Five ways hospital associations can drive quality improvement

WarrenStauss-LinkedIn

Warren Strauss, Battelle

Performance benchmarking provides hospitals with invaluable information to drive quality improvement initiatives: how they compare with similar hospitals across different quality metrics, which mitigation strategies have proven to be most successful among their peers, and how these trends are evolving over time. To get the most benefit from healthcare quality improvement initiatives, hospitals need comparative information that is timely, complete and actionable. State hospital associations may be in the best position to help their members design and implement an analytics system that meets all of these needs.

Power in numbers: Lessons from Ohio
Since 2013, the Ohio Hospital Association (OHA) and Battelle have been working together to create a shared analytical platform for Ohio hospitals. The result is Battelle WayFinder QI Dashboard, a cloud-based quality improvement analytics tool that allows hospitals to organize, display and analyze Agency for Healthcare Research and Quality (AHRQ), Centers for Medicare and Medicaid Services (CMS) and other quality indicators, and benchmark performance among hospital types.

The collaboration has allowed the OHA to provide their member hospitals with unique benefits that would be hard to achieve without the umbrella of the state organization and its voluntary data program, in which 100 percent of OHA members have opted to participate. The collaboration with Battelle has brought more than half of OHA members to new level of sophistication in quality improvement analysis. Here’s why.

  • A common platform for analysis: Comparative benchmarking requires a common analytics platform. In many states, individual hospitals or hospital systems each select their own analytics package. That makes it difficult or impossible to share data for benchmarking purposes with other hospitals or systems in the region or state. A standardized, validated analytics package that is used across all hospitals in the state provides a broader base for more reliable benchmarking. OHA members all use the same platform to submit, validate and analyze data. OHA also compiles this data to generate risk-adjusted benchmark quality reports across the entire state. In recent years, OHA has partnered with Battelle to provide actionable analyses based on sophisticated statistical modeling and predictive analytics.
  • Timely information: Many hospitals rely on CMS and AHRQ data for benchmarking purposes. However, reports from these Federal programs lag behind data collection by roughly seven quarters—meaning reports are nearly two years old by the time hospitals get their hands on them. This data lag makes it difficult for hospitals to monitor ongoing progress or measure the impact of quality improvement initiatives as they are implemented. OHA currently provides quarterly reports to member hospitals and plans to move to monthly reports if hospitals desire more timely results. Additionally, OHA is investigating the potential to work with both Battelle and CliniSync, Ohio’s Health Information Partnership, to provide near real-time access to electronic clinical quality measures that are based on the electronic health record. Faster reporting makes it easier to trace cause and effect and understand how hospital performance is changing compared to peers over time. The closer to real time we can make data analysis, the more easily hospitals can adjust their processes in response to that data.
  • A larger comparative cohort: Meaningful benchmarking requires the ability to compare performance to peers of similar demographic characteristics, size and areas of specialization. A large urban teaching hospital is very different from a critical access hospital serving a widespread rural area. Hospitals with distinctive missions, such as a children’s hospital specializing in cancer care, are not likely to have peers within the immediate geographic area or their hospital system that provide a close match for their specific characteristics. Benchmarking against peers within their system with very different patient profiles and areas of specialty will provide limited insight into their unique challenges. However, within a statewide hospital association, it is easier to find facilities with similar characteristics for benchmarking and sharing of best practices. The shared statewide database provided by OHA becomes a conduit where Ohio hospitals with comparable profiles can find each other.
  • Better statistical analysis for rare events and subpopulations: One of the biggest benefits of a shared statewide database may be the ability to generate statistically significant data for specific subpopulations or for rare or infrequent events. For example, a single hospital or hospital system may not have enough data to draw conclusions about rare adverse events such as mortality rates in low-risk diagnosis-related groups (DRG) or the hip fracture mortality rate. However, when looking at data across all members of a state hospital association, it’s easier to see patterns that can help hospitals design effective mitigation strategies. Larger pools of data will also generate more meaningful and statistically significant results for small subpopulations and rare disease types. A small rural hospital system may only have a single patient with chronic myeloid leukemia, but shared data will allow them to look for patterns among similar patients across the entire state.
  • A shared mission: Perhaps the most important benefit a state hospital association brings is a deep understanding of the issues faced by hospitals in the state and a strong commitment to the success of each and every member. Commercial entities that provide shared benchmarking data among other hospitals that use their analytics platform can provide some of the benefits described above. However, organizing data collection, analytics and reporting at the state hospital association level provides distinct advantages over purely commercial entities. Instead of comparing to a non-representative selection of hospitals nationally who happen to share the same analytics platform, hospitals will be able to look at data for all of the hospitals within their state. These data are both more complete and more comparable, as state-level initiatives or reporting quirks may lead to differences in reported outcomes between hospitals across state lines. Perhaps more importantly, hospital associations are non-profit entities focused on the needs of their members. The association has bargaining power to influence the development of statewide reporting tools so that they better meet these needs. For example, OHA and Battelle have worked very closely together to
    develop new reports that make it easier for Ohio hospital administrators to find the data they need and understand what it is telling them.

Getting started: Factors for successful association initiatives
How can other state associations emulate Ohio’s success? Here are the steps that OHA recommends.

  • Define a data collection mechanism: In some states, the state hospital association may already be the collecting entity for hospital quality data for state reporting purposes. In other states, the state government may have a data collection mechanism that associations can tap into. If these mechanisms are not already in place, the first step is to establish the state hospital association as a trusted data steward and define a method for data collection.
  • Find a partner with statistical expertise: Analyzing the vast amounts of data provided by members requires sophisticated analytics that is beyond the capabilities of most individual hospitals, associations or state government agencies. Partnering with an organization with expertise in statistics, predictive modeling, algorithm development and data visualization is critical. For OHA, Battelle’s status as a non-profit research organization was an important factor in the selection. Rather than simply purchasing an off-the-shelf analytics package, OHA was able to work with Battelle to present predictively modeled quality measures, risk stratification analyses and comparative benchmarking in a way that is actionable by hospital quality leadership.
  • Make sure data collection and reporting meet all federal requirements: Your data collection initiative should not double the work hospitals are already doing for CMS or state agencies. It’s also critical to make sure that your data collection, storage and reporting systems don’t put protected patient data at risk. Both Battelle and OHA have deep experience with hospital QI data, and cybersecurity, ensuring a HIPAA- and FISMA-compliant platform to meet hospital needs for advanced analysis of patient encounters.
  • Get hospitals on board: The most important step for launching a statewide benchmarking and quality improvement initiative is getting member participation. For OHA, that meant a high-touch approach when rolling out the initiative to member hospitals. From the beginning, OHA engaged in active conversations with members to gather their input, demonstrate the program and discuss the benefits. That conversation continues to this day as the association gathers feedback from members and provides additional training and support.

The OHA data program continues to grow and evolve as national healthcare policy and member needs change. Over time, we are generating a rich dataset that will provide members with unprecedented insights into risk factors and best practices. Ohio hospitals are already using this data to perform predictive analysis and project future patient outcomes and financial returns. The benefits of the initiative are only beginning to be realized.

Print Friendly, PDF & Email

Tags: , , , , , ,

Trackback from your site.

Leave a comment

You must be logged in to post a comment.

Get our eNewsletter

Email Marketing powered by StreamSend

Please follow & like us :)

RSS
Follow by Email
Facebook
Google+
https://us.hitleaders.news/five-ways-hospital-associations-can-drive-quality-improvement">
Twitter
©2017 HIT Leaders and News LLC. All rights reserved.