This article is the second 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.
The simple definition of data mining includes the process of knowledge discovery from data. “Mining” further elucidates the work of uncovering trends, commonalities, and patterns in big data. There has been a steady rise in leveraging data analytics in healthcare to benefit patient outcomes. The Covid-19 pandemic has only strengthened the collective need and opportunity to corral big data, to study it, to make informed decisions, as well as to disseminate it for transparency and collaborative initiatives.
Although these are some of the great benefits to data analytics, there are unfortunate perils that affect medical providers based on data mining when used to analyze claims submission data. On a positive note, the sheer sophistication in data mining techniques, with their ability to precisely isolate algorithms, has allowed both our government and commercial payors to gain the knowledge and the freedom to flush out the true fraudsters. While snuffing out fraud is truly commendable, coincidentally, these very techniques can potentially devastate many honest providers as well.
Post-payment audit reviews are conducted through advanced data mining directives, like process-based systems that focus on the complex statistical analyses of billed services and supplies on claims against the providers of those claims. Data mining leads to figurative fingers pointed at physicians; they are labeled outliers; and they are called bad doctors simply because of someone’s interpretation of big data. More often than not, unintentional errors are made when billed claims leave the private practice or facility setting. These errors often involve the appendage of unnecessary modifiers, or inaccurate coding choices. Or, highly specialized physicians unique to a geographic area have higher incidences of billing certain codes compared to their peers because they are the only physician in the area. Unfortunately, none of these explanations or facts are taken into consideration on the payors’ side. Instead, oftentimes the sheer volume of these “mistakes” in coded form sound the alarm bells to the payors’ team of auditors.
Moreover, once healthcare providers receive demand letters from payors wanting to take back original reimbursements based on investigations using data mining, it is often a difficult and up-hill battle. The process to appeal these requests is multi-faceted and complex, with numerous deadlines that cannot be missed. Providers must also bear the burden of additional expenses to retain legal counsel and clinical and coding experts, as well as pay overtime to trusted staff to assist with the lengthy appeal process(es).
There is not a current classification for these types of unintentional errors; nor is there a current classification level for providers who are highly specialized. Rather, they are both simply thrown into the category of fraud, waste, and abuse. It is evident that medical providers that fall into the category of unintentional error need continued education on best practices in coding compliance rather than the harsh punitive measures they are currently experiencing. Further, specialty provider outliers do not deserve to receive overpayment demand letters simply because they are billing more services than their peers due to a lack of competition.
Other examples of providers who face difficult battles defending their work include hospice and palliative care entities that have patients who live past Medicare’s life expectancy of six months or less. These physicians receive overpayment demands simply because auditors data mine and determine that too high of a percentage of patients’ claims are still being submitted after 6 months. Billing companies are also known to append modifiers to charges the provider did not document them to append. These providers face voluminous overpayment audits due to mistakes when the data shows too many modifiers are being billed. Lastly, physicians are often prey to scrupulous vendors that sell their latest devices and new technologies and advise them how to code . These physicians are left holding the bag when the post-payment audits come in when auditors notice a sudden influx of these codes simply because the providers trusted the wrong people and did not do their research.
Unfortunately, the current system of understanding big data through data analytics to ensure fraud prevention allows payors to maintain their positive revenues while severely hurting reputable, yet naive providers.
The vast majority of physicians and qualified healthcare professionals are rendering services to benefit their patients and work tirelessly to follow and support their patients’ continuum of care. The administrative burdens our physicians face are many, and often strangle-hold them. This leads to their disdain and dissatisfaction for their original purpose – providing excellent patient care based on their education and expertise. It is unfortunate that the current data mining trends to better understand big data is becoming the financial downfall of many excellent healthcare providers.
