By Warren Strauss, Director, Advanced Analytics Applied Research Group, Battelle

It’s easy to claim that your hospital quality improvement plan is “data driven.” But do hospitals really have the data that they need to drive effective and meaningful improvement?

Today’s hospitals have access to more data than ever before—and the volume of data collected is only growing. As Centers for Medicare & Medicaid Services (CMS) and private insurers move to value-based purchasing models, hospitals are under tremendous pressure to track and report vast amounts of Quality Indicator (QI) data, from treatment outcomes for specific subpopulations to the frequency of various adverse events.

In theory, all of this data tracking and monitoring will help hospitals identify areas of improvement and make changes that lead to better patient outcomes, lower costs and greater efficiencies. In practice, many hospitals find it difficult to cut through the clutter and find actionable, meaningful information among the data points.

The Problem With Hospital Data
With innovations such as hospital quality indicators from the Agency for Healthcare Research and Quality (AHRQ) and electronic clinical quality measures (eCQMs) from CMS, generating data is easier than ever. However, calculating these indicators and measures in a timely and accurate fashion, and in a format that is usable for decision makers has proven to be a challenge for many hospital systems. Too many hospitals are relying on data that is outdated, hard to use or missing essential pieces of information necessary to give it meaningful context.

In particular, much of the data used for hospital quality improvement is:

  • Too slow. For example, it can take up to seven quarters to get comparative quality data back for benchmarking purposes. This lag makes it difficult to go back and investigate areas of concern or to understand whether the hospital is improving or declining in comparison to its peers.
  • Too complicated: The sheer volume of data produced makes it difficult to analyze and draw connections between data points. Interoperability is a large challenge as well – in standardizing and normalizing data within and between hospitals for the significant variations in record-keeping and coding practices. Unless hospitals are able to extract reliable and trustworthy meaning from the data, they will not be able to use it to guide appropriate quality-improvement decisions.
  • Hard to access: Security and privacy concerns dictate that more sensitive, patient-specific data is delivered through secure channels in accordance with HIPAA compliance standards. This often means that hospital quality data is “locked down” and inaccessible to those who need it most.
  • Not actionable: For example, it’s not enough to know that a specific adverse event is rising. Hospitals need to know what that means, what the possible causes are and what they should do about it.

Making Hospital Quality Data Meaningful
Fortunately, there are new answers available for hospitals drowning in QI data. Sophisticated analytical programs can help hospitals turn piles of disparate data into concrete, actionable insights that can be used for QI planning and decision making. Here are four critical factors that hospital, health network and state association leaders should be looking for when it comes to leveraging analytics for quality improvement:

  • Actionable Insights: Hospitals need analytics tools that allow them to correlate different kinds of data in order to understand causes and effects and predict future outcomes. Models that employ both predictive analytics and risk adjustment can help hospital leaders make proactive decisions to improve patient outcomes and financial returns.
  • Time-to-Value: To make a real difference in hospital outcomes, data needs to be available for analysis very close to the time that it is collected. The closer we can get to real-time data analytics the easier it is to link actions to outcomes and monitor whether or not improvement activities are generating the desired results.
  • Accessibility and flexibility: Everyone across the spectrum of care − from hospital executives to practitioners to those involved in monitoring quality − should have the ability to access the data appropriate to their role when, where and how they need it. Secure, HIPPA-compliant cloud-based programs can provide fast, easy access to clinical performance data and analysis, from system-level insights to patient-specific information.
  • Benchmarking and collaboration: Comparing hospital performance with peers within the system, state or nationwide can help hospital administrators identify areas for improvement and develop meaningful benchmarks. For hospital networks and state associations, these comparisons are a catalyst for increased dialogue and sharing of best practices among members.

With the right tools for predictive analytics and data modeling, hospitals can finally make effective use of the mountains of data they are collecting. By making data more actionable, timely and understandable, hospitals can finally start to use it to drive improvements in patient outcomes, health care costs and financial returns.

Warren Strauss, Sc.M., is director of the Advanced Analytics and Health Research Resource Group at Battelle, a private nonprofit applied science and technology development company. He is a biostatistician with more than 20 years of experience in applying statistical and mathematical design and analysis strategies to problems related to the fields of population health and health care.