departments were closely involved and able to prioritize their work based on when each patient was likely to leave. That alignment helped overall length of stay.
Real-time alerts to reduce avoidable delays
We focused heavily on streamlining home discharges—patients who needed little beyond paperwork and coordination. With real-time visibility into these cases, the system flagged any remaining barriers and triggered automated alerts when patients crossed key wait-time thresholds. This enabled expediters to act quickly and shave hours off of each discharge.
Visibility into outpatients and observation patients
Prior to adopting the new system, we lacked a clear view into our bedded outpatient and observation populations—a major blind spot. These patients often represented a missed opportunity to free capacity. Now, we can identify these cases in real time, escalate them for review, and safely move them out of inpatient beds, easing our overall capacity crunch.
Early identification of likely ED readmissions
We also use the system to flag patients who may soon return via the ED — allowing our case management team to intervene early, support discharge stability, and prevent unnecessary readmissions. That upstream focus has helped reduce bottlenecks at the ED entrance and supported smoother throughput overall.
Remarkable results in 60 days
Our transformation wasn’t just fast — it was measurable. Within two months of go-live, we saw significant improvements in discharge efficiency, patient throughput, and team coordination across the hospital. Here’s what we achieved in the first 60 days:
- 50% reduction in discharge processing time (more than two hours saved per patient discharge)
- Half-day reduction in average length of stay
- 90% compliance in entering estimated discharge dates (EDD)
- 50% drop in outpatients occupying inpatient beds
These improvements reflect more than just operational gains. They represent a shift in how our teams collaborate, prioritize, and plan. By introducing real-time insights and more reliable discharge forecasting, we empowered our frontline staff to take proactive steps throughout the day—not after the fact.
Our partnership with LeanTaaS has played a valuable role in this journey. Their team worked closely with us from day one, helping us interpret our data, make informed decisions, and get the most out of the platform.
Planning the next round of transformation
Moving forward, we continue to push the boundaries of what’s possible at Springhill. First, we plan to tackle two metric-based goals: reducing our geometric mean length of stay (GMLOS) variance by 1.75 days, and further reducing our discharge processing time to 90 minutes. Second, we plan to use the system’s insights to continuously review and optimize processes, with an eye toward increasing surgical volume and further reducing ED boarding times.
This isn’t just a story of technology—it’s a blueprint for how AI, when paired with strong clinical leadership, can unlock sustainable capacity. By embracing proactive planning, real-time visibility, and collaborative workflows, we’ve not only improved key performance metrics but laid the foundation for a culture of continuous improvement. Our journey is far from over—and the results so far are only the beginning.
