Unlock the Full Potential of Inpatient Bed Capacity with AI

Every day, hospital staff do the best they can to navigate the daily chaos of bed management by making educated guesses as to what is going to happen over the course of the day. Relying on team huddles throughout the day, staff pore over Excel or paper spreadsheets to predict how many beds will open up and when. They try to estimate demand for those beds by the time of day, unsure when to deploy “surge capacity.” On some days, this method works out well. However, more often than not, the staff’s best efforts result in long patient waits, unwanted staff overtime, and ultimately lower access to care. 

The problem with traditional bed management is that the common approach of using spreadsheets to get a periodic read of patient flow, then trying to unlock capacity by discharging patients faster, simply does not work. It takes sophisticated algorithms and real-time predictive and prescriptive analytics to shape demand, successfully match bed supply, place the right patient in the right bed at the right time, and identify and address discharge barriers. 

The ROI of Improving Patient Flow 

Inpatient beds are a substantial economic investment — a single bed can be worth $10,000. Keeping a steady flow of patients into and out of beds is a tricky yet vitally important element of the overall management and efficiency of a hospital. It’s also core to providing a positive patient experience. 

Historically, health systems have directed extensive resources to improve patient flow and reducing the length of stay. Avoidable days, or the number of days a patient remains as an inpatient even though he/she/they are medically ready for discharge, can cost a hospital thousands of dollars each month. Avoidable days generally occur because of an avoidable delay, such as not securing necessary durable medical equipment (DME) for a patient post-discharge or not securing a room at a post-acute care facility (i.e. skilled-nursing facility (SNF) or rehab) for the patient to enter once he/she/they are no longer an inpatient. Addressing these barriers as early as possible in the patient stay, so the patient can be discharged to the right next step of their care journey at the right time, is critical to avoiding lengthy and costly delays and turning over new beds.

If hospitals invest in the right easy-to-use tools, backed by a predictive and prescriptive analytics engine combined with the use of Natural Language Processing (NLP), they can proactively identify and address discharge barriers earlier, streamline patient flow, and ultimately improve patient outcomes and the bottom line. Here are four opportunities to improve bed management with AI and NLP-based analytics: 

1) Using sophisticated demand-supply models to assign patient beds

The best way to optimally place patients is to accurately predict and match supply and demand – on a unit-by-unit, minute-by-minute, day-by-day level – every day. Similar to how apps like Waze take baseline predictions from the speed of traffic for each section of the road for each minute of each day of the week, solutions are now available that model current and future bed availability in each unit. Supply and demand must be approached in different, yet compatible, ways. 

Supply: Model the availability and timing of beds that will become available in each unit. By using historical data to mathematically create a “fingerprint” (a model for each unit that predicts the likely number of patients that will be discharged), hospital staff can make concrete placement decisions about individual patients. Since the predictions are augmented by real-time feeds, these decisions will be more accurate and less speculative. 

Demand: Similar to the supply side, create specifically tailored models for “upcoming demand signals” at any time of the day for each element of demand. These elements can include various factors, such as incoming volumes from surgical and emergency departments, as well as external transfers. Models can be updated by real-time feeds that capture any delays or cancellations of surgeries to ensure updated accuracy.

Side-by-side supply and demand models can then be elevated to patient-placement leaders, giving them visibility into upcoming demand and supply for beds. This leads to dramatically better outcomes than a system purely based on reaction. 

2) Make data-driven internal transfer decisions

Internal bed transfer requests are often viewed as an added burden, pushed to the side to be executed only when convenient. However, transfers can actually serve as a strategic lever since they can free up a bed that will be needed in the near future. By utilizing the predictive modeling tool described above, plus moving the right patients to appropriate open beds, placement teams can open up the right slots to meet the expected demand for high-value beds.

3) Forecast demand with surgical smoothing 

On a given day, 20-25% of bed demand is the flow of patients from the OR into inpatient beds. This often results in spikes in the inpatient census. Surprisingly, this flow is in fact more “controllable” than the census contribution from the emergency department, as optimizing the elective surgery schedule with respect to recovery time can yield a flatter inpatient census. The practice of “surgical smoothing” can be done by forecasting the volume and case mix of surgeries, using AI-based tools to develop templates for scheduling. 

4) Use predictive discharge planning to focus case teams and social services

The most common discharge delays occur towards the end of a patient’s stay – typically surrounding insurance, transportation, follow-up outpatient or home care, or if necessary, availability at skilled nursing facilities (SNFs) or extended care facilities. This last discharge barrier has become especially challenging in recent years, as SNFs are suffering from severe post-pandemic operational and staffing constraints.

Many discharge delays can be avoided if case managers were alerted to the problem earlier in the patient’s stay. Historical data regarding avoidable discharge delays can be collected, and a machine-learning model can be used to identify key case attributes that indicate possible discharge delays early on. The most powerful solution also uses NLP, which makes this information actionable to all care team members and unit staff so they can proactively remove barriers and optimize patient flow. This helps reduce costly avoidable days and provides patients with a smoother care journey. 

Each of these pillars of bed management is critical when it comes to improving processes and asset utilization. While each health system will have different criteria and contributing factors for proper bed management and patient outcomes, and while new tools require learning and patience, the investment is definitely worthwhile.

About Sanjeev Agrawal

Sanjeev serves as the President and Chief Operating Officer for LeanTaaS, the leading AI / ML analytics company in healthcare operations. LeanTaaS’s predictive analytics software powers over 130 health systems and 500 hospitals to improve access and lower costs.

Sanjeev is also the co-author of the book “Better Healthcare Through Math”. Before LeanTaaS, Sanjeev was Google’s first Head of Product Marketing and led three successful startups – CEO at Aloqa (acquired by Motorola), VP of Products & Marketing at TellMe Networks (acquired by Microsoft) and Founder & CEO at Collegefeed (acquired by AfterCollege).

ai
apps
software
healthcare
hospital
medical