I had the privilege to moderate a Healthcare Informatics webinar yesterday, August 14, and it impressed upon me the true power of data—even in small amounts—and how it can be harnessed in ways to inform and transform the practice of healthcare. In our post-ACA world, healthcare organizations large and small are on the gradual path toward forming accountable care organizations (ACO) to provide value-based care for patient populations. In July the Department of Health and Human Services announced a list of 89 ACOs joining an initial group, for a total of 154 to date (including 32 Pioneer ACOs), now participating in the Medicare Shared Savings Program (MSSP) under the federal Centers for Medicare and Medicaid Services (CMS).The webinar I moderated, “Using a Clinical Registry to Track Outcomes & Quality Measures,” provided a case study on a small pilot at the Cleveland Clinic Heart & Vascular Institute and the Cleveland Clinic Bariatric & Metabolic Institute performed on data mined from a clinical registry (Remedy Informatics, Sandy, Utah). The principal researcher and cardiovascular medicine fellow at the Cleveland Clinic Heart & Vascular Institute, Amanda Vest, M.D., wanted to find out if the heart structure and function improved in a group of obese patients who already had heart failure, after they underwent weight loss surgery. “The formulation of the question is really the key to the entire process,” said Vest.
Vest noted that researchers already knew the benefit of weight loss on the diastolic function of obese patients, but not much research had been done on the benefits to systolic function. Instead of doing a time-intensive randomized study, Vest did a retrospective study using data in the clinical registry gathered from various sources including Cleveland Clinic’s electronic health record (Epic, Verona, Wis.). Twenty-six subjects in the registry were identified as fitting the pilot’s systolic parameters, but only 10 of those patients had pre- and post-op echocardiograms to study. Mining of Cleveland Clinic’s echocardiogram database yielded five more eligible patients for the study.“When myself and a colleague looked at the ejection fraction [the volumetric fraction of blood pumped out of the ventricle of the heart with each heartbeat] and looked at how hearts were pumping on the moving images of the echocardiograms, we were very interested to see that in many of patients, the ejection fraction actually improved from the preoperative pictures, to those more recently,” said Vest. Vest and her colleague Philip Schauer, director of the Cleveland Clinic Bariatric & Metabolic Institute, overall saw a mean increase in ejection fraction of 6.4 percent and a mean BMI decrease of 9.2 kg/m2 in patients post bariatric surgery.Not only will this research using standardized clinical registry data allow Vest and her Cleveland Clinic colleagues to audit their own clinical practices to see assess and treat high-risk patients, results from this pilot study are being prepared for publication and are being submitted for a grant for prospective research. Vest said eventually she’d like to start storing data on readmissions, quality of care outcomes, biomarker information, and more, to help inform future research.
Vest did note the challenges of standardizing data and using unstructured clinical data in clinical registries. “We need to make sure the quality of that data are excellent,” said Vest. “For example those five patients that we had to use another external database to find were missed because they had not been coded with the ICD-9 code for heart failure prior to their surgery, so that’s an area of clinical improvement we’re learning.”The possibilities for clinical registries to synthesize data from different sources—various standalone clinical information systems, EHRs, HIEs—into a common understandable format to track high-risk patients and manage patient populations is truly exciting. Clinical registries will become the platform for this future state of healthcare, and I look forward to reading and reporting more about all of it.
