Health First streamlines discharge with AI-powered analytics

In two years, the organization eliminated 517 avoidable days monthly, reduced length of stay by six hours on average per patient, and eliminated 200 hours of data collection per week.
Health First Central Florida AI analytics

Health First's Heart Center in Central Florida. (Credit: Health First)

Health First is a nonprofit, Central Florida health system consisting of more than 900 beds across four hospitals, with the only Level II trauma center in the county.


Streamlining and automating the patient discharge process is a major focus in gaining efficiency among hospitals. Health First is no different, as it has struggled to manage patient flow and manual discharge processes across its four hospitals.

Manual discharge processes, dependent on paper trails and ad hoc verbal communication, resulted in siloed information among different care teams, creating patient-flow bottlenecks and discharge delays. This clouded staff's visibility into attaining the latest status updates and identifying existing barriers for discharging patients.

"For example, we had multiple systems where ancillary departments relied on EHR charting, but they couldn't tell what was outstanding or what was complete," said Patti Canitano, division director of patient throughput at Health First. "All discharge documentation was paper-based, and that paperwork had to be reviewed in a huddle or shared with other care teams via scanning, emailing or faxing to ensure timely communication."

Supervisors were creating 20- to 30-page patient lists daily that care teams relied on to see who was being discharged and when. However, the data quickly became outdated and inaccurate. Support services did not know when to expect admissions or when discharges were in progress, leading to a shortage of staff struggling to support the patient flow.

"We needed to automate this process to gain real-time insights and create meaningful impact by reducing avoidable days and lengths of stay," said Canitano. "These inefficiencies came to a head during COVID-19. Accurate, timely and proactive operations were more crucial than ever. We needed to optimize length-of-stay management to better plan for capacity constraints, avoid capacity expansion and track our COVID patients versus the normal population."

Health First did not have the digital means or data transparency to complete this in a timely fashion or function with patient throughput processes at the level of efficiency that COVID demanded, she added.


Hospital IQ is a vendor of predictive hospital operations software. Health First opted for the company's system, which offers a way to streamline discharge management. The system uses artificial intelligence to drive action across care teams in real time by providing situational awareness to future capacity constraints, identifying and orchestrating discharge activities, and communicating actionable insights from discharge barriers to bottlenecks.

"The solution's prescriptive recommendations and AI-enabled workflows would help our staff across the hospital to expedite the discharge process, reducing unnecessarily long patient stays and avoidable days," Canitano explained. "By integrating machine learning-based AI and forecasts with our own data collected across our four hospitals, Hospital IQ predicts future patient demand while automatically prioritizing specific patients for discharge."

"Our success to date proves that the use of an advanced analytics platform can play a huge role in helping hospitals switch from a reactive approach to a proactive approach that prevents patient flow problems before they arise."

Patti Canitano, Health First

By predicting and prioritizing discharges, the care teams quickly could align on the high-priority patients and would be provided with the insights needed to manage throughput and remedy capacity challenges before they arise, she added.

"Additionally, the vendor introduced us to their COVID-19 Regional Forecast Dashboard tool, which is currently helping other hospitals plan for upcoming surges of patients," she said. 

"Lastly, to best serve our specific data-tracking and COVID-related reporting needs, the vendor created a COVID Command Center specifically tailored for our health system in just a few days – putting proof behind the vendor's promise of technological agility and marking the start of our growing partnership."


The provider and vendor organizations implemented the platform and focused on streamlining and automating inpatient operations.

"We converted manual processes to digital through intelligent automation – using technology that streamlines standard operating practices while using AI to detect risks and direct actions in real time," Canitano said. "This way, our multifunctional teams, who traditionally operate independently, became easily connected with streamlined workflows for enhanced collaboration to speed up patient throughput processes.

"Implementation was easy and seamless," she continued. "The discharge automation-focused rollout started with our patient flow administrators, or house supervisors, who used Hospital IQ during rounds to identify patients ready for discharge. 

"Some associates initially had some reluctance about adopting this new tool," she added. "But after the first week or so, the team reached a level of confidence in their use of the digital solution, which completely abandoned their paper and fax processes."

As usage of the tool increased, Health First has worked with the vendor to integrate additional data sources from across its IT infrastructure. The result provides teams with a consolidated view of each patient they could not achieve otherwise, she said.

"To serve timely tracking and reporting needs driven by the COVID-19 pandemic, Hospital IQ also provided automated tracking of our COVID-19 patients," she noted. "The platform combines dozens of data sets to power our COVID-19 command center with a centralized, real-time and predictive view of operations."

Through a single source, Health First has immediate access to patient population data and insight into bed availability, which is needed to successfully navigate patient surges, while maintaining efficient throughput. The COVID-19 command center is also enabled to deliver on state and federal reporting requirements because of these data insights.


The AI-fueled technology is changing the way Health First operates, moving away from delayed data analysis and siloed improvement initiatives to digitized, more coordinated efforts.

"Within the first 90 days following implementation, we exchanged more than 15,000 notes across clinical care teams that previously depended on paper processes and phone calls," Canitano reported. "In the two years since implementing Hospital IQ's solutions, we eliminated 517 avoidable days monthly, reduced length of stay by six hours on average per patient, and eliminated 200 hours of data collection per week."

By developing a digitized transfer of information across all staff and shifts, Health First saved over an hour per person per shift, prioritizing patient discharges, and drove two additional discharges daily, on average.

"Our success to date proves that the use of an advanced analytics platform can play a huge role in helping hospitals switch from a reactive approach to a proactive approach that prevents patient flow problems before they arise," she stated.


Understand that one cannot improve what one cannot measure, Canitano advised.

"Assess all the paper processes currently in place and the time frame that it takes to complete and collect data," she suggested. "Then, look at when that data is used and for what. We saw that we would start to collect data on our discharge patients and outstanding issues prior to 7 a.m., which usually happened overnight, and that information was used at our 8:30 a.m. huddles and 10 a.m. rounds."

Staff found there was a huge gap in the data.

"Our ancillary departments were missing the most updated discharge list and were performing morning labs, physical therapy and diagnostics on patients that were no longer pending discharge, or they were missing patients added to the discharge list during that time frame," she noted. "There was no reliable source of data control or single source of truth."

Start small, she added, by identifying the largest manual opportunity for improvement and tackling it. Do not try to take on everything at once.

"As you start to have your initial wins and your team's comfort level and product knowledge grows, and your support team develops, begin having brainstorming sessions to ultimately form an interdisciplinary team that proactively identifies any remaining barriers for discharge management, cross-team communication, and data metric collection for continuous optimization," she said.

"Collaboration across locations is key to optimizing rollout."

Finally, put training processes in place to ensure the pilot location is fully capable of training other hospitals or locations as its staff was, she said.

"One way to do this is by choosing a champion of the technology to oversee rollout across your full enterprise," she concluded. "Internal champions make adoption more successful by not only ensuring the proper training is completed, but by advocating the benefits and value of the technology as seen thus far at the pilot location."

Twitter: @SiwickiHealthIT
Email the writer:
Healthcare IT News is a HIMSS Media publication.

Women In Health ITResource Center

Stay Informed

Subscribe today to receive our FREE monthly e-newsletter