We have stumbled our way into the new healthcare information age where population and individualized data can be viewed instantly, helping us see into the past, helping us understand the current, and helping us predict the future. Regardless if the drive comes from the organizations’ own desire to improve quality or driven by government regulations, the fact remains that healthcare organizations today are capturing more data now than ever before.
Understanding the healthcare environment, and what a hospital leader wants to measure is not a straight forward exercise. It takes years of knowledge and experience to just understand the terminology, and it takes even more experience to understand how the data can roll up to a meaningful performance indicator. A few of the key performance indicators (KPIs) hospital leaders are looking to better understand with the promise of fingertip analytics is length of stay (LOS), appropriateness of care, discharge delays, and readmission rates. Of course it doesn’t stop there, ultimately the hospital leaders are looking to dig into these KPIs to provide the data they need to make appropriate and timely decisions. We know that hospitals are collecting large amounts of healthcare data, but the complete picture does not always show up in the analytics solutions because of the complexity of the data silos in healthcare. For the analytics to have real relevance to performance improvement, the data needs to be accurate, it needs to be timely, and it needs to be complete.
Health data can become overwhelming very fast if it is not appropriately put into context. In order to make sense of it all, many healthcare providers invest lots of time and money in systems that provide reports, intelligence, and analytics. The solutions have wonderful features from a display, reporting, interaction and mining perspective but providers are still confused with the results. They are questioning the completeness, accuracy, and age of the data; they don’t trust the data, so the health analytics solutions quickly become ignored.
Building Data Trust
Data trust is built by providing your healthcare providers, your decision support team, and your hospital leaders with high quality accurate data paired with near real-time updates. Access to near real-time accurate data allows your team to utilize a health analytics solution to provide supportive and actionable conclusions across your entire organization. The overall analytics solution is more than just a report generation engine, the solution needs to be backed by and used by the people who are your partners in supporting and leading behavioral changes within the organization. When your employees view a report, and decide to change behavior based on that report, the follow-up report needs to reflect the behavior change as soon as possible.
Further insight into the benefits and repercussions of care quality strategies can be gained from having reliable data over a large period of time. Solutions need to link discreet events across a health system, or health region, and also need to link patients and care providers over time and across a system. A quality longitudinal data set, properly linked, will provide insight and conclusions that before simply could not be seen. A quality longitudinal data set also allows for accurate extrapolation that results in better prediction and forecasting.
Readmissions rates is a great example of using a properly linked longitudinal record to spot trends earlier and more accurately. If a large health system, or health region has many HIS solutions that are not properly integrated, it could be days, weeks, or even months before a readmission is reported. If the multiple HIS systems were properly integrated into a data warehouse that built longitudinal records it could lead to readmissions being detected sooner, and more often! Taking the concept even further, with near real time integration, the readmission pattern of a patient could be caught and properly handled before it becomes a costly additional stay.
Real-time data can be the sidekick for an effective health analytics strategy that can lead many additional benefits including: creating competitive differentiation, reducing readmissions, detecting poorly timed discharges, and ensuring compliance.
Painting a Complete Data Picture
It’s hard to know what you don’t know. Is there health data sitting in your organization that is not being utilized? Does your current or planned health analytic strategy integrate all data across the organization? It is very difficult to paint an overall picture of health while missing pieces of the puzzle. A data source that is not relevant today, may be relevant in the future, and you may realize that the ignored data source is the missing puzzle piece that is the catalyst to transformational change.
Including as many data sources as possible, and providing for future flexibility into your integration strategy, provides you a great opportunity to leverage longitudinal decision well into the future. A complete integration strategy that properly supports analytics allows you to include data from various discreet sources. A supportive integration strategy will also allow you to easily connect and include future data sources that were either not in-scope or did not exist when your initial solution was being implemented. This approach also ensures that no stone is left unturned when trying to find answers to questions that have typically not been asked before.
Overall, it is imperative that an organization builds an integration strategy that is comprehensive yet still employs a strong sense of usability and balance. A comprehensive integration strategy is key to success with health analytics and the balance of value versus noise is achieved with effective data warehouse design associated with a comprehensive integration strategy. So, give health analytics another chance, ask yourself if your health data integration is properly supporting your investment in health analytics.