Using Analytics to Benefit Population Health

The United States is experiencing significant shifts in the way healthcare is accessed and delivered. These changes are disrupting traditional approaches to collecting, analyzing and using health data. Ensuring quality health and care provision for an entire community is no easy task, and it is clear that reliance on electronic medical records (EMRs) that are designed and built in an inconsistent way and do not always provide timely, accurate access to patient – and broader health – data is not sufficient. The global healthcare industry is moving towards more comprehensive and powerful analytics at an unprecedented pace, with a consequent need for better execution and the ability to use more complex analyses.

The implementation of a robust, inclusive analytics strategy is complex and needs to be planned carefully. The general view is that technology is more able, today, to support large scale analytics of the type needed to drive most aspects of care and population management. However, underlying infrastructure and systems architecture may not be as capable of providing the necessary quality, reliable and consist data to ensure accurate and effective analysis.

Further, the federal government has mandated the use of new analytics and EMRs for healthcare. Meaningful Use, for example, incorporates criteria that organizations not only have to collect data on individuals and entire populations, but also meaningfully use the data. With the necessary forethought and deliberation, the US health sector could gain significant public and private revenues each year through the ability to more effectively mine data.

There Is No Shortage Of Data, Just Knowledge Of How To Use It

Although the amount of health-related data being collected continues to grow, most information gathered is not being used in an optimal way. Inconsistencies and contradictions in data definitions, terminology, collection, storage and organization, make determination of best practices near impossible. For example, the act of gathering data is very different from the task of synthesizing, harmonizing and using that data to plan and implement improvements in care and outcomes.

This is true across the range of applications from predictive analytics for individual patient care to effective macro-level management of cohorts and populations. If it is difficult to achieve consensus on the definition of a “population” or any relevant terminology then it is unlikely that any tool or care model can be deployed to manage an entire grouping of people.

As is the case with most analytical evaluations, if the input data set is of poor quality, then it is highly likely that any outputs will be equally poor – garbage in, garbage out. The establishment of data and terminology standards, or at a minimum an agreed upon approach to harmonizing data, methodologies and intentions are all imperative for proper use of data. Additionally, not only must care providers understand the population they are working with and their needs, but also reconcile those with the often-competing and potentially conflicting goals of patients, providers and payers.

How Will You Use Your Data And Analytics To Drive Change?

Improved analytics can enable better understanding of medical and clinical conditions and processes, as well as more effective management of care through the adoption and compliance with protocols. Patients will also be better equipped and more informed as they seek to recognize and access the most appropriate care for themselves and their families.

Additionally, insight into how communities are defined, the rules and the criteria by which they are governed further helps with the alignment of data goals and creation of an effective roadmap for care coordination. This is because effective population health management requires engaging patients and communities correctly through added analytic knowledge. Specifically, access to real time data can improve the in-person ten remote patient-physician relationship as well as increase cost-effectiveness in scheduling and resources. Further, additional healthcare knowledge, confidence and access will inherently drive patient activation.

The Analytics Reformation Begins Now

Whether the organizational goal is to improve individual patients’ experience, overall health of a community, lower costs, or to achieve all three components of the Triple Aim, this cannot be achieved without better data and analytics that are focused on supporting the specific mission. Therefore, the collection and use of large volumes of relevant data will become a key basis of competition and growth for organizations in the health sector.

While improved electronic health and medical records systems are required under the HITECH Act and provide the storage for collected information, they are not the complete answer for better data analysis and utilization. Collection methods must become more robust and sharing data amongst stakeholders has to become standard practice. Further, data must be used meaningfullyto achieve improved quality of care, safety, and efficiency.

The entire health ecosystem is at a point of reformation in the United States, with population health at the forefront. To obtain the best results for the greatest number of people, analytics and the ability to support thoughtful analyses will have to become a primary goal of leadership, with those able to act quickly being in the best position for success.

How do we get there?

As was intimated above, the achievement of a clear framework for analytics does not happen automatically, even when an organization uses one or more EMRs or other relevant source systems. Understanding the linkage between clinical, financial, resource and cultural data is even more important as organizations plan for and implement comprehensive analytic capabilities.  Organizations need to develop Analytics Strategies that define the business objectives and processes that need to be supported. Further, the data, reporting and storage requirements to support the business objectives – and the ways in which source systems and their data need to be manipulated – need to be appropriately designed, harmonized, imported and assessable into the analytics engine.

Strategy development takes time, requires detailed engagement with all stakeholders and needs to determine the type of data to be accessed. It additionally needs to be used in a way that can answer the types of questions that an end user will inevitably expect to answer through the use of large-scale analytics, with the goal of bettering individual and community health outcomes.