Medical laboratory test data is one cornerstone of this data mining activity and pathologists are well-positioned to take a more prominent role in helping clinicians use lab tests more effectively
Data mining has arrived at many hospitals and health systems. The goal is to mine large quantities of clinical data to identify useful patterns that can guide clinicians to intervene with specific patients. This trend creates a big opportunity for pathology informaticians to step forward and contribute to improved patient care in significant ways.
Progressive healthcare networks, such as the Carolinas HealthCare System, are creating a lot of buzz for their aggressive and smart use of technology and data. The idea is to ultimately reinvent care—in response to new payment schemes—by preventing acute and critical episodes that may require inappropriate readmissions.
Clinical laboratory test data will always play a role in this initiative. Pathologists can play essential roles in these efforts to analyze clinical data and identify useful clinical interventions.
In-house Data Warehousing Helps Reduce Readmissions
For example, the Carolinas HealthCare System predicts it will lose $5 million in 2014 on $4.8 billion worth of revenue. This would be its first loss in 30 years, according to a story published by the Charlotte Observer in early March.
That is why this health system, with the dual goals of cutting costs and keeping its population healthy, has turned to big data to flag people most likely to be unexpectedly readmitted. In fact, it was in 2012 when the Carolinas HealthCare System created its own advanced analytics in-house data arm, called Dickson Advanced Analytics Group.
This business unit is organized to focus on population health management, individualized patient care, and predictive modeling. It integrates clinical, financial, and operational data across the healthcare continuum, according to the health system’s Website.
Predicting Readmission Risk with 79% Accuracy
The Dickson system uses a 10-terabyte enterprise data warehouse to see trends. It has predicted readmission risk with 79% accuracy by using statistics that are updated hourly to correspond with patients’ records, reported Modern Healthcare in a story published in December.
“This is where the ability to analyze data sets becomes very valuable. We are able to risk-stratify patients. It allows clinicians to focus on those patients who are at the highest risk of coming back to the hospital,” said Allen Naidoo, Ph.D. (pictured above), Vice President of Advanced Analytics for the Carolinas Healthcare System, in a story published by Modern Healthcare. (Photo copyright Carolinas Healthcare System.)
The health system has already generated a return on that investment. Since July 2013, 20,000 Carolinas patients were reportedly assessed, resulting in more than 30,000 care interventions.
In describing a care intervention, Zeev Neuwirth, M.D., Chief Medical Officer and Senior Vice President, Physician Services Group of the Carolinas physician practices, paraphrased caregivers in the Charlotte Observer story: “Nothing’s happened yet, but you are at risk based on data from your medical charts. I know you feel fine, but we want you to go into our care management program.”
Providers Turn to Outsourcing to Alleviate Readmission Penalties
Other hospitals and health systems—after getting stung with penalties associated with readmissions of Medicare patients—have hired emerging new companies in the data warehouse realm. For example, St. Dominic Hospital in Jackson, Mississippi, engaged Engineered Care, of San Francisco, according to a story published last summer in The Wall Street Journal.
The firm reportedly helped the 535-bed hospital reduce Medicare readmissions for selected types of healthcare by 1.25%, to about 8.5%. Engineered Care was established in 2009 specifically to bring to market Project RED, an evidence-based discharge protocol developed at Boston Medical Center. The company said it has developed evidence-based care protocols based on that earlier model.
Effective in 2012, the Affordable Care Act (ACA) penalizes hospitals for excessive Medicare patient readmissions related to select conditions. These include acute myocardial infarction, heart failure, and pneumonia. In fiscal year 2015, ACA will expand this law to four more conditions, including chronic obstructive pulmonary disease and coronary artery bypass graft surgery.
Several Thousand Hospitals Hit with Medicare Penalties in 2013
About 2,225 hospitals were fined more than $227 million by Medicare last year, according to federal data. At least 2,215 providers faced penalties reaching about $280 million in 2012. Such penalties have caught the full attention of hospital and health system administrators.
This was a powerful motivator that encouraged innovative health systems, such as those mentioned above, to begin mining clinical data specifically to address the problem of unnecessary readmissions. In so doing, these same institutions are opening a door for those pathologists who want to assume greater roles as consultants to physicians. This can be a win-win for both the pathology profession and health systems, since pathologists have the training and experience to help physicians better utilize medical laboratory tests for diagnosis, prognosis, and patient monitoring.
—By Donna Marie Pocius
Related information:
Carolinas HealthCare seeks to cut costs through prevention, technology
Carolinas centralizes data analytics to reduce readmissions and redesign care
