Predictive Analytics And Business Intelligence: What Every Provider Needs In Its Labor Management Arsenal

By Chris Fox, CEO, Avantas

To effectively manage labor in healthcare today, organizations need to be proficient in a number of areas. First and foremost, they need to have the right strategies and methodologies to reduce variances in practice and eliminate wasteful redundancies. The right strategies set the foundation for predictable and sustainable results. Once that is established the conversation can open to the elements everyone usually wants to talk about first: technology, specifically predictive modeling, analytics, and business intelligence.

These are the shiny tools that every health provider needs in their arsenal. Layering these on top of a bedrock of best practices and incorporating them into a well devised resource management end-to-end process provides the best approach to positively impact the bottom line while enabling better, more efficient care.

Predictive Modeling
The promise of predictive modeling in healthcare is immense and its application far reaching, but so far the bulk of possibilities lie in the future. Most of the conversation around the application of all of these algorithms is related to population health, predicting disease, foreseeing the likelihood of readmissions, etc. While it is being employed in general ways here and there and some success is being enjoyed, the bulk of benefits are yet to come.

One area where predictive modeling is going on its second decade however is the forecasting of staffing needs and outcomes. There are dozens of elements that go into forecasting the need for care staff, including CDC and Google flu data, historical census, temperature, and a number of customized variables and input on anticipated events at the local level. Accurate predictions must be established at the time schedules are created so units begin with the best possible starting point. The key then is to continually update the prediction as the weeks tick by and the days and shifts get closer. In doing so, an organization can begin to align its various layers of contingency to meet that demand. Proper contingency layering is one of the strategies I was alluding to above. Strategies must always work in tandem with the tools, supporting them as they automate processes.

Analytics
Hospitals are swimming in data, but most are not utilizing it anywhere near its full potential. Access to data by healthcare providers is a continuum. On one end you have the lack of accessibility. At the other you have data being leveraged for decision support, i.e., business intelligence. All organizations are somewhere along this spectrum.

For data to be of any value it must first be accessible, and then able to be processed in a timely manner, real-time to next day in most cases. There are numerous staffing metrics that provider organizations need to be tracking, including fill rate, contingency usage, extra and overtime, and core staff floating, among others. Two of the main metrics we encourage our clients to focus on are FTE (full time equivalency) leakage (staff not being scheduled, or not working up to their FTE commitment) and incidental worked time (time on the clock before or after a shift or during a scheduled break). These metrics are two of the easier ones to spot and improve on.

Business Intelligence
The main data sources that can be used to improve the resource management end-to-end (monitoring the metrics mentioned above) are time and attendance and payroll feeds. Once those two feeds are processed with scheduling data within your software solution and displayed in an easy-to-understand manner you have the beginnings of business intelligence. Then next step is to figure out how to use this data, who gets it, when, and what will they be allowed to do with it?

With these particular data, the people who first need it are the frontline managers and directors. These are the people who have the biggest day-to-day impact on the provider’s operating expense and are invariably the ones who can be major players in the effort to improve. It’s important to remember that these essential members of the team do not typically have the training to know exactly what to do with the data. This is where finance in particular needs to partner with nursing to provide an educational background to understand the data. Once they know this, they’ll have the wherewithal to understand its potential. Then they have to be empowered to make the adjustments to curb the emergence of negative trends before they affect productivity.

It’s often very easy for others in the organization to dismiss the role nurse managers can play in improving the bottom line with the use of BI tools. Nurses however are surrounded by data all day. From the start of the shift to the end of it they are constantly processing data and making decisions. They are more than capable of diving into the world of scheduling and finance analytics, they just need the initial training. The organizations that I see effectively using data to improve patient care are those that have a high degree of cross-departmental camaraderie.

It’s this idea of cooperation across various groups that must be the cornerstone if any real change is going to happen. No element of technology or strategy is going to make an impact unless there is a focused team working together to push the solutions to their potential.

About the author
Chris Fox is CEO at Avantas, a provider of strategic labor management technology, services, and strategies for the healthcare industry. He is an industry veteran and proven leader who has played a critical role in the company's rapid rise to leadership in healthcare enterprise labor management. Contact him at cfox@avantas.com.