Learn how analytics can detect healthcare fraud and how it can help insurers and payers.
The National Health Care Anti-Fraud Association (NHCAA) points out that financial losses due to fraud amount to 3-10% of annual healthcare expenditures. Healthcare fraud analytics identifies fraudulent activities and helps organizations proactively prevent them. Research and Markets highlights the global healthcare fraud analytics market is expected to grow at a CAGR of 20.45% from 2022 to 2028.
The NHCAA writes that most healthcare fraud occurs due to the actions of unscrupulous healthcare providers or people who impersonate them. For example, a provider or fake provider may bill for services that never have been provided to the patient. Fraudsters can also “inflate” the cost of treatment. To do this, they code the patient’s diagnosis as more serious than it is and prescribe more expensive treatment procedures.
Also, a dishonest provider may misrepresent a non-covered procedure as covered to obtain insurance payments. Another example of provider fraud is kickbacks and bribes from providers, pharmacies, and other healthcare organizations. The damage from fraud can amount to hundreds of millions of dollars. Patients themselves may also be the fraudsters. They may steal the health information of an insurance policyholder and impersonate them.
In addition to the briefly mentioned cases of financial losses, there are more detailed examples of how fraud affects consumers’ wallets and what other far-reacting problems it can cause.
Insurers utilize ready-made or custom insurance software to conduct annual reviews of clients’ insurance policies. Without incorporated fraud detection and analysis, they may increase premiums for consumers to compensate for preceding losses in the form of inflated payments for fraudulent insurance claims. As a result, consumers may have to “pay” for insurance fraud out of their pockets. It costs approximately 900 USD per consumer annually.
The fraud schemes affect the financial state of insurance and payers and put the patient’s health at risk. The provider wants to make money on the difference between the bill for an expensive service and the cheap service provided. Also, treatment by unscrupulous doctors is fraught with emergency hospitalization for patients more often than it would be with treatment by reputable providers.
It happens that a doctor performs an unnecessary and even destructive medical procedure on a patient to make a profit from the insurance company. And besides the irreversible effect on health, it undermines the patient’s trust in all doctors. They may not fully follow the provider’s recommendations or not want to seek medical care even if there is a threat to health.
The sheer scale of big data and the complexity of healthcare processes make it difficult to monitor the entire medical journey: from new customers’ registration and their diagnoses coding to their bills and claim payments. Healthcare organizations increasingly recognize the importance of analytics solutions to detect and prevent activities that may indicate a fraudulent strategy.
This type of fraud occurs when a healthcare provider adds services to a patient’s bill that they did not provide. For example, a claim may show that the patient visited a doctor although this visit has never been. Another example of fraud is when a provider promises a patient services of a lower quality than what they added to the bill.
Fraud analytics considers data from all available sources, e.g., call center records, patient electronic health records (EHR), and healthcare billing systems. The EHR contains a record from a doctor that the patient was present for the appointment. However, the call center records do not confirm that the front desk staff scheduled the patient for the appointment. So, analytics highlights a discrepancy that insurance may investigate.
Another type of healthcare fraud is when a provider “upcodes” a patient’s diagnosis to make the treatment more expensive. Belitsoft experts confirm that each diagnosis code in the patient’s EHR corresponds to a procedure code (diagnostic, medical, or surgical interventions) compatible with this diagnosis. In case of fraud, a doctor can often prescribe unnecessary knee surgery to patients, although in these cases simple physiotherapy would be enough. They call patients’ conditions serious and enter an “inflated” diagnosis code in the EHRs, which justifies the surgical intervention.
Through EHR data analysis, an insurer sees a surge in claims for knee surgery performed by a specific doctor.
This type of fraud is common in cosmetic surgery schemes. For example, “nose jobs” are not medically necessary covered treatments, while deviated-septum repairs can be performed for medical reasons and are considered an insurance case. Unscrupulous providers can “disguise” a cosmetic procedure to change the shape and size of the nose as a surgical intervention for medical reasons.
Analytics checks the data on the procedures performed by the provider. If in their practice there are many more cases of septum correction (septoplasty) than cases of nose plastic surgery (rhinoplasty), then the insurance company can investigate these cases.
A provider may receive kickbacks and bribes from other providers and pharmacies. For example, a provider prescribes a drug for a patient that they can buy only in a certain pharmacy or recommends to consult a certain doctor. The patient follows the recommendations diligently, and then the “kickback partners” share the profits.
Fraud analytics examines data from the patient’s EHR (in this case, previous and current diagnoses and doctor’s prescriptions) and identifies suspicious mismatches. In addition, the predictive analytics function makes it possible to prevent fraudulent schemes. It analyzes not only historical data but also data in real-time and highlights mismatches in procedures and diagnoses as soon as they appear in the system.
Previous cases of fraud arise due to the dishonesty of the provider. Medical data theft is most often committed by those who seek medical care. This could be willing “card sharing”. An insured patient gives their data to an uninsured person who pretends to be the policyholder. It also happens that an insured patient doesn’t know someone else uses their private insurance data.
Fraud analytics uses historical claims data, such as the type and amount of the claim, the geolocation of the insured person, their description, etc. The reports provide claim details with trends and scores, and analysts can use them to determine whether a claim may be a potential case of medical data theft.
As big data grows, so does the frequency of healthcare fraud cases. Fraud analytics uses contemporary tools and methods to analyze big data, identify trends, and detect suspicious patterns. And users of these software solutions can use the reports to prevent potential fraud. Also, learn about Internet of Things in the Healthcare Industry by reading this article.
Thanks for choosing to leave a comment. Please keep in mind that all comments are moderated according to our comment Policy.
