Banner
  • Utilizing Medical Malpractice Data to Mitigate Risks and Reduce Claims
  • Industry News
  • Access and Reimbursement
  • Law & Malpractice
  • Coding & Documentation
  • Practice Management
  • Finance
  • Technology
  • Patient Engagement & Communications
  • Billing & Collections
  • Staffing & Salary

The future of capacity management lies in predictive analytics, digitization

Article

Data driven solutions using predictive analytics, AI, and ML are leading the path forward.

The future of capacity management lies in predictive analytics, digitization

There’s no doubt that the U.S. healthcare system is one of the best in the world from a clinical perspective. It’s far from perfect, however,as for years the industry has struggled to overcome the fundamental issue of matching supply with demand. The truth is most health systems operate at the edge of their capacity – just like freeways during rush hour. Meaning a shock to the system on either the demand or supply side will push the health system into a state of chaos and gridlock.

The pandemic was a shock to both the demand and the supply side. On the demand side, tens of thousands of people in a small geographic area suddenly needed intensive medical care. On the supply side, there were shortages – first in PPE, followed by ventilators, ICU beds, regular inpatient beds, and now ultimately for nursing staff. The crisis has also put enormous financial pressure on many health systems. This makes it difficult to maintain the old approach of building more hospitals, more operating rooms, more inpatient bed units, and more ambulatory care facilities. The balance must shift from building more assets to getting more out of the existing assets.

Recently the team at LeanTaaS hosted its second Transform Hospital Operations Virtual Summit. Over 20 speakers from some of the most prestigious health institutes in the country hosted presentations and had detailed conversations to illuminate capacity management for hospitals and health systems alike. As discussed during the summit, though the pandemic forced the healthcare industry to embrace digital tools like telehealth seemingly overnight, there is much more work to be done to digitize healthcare. Traditional scheduling methods and manual workflows are still being used to manage capacity and will no longer cut it in today’s world.

Data-driven solutions that utilize predictive analytics, AI, and ML are leading the path forward. Below are the top reasons why the future of healthcare capacity management lies in digitization.

Creating operating room efficiency

It’s no secret that operating rooms are the financial backbone of any hospital. They are also one of the most challenging areas to effectively manage. In a given hospital, one or more operating rooms will be idle for several hours during the day, yet many surgeries will be forced to take place late into the evening and night. Also, up to a third of the surgeries completed each day are performed by a surgeon who wasn’t the original assigned owner of the block in which it was performed. Finally, many of the surgeries performed each day are classified as urgent add-ons only because they were squeezed in at the last minute to take advantage of supply suddenly coming available.

To avoid these issues, hospitals must stop relying on simple-minded average utilization calculations. The future lies instead in using sophisticated data science to analyze the patterns of actual time used by surgeons to identify large, contiguous blocks of time left unused, blocks that were abandoned at short notice, and blocks that are consistently being released.

During Transform, Novant Health shared how they deployed tools to combat these issues. In May 2020, within a period of six weeks in the midst of the COVID-19 pandemic, they implemented a solution across 138 operating rooms at 16 medical centers, involving over 1,000 physicians and hospital staff members. During this time they accommodated their entire backlog of surgeries postponed from the pandemic in less than 3 months. They also increased their volume by 8% and even more impressively, increased splitter surgeon volume by almost 13%.

Reducing long infusion wait times

Infusion centers face the same unique challenges every single day. In the morning and late afternoons, the centers are quite empty. But from about 11 a.m. to 2 p.m. it can be busier than an airport during Thanksgiving break. Infusion nurse schedulers and managers rely on traditional methods to reserve chairs, such as looking at calendars as if they were reserving a conference room for a meeting. Because many more factors are involved in an infusion appointment than in that meeting, such methods aren’t efficient in creating a supply and demand balance in a scenario with this many moving parts.

Infusion centers need to better predict the incoming demand pattern for every single day going six to eight weeks into the future. Additionally, a detailed understanding of the individual components of the supply – nurses, chairs, pumps, pharmacy and the rules that govern their usage – is essential to figuring out the equation. Lastly, centers need a scalable way of guiding the scheduler to place each patient into the best possible slot for the day on which the infusion treatment is being scheduled. Prescriptive and predictive analytics can perform these advanced calculations and analyze real-time data to help staff safely accommodate patients. At Transform, Michigan Medicine shared how it used an analytics solution to unlock 8% higher volume with 20% fewer chairs during the pandemic.

Unlocking inpatient bed capacity

As a result of COVID-19, the number of available inpatient beds has become the utmost concern for hospitals and health systems around the country. Beds are organized into small units of 12-24, based on the specific needs of patients. Units are constrained to take only certain kinds of patients, and can only operate at a level of capacity dictated by their ability to staff the unit with the right number of nurses. On the demand side, there are three major sources of patients – from the ORs, from the ER, and from transfers from other hospitals.

The surgery roster is clear several days in advance and can help estimate the demand for beds–but this accounts for only 15-20% of total demand. Predicting demand volume from the ER and inbound transfers is much more difficult. Typically, nurses try to manage ongoing bed capacity through nurse huddles. In short, they try to solve an incredibly complex math problem manually through spreadsheets, discussion, and sheer intuition. While these methods often work, they are also high stress and time consuming. Nurses need a better prediction of the incoming volume for each unit, an accurate prediction of discharges, and a scalable way to surface units that are likely to experience pressures hour-by-hour.

Another Transform speaker, UCHealth, deployed predictive analytics tools at the start of the pandemic in 2020. As a result, they have since seen a 37% reduction in the time to complete an ICU transfer, an 8% reduction in the number of excess bed nights, which is worth millions of dollars each year to most hospitals, and a 4% reduction in the time to place a patient in a bed even though the volume of admissions went up by 18%.

It can be incredibly difficult to optimize asset utilization in health systems with simple calendars and process improvement pushes. Dashboards with alerts just do not do enough. But we do have the digital tools available to effectively manage capacity within healthcare.

The path forward is to leverage sophisticated math and algorithms that are intelligent and can continuously learn. It will enable the front line at hospitals to automatically and rapidly make smarter decisions that consistently match the supply and demand for each hospital’s assets throughout the day, and on every single day – not only unlocking capacity, but leading to improved patient access and reduced administrative burdens on hospital staff.

Recent Videos
MGMA comments on automation of prior authorizations
Erin Jospe, MD gives expert advice
A group of experts discuss eLearning
Three experts discuss eating disorders
Navaneeth Nair gives expert advice
Navaneeth Nair gives expert advice
Navaneeth Nair gives expert advice
Matt Michaela gives expert advice
Matthew Michela gives expert advice
Matthew Michela gives expert advice
© 2024 MJH Life Sciences

All rights reserved.