Leveraging predictive analytics to manage the certainty of constant change

In a 1789 letter to Jean-Baptiste Leroy, Benjamin Franklin famously quipped, "In this world nothing can be said to be certain, except death and taxes."

If Franklin wrote that letter, or email, or tweeted that statement today he might include a mention of any number of certainties about healthcare: the certainty of constant change, continuing M&A activity, cost pressures, etc.

What about certainty with regard to patient volume? Can a healthcare provider ever know for certain what patient or visit volume will be?

In short, yes, and data is the key. Today we have the ability to process huge amounts of data. And, our ability to do this is only increasing; more data, faster processing. A decade ago it was possible to predict volume, but it was laborious and not scalable, at least not without incurring significant costs.

Today it is possible to predict volume months in advance of a shift, with a small margin of error, and do so for a variety of departments, from nursing units to clinics. Additionally, this technology can be used in areas like housekeeping to predict things like pounds of laundry, or any driver of work.

To forecast patient census data, multiple inputs including CDC and Google flu indexes, historical census data, and temperature are pulled in and run through algorithms to produce various time series, vector, and structural models. The inputs used vary by client/area, but essentially anything that can be quantified and results in statistically significant impact can be included. While I've oversimplified it, this process results in a volume prediction, and when run through a departments staffing or workload matrix, a forecasted need by skill by shift is produced.

But the forecast is just the first step in the process. Knowing how many staff you'll need to handle demand is far from actually being able to staff to that demand and do so in a cost-effective and sustainable manner. This seems like it should be the easy part, but there are a thousand moving parts to consider as the weeks tick by and the story of how staff get to the patient's bedside unfolds.

Florence Nightingale, born 30 years after Franklin's death, is known as the founder of modern nursing. She was also a pioneer in healthcare statistics, i.e., actionable data. Data analysis plays a vital role in setting up a framework capable of getting the right person to the right place at the right time (and at the right cost).

By performing an analysis of metrics such as workload, FMLA, "FTE leakage," incidental worked time, as well as analyzing demographic and generational trends, a provider organization can better align core staff (by skill) unit-to-unit in a way that minimizes staff dissatifers and stress-points, like floating and overtime or cancellations, to handle the majority of patient demand.

This same type of analysis can result in contingency staff strategies to supplement your core team. Contingency staff (typically thought of as float pool and PRN staff) are utilized when core staff is unavailable to take a patient or when demand spikes. A solid contingency strategy is one that can expand in times of need and contract down to near zero in times of low demand.

The idea of the "float pool" has evolved from the decades-old model of a handful of nurses who pick up shifts as needed. Today, with various types of skills and personality profiles to consider and the expansion of the hospital to encompass numerous locations, medical practice sites and specialty clinics, and retail care settings, the sophistication of the "float pool" has reached a level that necessitates multiple layers with varying competencies. These layers are best managed centrally utilizing real-time staffing and demand data and deployed at the last possible minute to the areas of greatest need.

The conversation around predictive analytics and workforce analytics is evolving from "what if" to "what now" – how can we really leverage this to make an impact? It's important to remember that these two areas are tools and not stand-alone solutions. At best, they can help provide the blue print, but they must be additions to a bedrock of best-practice strategies. Knowing what will happen, having the resources to handle it, and doing so in a cost-effective and sustainable manner are not the same and take different approaches.

This is where the power of predictive analytics, workforce analytics, and best-practice strategies can converge into the making of a continuum of care that is stable, sustainable, and of the highest quality. Of this, we can all be certain.

Chris Fox is President at Avantas, a leading 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 chris.fox@avantas.com.

The views, opinions and positions expressed within these guest posts are those of the author alone and do not represent those of Becker's Hospital Review/Becker's Healthcare. The accuracy, completeness and validity of any statements made within this article are not guaranteed. We accept no liability for any errors, omissions or representations. The copyright of this content belongs to the author and any liability with regards to infringement of intellectual property rights remains with them.

© Copyright ASC COMMUNICATIONS 2019. Interested in LINKING to or REPRINTING this content? View our policies by clicking here.

 

Top 40 Articles from the Past 6 Months