ER staffing coverage woes? Data shows the solution is strategic deployment, not new hires

Here’s a riddle: how can you improve your hospital’s staff coverage without hiring new employees? 

Answer: by working smarter with what you’ve got.

Seems obvious, but our evidence shows that it’s true. The answer to understaffed ERs isn’t to chase ever more hires (if you can even find and afford those new hires.) The solution is simpler and more solvable than most people realize. Hospitals across the country can go far in fixing their coverage issues by more efficiently deploying their existing staff. 

How do I know this? 

Because I’ve been conducting retrospective analyses of ERs across the country to understand the difference between manual staffing schedules and those performed with customized AI modeling for each hospital.

After loading in datasets and customized variables provided by these hospitals, my team and I have used AI to predict patient volume, including the timing of expected peaks and valleys. Based upon those analytics, we have then generated tailored staffing models that optimize each hospital’s existing resources. With each analysis, we have been able to compare the hospital’s traditional ER/ RN staffing models to the ones generated by our AI solution, Polaris AI.  

The results have been remarkable.

We have revealed huge opportunities for each hospital to immediately improve coverage, reap tremendous annual savings, and decrease burnout and turnover. All with quick implementation handled by the Polaris team. 

It turns out that the traditional staffing models aren’t just a laborious exercise for ER Directors. They also don’t perform as well as models generated through machine learning.

“We’re covered up with Covid”

Healthcare staffing is in such a crisis that nobody feels they have time to fix the problem. Time and again, the beleaguered excuse we hear when we point out staffing inefficiencies is: “We’re covered up with Covid.”

And this is plainly true. Emergency rooms everywhere don’t have enough beds on the floor because they don’t have enough nurses to staff those beds. Hospital managers are overwhelmed dealing with soaring patient volume and staffing turnover at the same time. 

Staff and administrators are frustrated. Because hospitals are under so much pressure, staffing leaders don’t feel like they have options. They know that their staffing models are rife with inefficiencies and excess costs, but they’re afraid to do anything about it for fear of upsetting the extremely delicate apple cart.

It’s as if staffing itself is in triage and everyone is just holding on for sheer survival.

The problem is that bandaids only make this problem worse. Faulty staffing models are increasing burnout. Turnover is expensive and leads to even greater staffing imbalances.

So let’s throw away the bandaids and talk about creating some options.

You may think that your situation is unique and unfixable without added staff, but our AI solution would beg to differ. Over and over, we keep finding untapped staffing opportunities in hospitals of all sizes, across all regions.

For example, many hospitals are short on nurses. A given ER may not have 12 nurses to fill a schedule; it may only have nine, and its staffing leaders need to do everything to make sure they don’t scare away those nurses. We show those leaders how they can find coverage with the same patient volume on fewer resources. For ERs that are used to a model with 12 nurses per day, we show how to function seamlessly with 10.

Increased capacity is within your reach right now

Here’s what this looks like in action.

We conducted retrospective studies on three different sized ERs in three different regions of the U.S., analyzing their current staffing models compared to the models generated by Polaris.

In each case, the result produced meaningful savings.

1. We looked at one hospital with 18,000 visits per year in the Southwest.
2. Another hospital with 30,000 visits per year in the Southeast.
3. And a third with 45,000 visits per year in the Midwest.

For each hospital, we compared the ACTUAL scheduling resources currently deployed by the facility for 90 ED visits per day using 12-hour shifts, with the Polaris AI generated staffing model for the same time period, using 12-hour shifts. And we also looked at the Polaris AI shift “start and stop” time distribution.

Here’s what goes into our AI modeling:

Our AI predicts the number of providers needed to meet patient demand. We identify the best distribution of existing providers. Then we generate an optimal schedule based on the recommended distribution. Our schedule balances provider preferences, organizational policies, and regulatory constraints.

Across multiple metrics, Polaris significantly outperformed the manually generated schedules of these hospitals:

• Annual savings for the ER in the Southwest: $268,000
• Annual savings for the ER in the Southeast: $600,000
• Annual savings for the ER in the Midwest: $1.1 million

Yes, we know that you’re afraid of alienating your existing staff if they don’t get the exact schedule they’re used to. But we’ve been able to show hospitals how to engage staff around efficient models.

We can quickly implement our platform and provide options so that you can immediately improve your capacity. Our system has the intelligence that will help make sure your troops are deployed in the right place at the right time 

Analyze my capacity

Intrigued? Let us conduct a free AI comparison of your staffing model.

Here’s how this works:

1. Polaris users provide the relevant historical data; about 10-12 data points. 
2. The AI “machine” integrates the historical data with AI databases to generate a highly accurate hour-by-hour volume prediction for the department.
3. We set scheduling parameters.
4. We review shift distribution.
5. We generate a staff schedule.

In today’s competitive staffing climate, you have to know your options and use your resources wisely. Chasing new hires is far more challenging than properly deploying the staff that you already have.

Contact us for a free AI analysis of your staffing model