There’s a lot of talk about best practices in radiology and artificial intelligence (AI) these days.  Exactly how can AI-driven technology support best practices for patient access and revenue cycle management (RCM) when it comes to radiology?

With today’s focus shifting toward maximizing the patient experience and alleviating any roadblocks to collecting timely reimbursement where due, harnessing AI-driven, cloud-based technology to improve administrative workflow makes the most significant impact on revenue  Dollar for dollar, improved automation generates more revenue and less administrative burdens, in a reduced time frame.

Specific Areas for Improvement Focusing on AI-Driven Automation in Radiology

Breaking down the patient encounter, we find two areas where administrative inefficiencies have led to increased costs and decreased revenue capture—first, the patient access process where patients present for treatment referred by an ordering provider.  Whether through a hospital-based department or an outpatient imaging center, prior authorizations need to be obtained, Clinical Decision Support Mechanism (CDSM) certificates must be available for Medicare patients, and valid estimation of the patient portion due must be collected.

And second, the RCM process entails not just strong coding and billing, but also AR optimization practices that manage denials and a robust insurance discovery process to identify unknown coverage for self-pay and charitable care patients.

Incorporating AI into Your Patient Access Workflow

Prior Authorization (PA)—The recently released 2019 CAQH Index alarmingly noted that costs associated with processing PAs manually rose to an average of $14.24 per incident while using an automated process estimated costs to be $1.93.  With advanced testing referrals on the rise, as well as the denials for those tests (the most denied test overall is an MRI of the Lumbar Spine), costs and processing time continue to soar.

Implementing an AI-driven PA software allows your radiology practice to determine if a PA is required at the time of service, submit a completed PA directly to the insurance payer, and monitor/follow up as needed in real-time.  Upon completion, notifications are sent directly to the scheduling department, allowing patients to be scheduled efficiently and significantly reducing the need to reschedule.

Clinical Decision Support Mechanism—While various specialties are impacted by the new CDSM guidelines implemented on January 1, 2020, it is the field of radiology (as the actual furnishing provider) that is at risk of being monetarily penalized through rejected claims for Medicare patients.

By automating the CDSM process with AI-driven software that can provide both a proactive and reactive workflow, your radiology practice (through your EHR, EMR or RIS) can link with your team of referring providers through a cloud-based integrated package.  This allows Appropriate Use Criteria to be obtained and CDSM compliance certificates generated so that billing are processed without delay.

Special Note: The CMS recently updated the mandatory implementation date from January 1, 2021, to January 1, 2022, recognizing the complexity of the program required additional time for training and technological workflow challenges.

AI-Driven Automation in Radiology RCM

AR Optimization—Once a patient has been seen and a claim has been coded and billed, there are inevitably denials and rejections that prolong payment if not outright stop revenue capture.  These rejected claims make up the AR and must be worked individually to ascertain the problem and then collect the necessary information before resubmission.

Unfortunately, according to the Medical Group Management Association (MGMA), somewhere between 50% and 65% of these claims are never reworked, and the associated revenue is simply abandoned.  But what if, through an automated AI-driven program, any unpaid claims could be curated through real-time analytics and predictive insights with the most likely to be paid resolved and revenue collected?  The improvements to the bottom line would be significant.

Insurance Discovery—The most frustrating component of RCM has to be bad debt and those accounts that must be reluctantly turned over to a collection agency for potential follow up.  These claims are often due to uninsured patients or unpaid self-pay accounts and end up being written off entirely or categorized as charitable care.

By using an AI-driven Insurance Discovery solution, these accounts are processed through an automated coverage identifier package where patient demographics, insurance profiles, and benefits are verified, and undisclosed coverage is identified.  These uncompensated accounts can then be submitted to the appropriate insurance and revenue retrieved.

Anytime best practices are being used as a barometer of practice efficiency, performance, and the means of getting there are critical to the evaluation.  By looking at patient access and RCM workflows honestly and seeking clarity on improvements that can make long-term sustainable improvements, the reduction in the administrative burdens associated with reimbursement in a third-party payer system leads to increased revenue and a better patient experiences overall.

Contact us to learn more about our multi-pronged approach for best practices in radiology.