It’s been said, “knowledge is power. Information is liberating.” We’re living in an era when it’s estimated the U.S. healthcare industry generates approximately 30 percent of the world’s data (BridgeHead Software). In today’s digital world it is not enough to just collect and report that knowledge. To move to industry leader status, behaviors must shift to analysis, prediction, and action to get the right information into the hands of the right person at the right time to decrease patient readmission rates and increase satisfaction.
Undoubtedly, over the last several years the industry has experienced unprecedented change. The Affordable Care Act has spurred new programs. One example is the Medicare Hospital Readmissions Reduction Program (HRRP), which provides financial incentives to hospitals to lower patient readmission rates.
Yet, reducing readmissions appears to be a riddle for many healthcare administrators.
The Centers for Medicare and Medicaid Services (CMS) reported in a health and policy brief from the Robert Wood Johnson Foundation in November 2013 saying, “historically about one in five Medicare patients discharged from a hospital are readmitted within 30 days.”
The Centers for Medicare and Medicaid Services (CMS) also recently fined a record number of hospitals – 2,610 – for having too many patients return within a month for additional treatments, according to federal records released in October 2014. In fact, CMS has estimated that total fines will hit approximately $428 million from October 1, 2014, to September 30, 2015, up from $227 million the previous year.
By no means is this a comprehensive outline of the issue, but it does highlight the need for administrators, clinicians, and staff members to be armed with real-time information that can liberate them to adjust processes and create patient programs designed to alleviate readmissions and increase patient healthcare.
After reviewing a wealth of written information on readmissions and how to reduce patient returns using a data-centric approach, it’s apparent there are distinguishing characteristics for best practices:
- It is important for a healthcare organization to build a framework to support a self-service analytics approach.
- The approach needs to start with the end in mind.
- Understanding the challenge of patient readmission, the decision making process of the healthcare provider and ultimately the data sources needed to support the decision process are critical to establishing the framework.
- The framework must be agile, so new data sources can be added as they are identified.
- The framework also should provide a user-friendly toolkit that’s aligned with the appropriate data governance measures.
Couple with each of these best practices: the patient’s quality of care is the priority.
Information on readmissions also has the following lessons learned and recommendations for administrators who want to leverage data are the following:
- Fund and manage analytics as an organizational priority; it is not a project.
- Healthcare organizations must view data as one of the most valuable assets that they hold.
- Stop the proliferation of redundant data silos. One of the biggest challenges for making full use of IT in healthcare is that vast amounts of data have been locked away in many disparate systems – or never digitized in the first place.
- Improve data quality by holding individuals accountable for specific data sources. Poor-quality data is the number one reason physicians are reluctant to participate in performance measurement programs.
- Add, or outsource, staff to develop a data management and governance approach.
- Create an approach that allows for experimentation and iterative learning within the bounds of security and compliance standards.
- Move toward a new business analytics architecture that goes beyond a relational database.
Creating a data-centric culture
Healthcare professionals need to find a way to ensure patients won’t get on the boomerang track. Predictive analytics solutions are designed to help anticipate a variety of factors to help improve both the quality and efficiency of delivering healthcare. This means creating a data-centric culture.
In can be done. Carolinas HealthCare System is leveraging predictive analytics to evaluate the risk of readmission for patients. Based on information culled from 200,000 patient discharges, the healthcare provider built a predictive model now used to evaluate the risk of readmission for each patient. The model has been applied to more than 100,000 patients, making it possible for providers to apply care based on actual factors tied into the actual risk of readmission. Now their knowledge is not only power, it’s liberating to be able to make critical decisions in patient care.
Andy Brockett, Director, Business Innovation