Healthcare financial leaders are constantly brainstorming ways to increase operating margins through better revenue cycle performance. These efforts often lead revenue cycle leaders to denied claims—when a payer doesn’t reimburse a health system for a service rendered. Although denials are a common reason for lost revenue, experts deem nearly 90 percent avoidable.
Effective denials management starts with prevention. Organizations can use revenue cycle performance data, combined with artificial intelligence, to predict areas within each claim’s lifecycle that are likely to result in a denial. With denial insight, health systems can optimize revenue cycle processes to prevent denials and increase operating margins.
Download
Health systems face mounting pressures to improve operating margins despite new challenges, such as COVID-19, that can erode revenue as quickly as it grows. As a result, healthcare financial leaders are examining revenue cycle performance more closely to find cost-cutting opportunities, including one of health systems’ most common revenue cycle challenges, denied claims.
Denied claims (denials) occur when a payer denies a hospital- or clinic-submitted claim to receive payment for a service rendered, leaving the hospital without payment. Denials result in lost revenue—something organizations cannot afford while surviving on thin margins, especially during a pandemic.
Although denials are a consistent pain point for health systems, Becker’s Hospital Review estimates that 86 percent of denials are avoidable. This is good news because it means that, with the support of artificial intelligence (AI)-powered tools and processes, health systems can predict and decrease denials and, therefore, increase revenue.
Denials occur for various reasons, from hospitals failing to provide documentation that supports the need for a medical procedure to an administrative error. For example, the insurance company (e.g., CMS or a private payer) could deny a hospital’s claim if the hospital submitted the claim with incorrect information, such as outdated patient insurance information or a misspelled name.
Although denials result in a lost $118 per claim, the impact is greater than the dollar amount. Denials also cost a health system the resources it uses to rework a claim. When a health system receives a denial from a payer, the organization must decide if it can fix the denial or if the claim is lost and unrecoverable. If the health system can fix the claim, it will rebill it and hopefully receive payment.
For example, if a health system thinks it can fix and resubmit a claim, the billing team will review the data and work to understand the reason for the denial. The team will then work with patient access teams, health information management, providers, and others to understand what went wrong in the claim submission process. Once the billing team has collected all of the information, it resubmits the claim, aiming for payment. However, payment is not a guarantee, and if the hospital receives payment for the second submission, its team members still performed twice the amount of work.
One way health systems can avoid this type of duplicate work and improve the healthcare revenue cycle is to improve the likelihood that payers will accept claims the first time. Organizations can decrease—and even prevent—denied claims by leveraging AI. Using advanced analytics algorithms, AI tools and processes can predict areas that are susceptible to denials by using predictive models to monitor the entire billing process and alerting stakeholders when specific processes have been interrupted. This allows members of the revenue cycle team to intervene and fix the problem before a denial occurs.
Health systems can use the following four steps to lay the groundwork for AI and then use AI predictive models to forecast denials. As a result, organizations can optimize healthcare revenue cycle performance and increase profits.
Health systems must first have the infrastructure to aggregate data from multiple sources to gain a comprehensive view of the healthcare revenue cycle. The organization’s revenue cycle team needs all the inputs representing the journey of a claim to see exactly when a procedure occurred and when the health system coded the procedure and billed the payer. With data from every step of the revenue cycle, team members have a complete picture of a claim’s lifecycle, allowing them to see all the activity within the system that should lead to cash collections and identify each potential problem area.
Once the data from the entire revenue cycle is in one place, health systems can use a data reporting tool, such as the Health Catalyst Revenue Cycle Advisor, to identify gaps in the current revenue cycle performance. This data will also be relevant inputs for the predictive model in step 4.
In this step, health systems can use the aggregated revenue cycle data to define a revenue cycle performance baseline so they can measure any change moving forward. An accurate baseline helps revenue cycle teams measure progress and change course if progress doesn’t occur at the anticipated rate.
After the health system defines the baseline, financial leaders can start using the denials data and other data elements (e.g., procedure information) related to that patient encounter to find specific behaviors that cause a variation in the revenue cycle. This data will also be relevant to include in the predictive model in addition to the data captured in step 2.
Because every healthcare revenue cycle follows a sequence of events, comprehensive data allows leaders to see what is actually happening in the revenue cycle versus what should be happening and, therefore, identify where the health system is losing money. For example, with each procedure, the health system schedules and registers the patient, performs the procedure, captures the charges, and bills for those charges. Any interruption or error in that sequence results in a denied claim (or lack of payment), directly impacting the revenue stream. Comprehensive data will reveal any variation in the claims process that could have led to the denial.
At step 4, health systems should be able to identify all the data inputs from steps 2 and 3 that ultimately influence the outcome of a revenue cycle denial and include them in the predictive model. With accurate data inputs, health systems can use the AI model to predict propensity to deny. This means that the model will flag potential areas where a health system is unlikely to get paid (e.g., between the insurance authorization and billing steps in the revenue cycle chain of events). With the predictive model insights, health systems know exactly which area of the revenue cycle to target to optimize workflows, leading to decreased denials.
Financial healthcare leaders should not overlook the power of AI to optimize the healthcare revenue cycle. Health systems can decrease some of the manual burdens of revenue cycle management and avoid the traditional reactive approach to denials by using AI to predict areas within the claims lifecycle that are likely to result in a denial.
With actionable information, health systems can increase the likelihood of receiving payment for every claim because revenue cycle leaders know exactly what to fix in the healthcare revenue cycle process. This insight allows organizations to receive full reimbursement for every service rendered so that they can focus on what matters most—providing quality care to the communities they serve.
Would you like to learn more about this topic? Here are some articles we suggest:
Would you like to use or share these concepts? Download this presentation highlighting the key main points.
