Data. We’re all swamped with it, and it can be a struggle to keep data under control enough to be useful. A recent report found that on average, hospitals produce as much as 50 petabytes of data each year (that’s over 50K terabytes of data!). And almost of that data (97%) goes unused or untapped. 
To get their data under control, Children’s Healthcare of Atlanta (CHOA) has developed an award-winning data strategy that has put them at the forefront of multiple CHIME categories. CHOA received a Level 9 in the Acute survey of CHIME’s 2023 Digital Health Most Wired Survey. This award is the eighth consecutive one for the organization.
CHOA’s superior rankings were for analytics and data management, population health, infrastructure and patient engagement. CHOA CIO Jeremy Meller attributed the CHIME achievement to machine learning algorithm implementation, robust data governance and use of a distinctive tactic for managing strategic data assets.
Meller said that while common approaches like dashboards are used, CHOA has worked to develop more advanced analytical capabilities that are integrated into clinical workflows.
A system of data governance that includes people from across the organization was established, encouraging leaders to become actively involved in what Meller calls their “learning health system.”
Strategic and trusted data assets are more than a spot on a dashboard. The data assets are closely looked at across the organization to determine how to best use them for improved operations. To be declared a strategic data asset, that asset must be an important part of the major goals of the organization and must be approved as such at the executive level. Usability has been improved with a standardized look and feel of self-service tools such as common elements of KPIs, data over time, outlier analysis, outcomes and process measures that affect a strategic goal.
All operational teams are required to have a data analytics team member embedded into their structure. This enables a better understanding of problems so that approaches can be fine-tuned to provide better outcomes.
Clinical decision support algorithms are being developed with machine learning to impact clinical operations. One example is the use of physiological or biomarkers to more quickly identify when a patient’s condition is starting to deteriorate.
Infrastructure investments include a health informatics core component of CHOA’s research organization. The grant-funded research has three units: a data delivery unit, a collaboration unit and an innovation unit. The three units will ultimately be combined into a technology innovation center within the next few years. This investment is important in exploring and developing innovations to improve clinical operations. It will be used to pilot and run real-world simulations of the models they’ve created to determine the impact on operations.
Other priorities related to analytics and data management, population health, infrastructure and patient engagement include hybrid cloud capabilities, dynamic security infrastructure and connecting everything. Meller said that one other huge priority is their digital patient experience, requiring tools and technology, along with redesigning organizational core practices.