The information in different electronic health records can be collected to create large scale predictive analyses to reduce unplanned patient readmissions, according to an article in Health IT Analytics.
The article showcases the efforts of Advocate Health Care, a 12 hospital system based in Illinois, which has a 700,000 patient pool and is involved in several accountable care organizations, putting it under pressure to provide high quality care at lower costs.
As a result, Advocate opted to make reducing readmissions a high priority, and created a tool that would collect EHR data from its systems in order to conduct population health predictive analytics. To achieve interoperability between facilities, which do not all use the same EHR, the program needed to match data to the right person, get the terminology right and formulate longitudinal records, according to the article. Advocate then developed an automated risk tool that creates a readmission risk score for each patient.
"We have a predictive analytics model for readmissions that allows the care manager or the clinician to review a risk score as part of their EHR workflow," Tina Esposito, VP of Advocate's Center for Health Information Services, said. "It's updated every two hours during the patient's stay. It's not a separate application they need to go to, and it's not information that was calculated based on claims data that is six months old. It's real-time, and it leverages the clinical data to really identify what that readmission risk is for the patient so their clinicians can act appropriately."
The tool is beginning to bear fruit; Advocate has gone from a 40 percent compliance rate to about 75 percent, according to Esposito.
Research has shown that EHRs can be used in predictive analytics to help pinpoint in real time which inpatients are at high risk for readmission and do a better job than claims based models. The data also can be pulled from disparate systems.
To learn more:
- read the article