Applying Evidence-Based Predictive Modeling to Cardiac Procedures


Predictive Medicine for Cardiac ProceduresPercutaneous coronary intervention (PCI) procedures, commonly known as coronary angioplasty, rank among the most common surgeries performed in U.S. hospitals, with an estimated 2.48 million procedures a year. While that figure has declined 27% between 2007-08 and 2010-11, it still represents a significant portion of all cardiovascular surgeries.

As hospitals adapt to new payment models and treatment modalities, a lot of attention is being paid to evidence-based practices, including those related to cardiac procedures. The ideal of evidence-based practice is to provide clinicians with personalized, actionable information they need to make care decisions that reflect the unique circumstances of each patient. That information should be presented as close to the patient bedside as possible to facilitate in-the-moment decision-making.

McKesson has partnered with Health Outcomes Sciences (HOS), a leader in personalized evidence-based intelligence, to integrate into McKesson Cardiology™ cardiovascular information system (CVIS) several HOS predictive models related to PCI:

  • American College of Cardiology (ACC) models to predict the risk of the following complications:
    • Bleeding
    • Mortality
    • Restenosis
  • 30-day post procedure readmission tool that identifies at-risk patients

These models pull data from McKesson Cardiology and other electronic systems to create a patient picture that helps clinicians make better decisions and improve patient outcomes and satisfaction at a reduced cost, says Max Reverman, vice president of sales for HOS. In addition, McKesson also sells standalone versions of HOS models for acute kidney injury and appropriate use criteria, as well as the company’s eLUMEN display software, which aggregates information for display in the cath lab.

HOS uses a technology platform called ePRISM to deliver predictive risk models, based on published and peer-reviewed research, to the point of care. ePRISM uses validated multivariable risk models and specific patient information to return a risk score that a physician weighs to determine the best course of action for that patient.

HOS cardiac models are based on research performed by medical research centers affiliated with the ACC. HOS has a business relationship with ACC “to translate the College’s risk models for use in routine clinical settings.” You can read more about the partnership here.

“For a PCI procedure, a doctor can determine whether the patient is uniquely at high risk for bleeding or at high risk for kidney injury,” Reverman explains. “With that information, the cardiologist can tailor their therapeutic choices based on those which are most likely to produce the best outcome. For example, anti-coagulation tailored to the individual; a radial versus a femoral approach or a varied amount of contrast dye.”

HOS clients that have implemented ePRISM have seen significant improvements in quality and outcomes. Examples include a client that reduced its bleeding rate from 5.5% to 1.2% and another that saw both bleed and acute kidney injury rates cut in half, according to company literature.

While the company got its start in the cardiovascular space, ePRISM readily accepts models from other clinical specialties. The ability to create predictive risk models from evidence-based research holds possibilities for many other facets of healthcare, Reverman says. That ability will be even more critical as payers turn their reimbursement focus from volume to value.

Just last month (January), Health and Human Services Secretary Sylvia Mathews Burwell announced ambitious plans to move 30% of all Medicare payments to alternative payment models that stress the quality of care over the volume of care by 2016. Just two years later, that payment goal is 50%.

Also last month, six of the nation’s top 15 health systems and four of largest 25 payers came together to form the Health Care Transformation Task Force, an alliance dedicated to accelerating the transformation of the U.S. healthcare system using value-based business and clinical models. While announcing the alliance, task force members “challenged other providers and payers to join its commitment to put 75 percent of their business into value-based arrangements that focus on the Triple Aim of better health, better care and lower costs by 2020.”

There’s no question that changes in reimbursement methodology are coming, Reverman says. “We help in the transformation from volume to value by focusing on quality outcomes and patient satisfaction. Increasingly, those considerations will be part of the compensation equation,” he says.

And you don’t need a supercomputer or a complex algorithm to realize that evidence-based practices that help improve outcomes are the future of healthcare.

Stay informed with medical imaging and image data updates by subscribing to the Medical Imaging Talk Blog today.

Leave a Reply