Data-based physician staffing stands to save millions of dollars for multi-hospital health systems

An innovative data-crunching method for staffing a key clinical department has reduced physician overtime by 13 hours a day and downtime by 14 hours at a large urban health system. 

The researchers who tallied the results estimate daily savings at $8,400 a day. 

Even after adjusting for additional compensation for on-call duties, the analysts figure, the net savings can cut wasted pay by 12%. 

That would translate to savings of more than $800,000 per year institution-wide if implemented across the experimenting health system’s 11 hospitals, the researchers state. 

And that’s just one department. 

The department was anesthesiology. The piloting enterprise was Pittsburgh-based UPMC. 

The researchers describe their project in a study published Feb. 12 in Operations Research, a journal published by the Institute for Operations Research and the Management Sciences (INFORMS). 

Mark Hudson, MD, MBA, and Aman Mahajan, MD, PhD, MBA, of UPMC—together with co-authors Sandeep Rath, PhD, of the Indian School of Business and Kumar Rajaram, PhD, of UCLA’s Anderson School of Management—state their system incorporates “on-call flexibility to address demand uncertainty.”

Practical, fair and accurate workforce planning 

The program pulls this off by breaking its game plan into three tactical stages. From the study: 

• In the first stage, anesthesiologists are assigned to specific locations or an on-call pool several weeks before the day of surgery. 
   
• In the second stage, on-call staff are deployed to particular locations based on demand forecasts received days before the surgeries. 
   
• Finally, in the third stage, overtime and idle time are realized. 

The authors say they designed the model to avoid bogging down in theoretical projections and to, instead, maintain a focus on practical aims with achievable ends. 

For example, the model considers constraints on commuting among and between sites by individual anesthesiologists, the authors note. 

It also incorporates “fairness considerations” for on-call assignments. 

Project managers tap historical data to estimate uncertainty in demand forecasts, and they use a calibration technique to balance optimal performance with conservative expectations. 

Designed for anesthesiology but applicable everywhere 

The resulting system—placeholder name “Multilocation Dynamic Staff Planning”—is generalizable to other areas of healthcare staffing that regularly face challenges in workforce planning, the authors state. 

In a news release, UPMC’s Hudson underscores the point. 

The model “provides a blueprint for other departments and health systems facing similar operational challenges,” he says. 

Mahajan adds: “This kind of dynamic, data-driven approach is critical to improving efficiency in modern healthcare.”

The study is available from INFORMS in full for free.