Project Description

Physicians Leverage AI and Dramatically Reduce Hospital Readmissions

Professional Health Care of Pinellas is an internal medicine and family care practice. With board certified physicians on call 24 hours, 7 days a week, PHC of Pinellas delivers high quality care to patients in various long term care communities on Florida’s Gulf Coast.

Finally…a solution to a chronic problem

Rehospitalizations among skilled nursing patients are a challenge, putting patients at risk for hospital-acquired infection, medication errors, disorientation, and overall deconditioning. Patients discharged from hospitals to nursing homes face a 21.9% chance of being readmitted. And many of these rehospitalizations are avoidable. By preventing unnecessary readmissions, PHC of Pinellas knew it could improve quality, increase patient satisfaction, and lower health care costs overall.

A colleague suggested the practice leverage SAIVA’s machine learning technology. Owner Fadi Saba, MD, immediately saw the potential to reduce readmission rates among their patients in nursing homes and agreed to try SAIVA in five locations, carefully measuring the results.

Giving physicians an added edge
The SAIVA technology applied artificial intelligence to the EHRs in those five communities. Each patient’s medical record is continually analyzed, looking for risk factors of decline and, ultimately, rehospitalization. Those patients at higher risk are included on a daily report for each location.

Physicians and nurse practitioners get an emailed report every day, listing at-risk patients for each location. “Our practice has 70 to 80 patients in some facilities,” explains Director of Operations Mike Eby, “SAIVA has helped us prioritize our work, giving us the opportunity to focus on high-risk patients.”

Instead of relying on nurses to share what they have observed, SAIVA detects subtle, often undetectable changes in condition that can quickly lead to a decline in condition. “With accurate predictions from SAIVA, our physicians can intervene early and prevent negative outcomes,” explains Eby. This is the first time PHC of Pinellas has tried a technology that pulls clinical data directly from the medical record—and clinicians are buying in. “There is very little effort required as the AI does all of the work in the background to produce the daily emailed list,” observes Eby, “and clinicians can use the reports in different ways.” Often these reports are printed off as physicians and nurses enter a care center, to prioritize their rounds. But the reports are equally as useful when clinicians are not visiting a particular location that day. “Using SAIVA enables us to direct the nursing home’s care team on which residents to watch particularly closely to avoid decline.”

Hitting it out of the park with the help technology
Among the five skilled nursing locations in which the physicians are using SAIVA, there has been a drop in hospital readmissions in every building. “We shoot for 10% or lower among our patient populations,” says Eby, “and we’ve reached that goal in all five pilot buildings.” In fact, PHC of Pinellas has achieved a 6.9% rehospitalization rate as compared to a national average of 21.9%.

Critical to the practice’s success has been team buy-in. “The reports are key, but then it’s what you do with the information,” explains Eby. “We have monthly meetings with our physicians and nurse practitioners to review their results, get their feedback, and ensure engagement. They feel like it’s benefiting their patients—and them.”

SAIVA uses artificial intelligence to improve outcomes and unnecessary hospitalizations by identifying your patients most at risk for near-term decline.

Back to Resources


6.9% rehospitalization rate

Quickly prioritized physician/nurse practitioner rounding

Equips clinicians to guide SNF care teams between visits

“We shoot for a rehospitalization rate of 10% or lower among our patient populations and we’ve reached that goal in all five pilot buildings.”
Mike Eby, Director of Operations

Request a Demo

Identify Patients Most At-Risk for RTH.
Reduce Short-Stay Rehospitalizations up to 52%.

    Get an Evaluation

    Identify Patients Most At-Risk for RTH.
    Reduce Short-Stay Rehospitalizations up to 52%.