Predictive analytics: increasing efficiency and financial vigor

Predictive analytics (PA) is becoming an important part of the revenue cycle in hospitals. As capitation and defined payment systems replace fee-for-service, looking at current practice and where changes can be made will become increasingly important.

“PA looks at a patient’s disease and characteristics and then at how we deliver healthcare,” said Charles Macias, MD, chief clinical systems integration officer at Texas Children’s Hospital in Houston. “In the revenue cycle we can focus predictive analytics on the physical care delivery but still be aware of the financial side. Our priority is to deliver effective and efficient healthcare, but (also) understand how to align that with reimbursements.”

[See also: UnitedHealth touts predictive modeling as solution to healthcare fraud and preventable hospitalizations]

PA looks at groups of patients, practitioners and diseases with an eye toward improving both hospital finances and patient care. It looks at patient populations and helps sort out efficient ways to treat. This speeds them through the hospital stay, prevents problems and converts into lower costs.

“With the advent of electronic medical records there are lots of available information and we can identify patients that are high users of hospital services,” said Jonathan Silverstein, MD, vice president for biomedical research informatics at NorthShore University HealthSystem in Evanston, Ill. “We can dig in and find where material and services costs are going. By helping decision-makers understand downstream cost implications, there can be discussions on doing things differently.”

NorthShore’s hospitals, for example, were testing every patient coming through the door for methicillin-resistant staphylococcus aureus (MRSA). While this practice reduced the incidence of MRSA, it did so at a high cost that was not being reimbursed.

Using predictive analytics, the hospital developed a model that looked at patient characteristics to identify likely carriers. This allowed them to reduce the cost of testing by 50 percent while still retaining a reduction in cases.

[See also: Fraud prevention through prediction]

“It is clear that getting the right things to the right people in a more efficient way drives down hospital costs,” said Silverstein. “PA is an area that currently has some hype, but is a real pathway toward efficiency in the system.”

Unlike some other methods, PA can also take into account quality issues. One of the main domains of quality outlined by the Institute of Medicine is that of efficiency which in turn, has financial implications. If a hospital provides the same or better services in a shorter period of time, it not only improves care for the patient, but is also a better financial model for the hospital.

“PA is the intersection where dollars and cents meet science,” said Macias. “We can predict which patients are going to need an intervention and which areas in the hospital have high ordering rates. We can tailor quality improvement strategies to address the best targets. The outcome is lower costs of delivery while creating higher quality care.”

Having the knowledge made possible by PA becomes more important as capitated payment expands and the hospitals receive a fixed amount for each admission.