This article is the first in a series of three that will explore the topic of data mining and its ability to highlight potential healthcare fraud, as well as its effects on healthcare systems and medical providers.
Data mining is the process of finding anomalies, patterns and correlations within large data sets to predict outcomes. Advances in processing power and speed over the last decade have allowed businesses like Integra Med Analytics (“Integra”) to use its data mining algorithms and data analytics to identify highly aberrant patterns that are indicative of Medicare fraud, waste, and abuse. Integra is an organization, unaffiliated with the government that investigates and shares its findings and research through academic publications and news articles. It is an outsider, a new kid on the block if you will, that does not fit neatly into the current definitions, protections, and provisions relating to whistleblowers under the False Claims Act (“FCA”). The traditional whistleblower is in fact an employee within an organization who has evidence of fraudulent activity and who typically recovers 15-30 percent of the government’s total recovery in FCA cases.
Integra acted as a relator to the government in filing a FCA complaint against a Texas hospital system, Baylor Smith & White. (United States ex rel. Integra Med Analytics, LLC v. Baylor Scott & White Health et. al., Case No. 5:17-CV-886-DAE (2020)). It based its entire case on the incidence of secondary diagnoses found in publicly reported data, not by firsthand experience or chart reviews. Mind you, these diagnoses were copious and taken from the data analysts’ goody bag of favorites: encephalopathy, respiratory failure, and severe malnutrition from alleged cases in 2011-2017. These diagnoses are known to reimburse hospitals at a higher rate based on the diagnosis-related group (DRG) it falls within. Further, these DRG payment amounts can be modified by patients’ complications or comorbidity (CC) and/or major complication or comorbidity (MCC). Referencing just the data alone, Integra identified possible upcoding, or inflated secondary diagnosis coding, that increases revenue optimization.
The specific allegations in the recent ruling in this matter involve $61.8 million dollars in allegedly fraudulent claims billed by Baylor Scott & White. The three-judge panel on the Fifth Circuit Court of Appeals summarily upheld the Texas district court opinion dismissing the case since the hospital system:
- adhered to clinical documentation improvement programs encouraged by CMS who state in guidelines, "we do not believe there is anything inappropriate, unethical or otherwise wrong with hospitals taking full advantage of coding opportunities to maximize Medicare payment that is supported by documentation in the medical record"; and
- no specific examples were given to support fraudulent claims, simply data analytics; and
- the contents of a medical coder’s complaints of unethical coding and billing practices, fraudulent directives, training, and guidance against the Texas hospital system were not produced
The court also found Integra’s lawsuit violated a provision of the FCA known as “the public disclosure bar” which states that a whistleblower lawsuit cannot be based on such publicly available information, with no extension into obtaining and receiving specialized expertise to analyze, interpret, and provide expert opinion.
Although Baylor Scott & White won in this case, the expansion and rise of the data-driven whistleblower should keep compliance officers, coding staff, clinical documentation specialists, and the C-suite on their toes. The trend of data miners continuing to use analytics to ferret out potential fraud, waste, and abuse is highly likely. It is imperative for hospital systems, organizations, and providers alike to conduct internal audits of their medical documentation, coding and billing to ensure not only compliance, but that reimbursement was appropriate.