It’s the rare hospital C-suite executive who doesn’t worry about the federally-mandated financial penalties that can result from not reining in avoidable 30-day readmissions. Several companies have recently developed analytics technology to help counter this costly problem.

ACOMSplus, for instance, has developed a readmission prevention system built on an algorithm that looks at more than 65 parameters that have been linked to the risk of readmission. Shengyong Wang, an assistant professor of systems engineering at the University of Akron, is working with ACOMSplus to commercialize the tool, which looks at risk factors for congestive heart failure such as CPT and ICD-9 codes, patient demographics, the number of previous hospital discharges for each patient, and several non-medical factors, including whether a patient has an assigned pharmacy.

[See also: Bigger readmissions fines hit hospitals.]

Wang points out that on average 25 percent of congestive heart failure patients are readmitted to the hospital within 30 days; the predictive model that he and his colleagues have created data mines 65-plus parameters to determine the probability that patients will come back into the hospital. The program allows health systems to import data from multiple EHRs into one digital interface. Then the software analyzes the data and recommends interventions that use minimal resources.

Since the system is still in its early stages, the company has yet to generate any clinical outcomes data to demonstrate that it reduces hospital readmissions, but initial testing at the Lake Health System in the Greater Cleveland area has yielded a 70 percent accuracy rating said Brian Barberic, vice president of marketing at ACOMSplus.

Other predictive models have taken a similar approach to that of ACOMSplus. Harvard Medical School investigators, for instance, used low hemoglobin and low sodium at discharge, along with non-elective admission, discharge from a cancer unit, one or more hospital stays in the previous years, and other risk factors to help generate a scoring system. The system was based on data from more than 10,000 discharged patients, nearly 2,400 of which were readmitted within 30 days. They used the scoring system to forecast which patients would need to be targeted for “intensive transitional care interventions. ”

Fusion Consulting has also entered the fray with its own analytics technology that the company says can help hospitals manage readmissions for acute myocardial infarction, heart failure and pneumonia patients. Its analytics tools can give “hospital administrators and clinicians the ability to quickly and easily stay on top of readmissions indicators and filter by location and provider including historical trending of actuals compared to goals,” according to CEO Jeff Wilhelm. Additionally, he noted, analysts can explore this information to find hidden trends in the data.

Although there are lots of predictive models out there that aim to detect hospital readmissions, Jeff L. Schnipper, MD, associate professor at Harvard Medical School and the co-author of the aforementioned scoring system, has his reservations about them. “There are so many predictive models out there; some are better than others. But at the end of the day, are we identifying patients who have avoidable readmissions or are we identifying patients who have all-cause readmissions, regardless of whether they are preventable or not?”

[See also: Watson-like tech tackles readmissions.]

It’s also not known whether any of these potential solutions will improve patient outcomes while lowering costs in the real world. Several research projects suggest, nonetheless, that the right combination of tools can have an impact. Project BOOST, a national initiative led by the Society of Hospital Medicine, has developed a toolkit that includes medication reconciliation forms, a checklist for discharge patient education and a checklist for post-discharge continuity checks. This approach reduced 30-day readmission rates by 3 percent in 11 hospitals after one year.

Similarly, a Yale University project was able to modestly reduce readmission rates using partnerships with community doctors and local hospitals, medication reconciliation spearheaded by nurses, and arranging follow-up appointments before discharge, among other interventions.