At the Houston-based Texas Children’s Hospital (TCH), Charles Macias, M.D., long believed in the power of data. But while the information was often at the fingertips of leaders at various healthcare systems, the ability to harness, integrate, and process that data was what would really lead to a transformation in the delivery of care.
Several years ago, TCH launched an overall quality and safety strategy, including implementation of an electronic health record (EHR) from the Verona, Wis.-based Epic Systems, to collect raw clinical and financial data. The objective was to transform the data into meaningful information the facility could use in guiding its clinical quality interventions and waste reduction efforts.
Clinicians, however, still weren't able to effectively leverage the data they needed to improve quality of care for individual patients and specific patient populations. What’s more, according to Macias, who is the chief of academic programs and the section of emergency medicine at TCH and the Baylor College of Medicine, paper and EHRs could not process the data into useful, actionable information for clinicians interested in improving quality and promoting adherence to evidence-based guidelines. Many people, Macias says, have the perception that with an EHR, all the data is there at your fingertips. “The reality is that yes, the data is there, but it’s just locked in a way that isn’t easily transformable for the clinicians.”
To meet clinicians' expectations, leaders of the hospital's quality, clinical, and IT departments knew they needed to develop an enterprise data warehouse (EDW), at which point they partnered with Salt Lake City, Utah-based data and analytics vendor, Health Catalyst.
At the CHIME 13 Fall Forum in Scottsdale, Ariz. in October, TCH was presented with the 2013 Transformational Leadership Award by the College of Healthcare Information Management Executives and the American Hospital Association. The award recognizes transformational IT initiatives that improve care and simplify administrative procedures. This year's award honors Texas Children's data-driven approach to healthcare improvement.
“We realized that we couldn’t proceed with the kind of care process improvement and the kind of strategy for condition specific quality improvement without retrieving the data more rapidly,” says Macias. “The ability to build metrics by coming to a consensus with our multi-disciplinary team and content experts was really critical to the whole process for several reasons.”
First, he explains, it created a common language to be able to drive care process improvement. And more importantly, he says, it allowed us to say, “This is how we will define high-quality care over the next 10 years for the delivery of care related to diseases processes.”
Using the example of asthma care, Macias says it was cumbersome at the beginning to come to this common language, as the process was new to a lot of people who prior to this time, operated in silos of service care. “But we have the ability to say, ‘Let’s look at the patient as a whole, as an entity, and let’s look at all the places care could be enhanced.’ And Health Catalyst helps us by transforming those metrics into a meaningful strategy so we could extract data directly relevant to the disease process. This strategy allowed us to archive [the data], create a data dictionary around it, and create alignment through a balanced scorecard that describes what was important and what defined high quality.”
For example, within the area of asthma care, Macias says it was discovered that TCH’s clinicians were ordering chest X-rays at a rate of around 65 to 70 percent, compared to the national benchmark of approximately 35 percent. What’s more, the cross-functional team that was selected to assess and manage acute asthma in the hospital from the time of arrival in the emergency department (ED) to discharge found that according to the evidence, only 5 percent of chest X-rays were indicated in children with asthma.
It’s the technology that makes it possible for the asthma team to analyze data on demand, rather than many months later. Equipped with this near real-time data, the team is drilling down into specific interventions, such as the delay between the time a child arrives in the ED and the time he or she receives the appropriate medications. In six weeks, the team produced a 15 percent reduction in unnecessary chest X-rays. “We knew we couldn’t completely create predictor models, as science isn’t that refined. But we have reduced the number of chest X-rays to a very respectable level. We have been able to minimize quite a bit of waste in other areas too, so overall, this approach in using the analytics has really helped us to manage our disease processes,” Macias says.
Another benefit of having an EDW is that as these data dictionaries were built, commonalities are found across different disease processes. “As an example, if I am looking at chest X-rays with certain chief complaints and outcomes, I can have that definition and migrate it because it is relevant to my patients with pneumonia as well. So that kind of translation of efficiency didn’t really exist with report writing— report writing in prior years was very awkward, and often were full of misinterpretations and confusion. The timeframe for getting elements back and the inefficiencies for the person-dependant report made the whole process cumbersome. Now, the reports are updated every 23 hours and 59 minutes, so the data being seen is not extracted from six months prior, but instead in real-time,” Macias says.