Last week, CMS issued a Request for Information to “obtain input on how the agency can better use emerging technologies, such as Artificial Intelligence, to ensure proper claims payment, reduce provider burden, and, overall, conduct program integrity activities in a more efficient manner”.  The RFI is consistent with movement toward the agency’s belief that healthcare data accessibility and utilization is crucial to revolutionizing the industry and slashing costs.

Like CMS, Advis believes that the transformation of data into value can take many forms. In day-to-day healthcare operations those forms include but are not limited to:

  • Improved Revenue;
  • Monitoring and Assurance of Compliance;
  • Reduced Costs;
  • Business Agility;
  • Increased Efficiency;
  • Enhanced Experience; and
  • Confident Decision-making.

Advis understands the barriers to implementing data science and analytics. Those barriers can range from budget shortfalls to overwhelming amounts of data. However, given the industry-wide push to mandate actionable and directed data driven insight across multiple and sometimes interrelated or overlapping areas of healthcare operations and compliance, Advis feels it’s a good time to audit your internal analytics practices now.

To help with this, according to key compliance and operational areas, Advis has created a list of Ten Data Science and Analytics trends that need watching. This list contains widely used applications of Artificial Intelligence, Data Science and Analytics across healthcare operations and compliance; they explore the use cases and benefits of robust internal data science and analytics initiatives.

Advis and our team of consultants and data scientists are uniquely positioned to combine analytics and data science best practices with unsurpassed industry knowledge and broad experience with regulatory matters to produce creative and impactful business solutions that optimize revenue while ensuring compliance.

Ten Data Science and Analytics Trends by Key Compliance and Operational Area

  1. Revenue Cycle Compliance Optimization

One of the most popular applications of data science, Revenue Cycle Compliance Optimization predictive modeling determines the key drivers of insurance claim denials. Real time interactive tools are developed to flag high risk claims. Additional models can monitor charge capture and entry timing, duplicate charges, and Medicare and Medicaid Claims Requirements, including place of service coding, modifier utilization, 855A exact matching, and 3 Day Window Payment testing.

  1. Strategic Planning and Finance

Data science gives leaders the confidence needed to make important decisions. Generating targeted insights through feasibility studies, market assessments, site selection models, service mix analyses, budget and revenue projections, and revenue optimization furthers decision making. By utilizing various advanced analytical data science techniques, like clustering analysis, anomaly detection, regression analysis, classification analysis, and natural language processing, confidence in having made the right decision is solidified.

  1. Needs and Prevention

One way that organizations can address social determinants of health issues is through the use of data science and analytics. Start by leveraging community health needs assessments as part of your 501(r) compliance regimen; various machine learning techniques identify specific patient populations and their needs; patterns form which ultimately inform strategic decision making that addresses unmet, even unrecognized, community needs. Additionally, data science and analytics can help develop hospital acquired condition reduction programs, enhance eligibility for bundled payment program pilots and capitated contractual requirements, as well as monitor managed Medicare/Medication monitoring and contractual requirements.

  1. Length of Stay Modeling

Artificial Intelligence has revolutionized the way patient length of stay is predicted. By analyzing patterns in timely ADT, bed utilization, One Day Stays, DRGs, Resource Allocation, Uncompensated Care Loss, and pattern recognition by time, patient demographics and payer class, predicting length of stay becomes a science.

  1. Clinical Appropriateness

Fraud detection and monitoring is another rapidly expanding area of data science and artificial intelligence utilization. Algorithms are built to monitor documentation completeness and requirements, use of diagnoses codes, InterQual Criteria, appropriateness of care venue, optimization of equipment and scheduling, as well as pattern recognition and outlier detection from physician evaluation and management coding.

  1. Physician Referral Patterns

Optimization and insight into physician referral patterns are made through use of data science solutions that monitor appropriateness of the referral, patient outmigration, utilization of specialty services, extended stay patterns, and optimization of advanced practice provider utilization.

  1. Medical Staff Credentialing

Advis has seen a rise in custom tools built to monitor timing, appropriateness, type of privileging, application requirements, excluded provider screening, and allied health professionals scope of practice.

  1. HR Analytics

Data Science and analytics can help organizations meet ACA reporting requirements by automating KPIs, such as employee work hours, as they relate to health benefits, certification renewals, upkeep and proof of inoculations, and annual reviews and performance evaluation tracking. Additionally, turnover and churn modeling can be used to determine staffing needs and to predict resource shortages.

  1. Readmission Management

Data science practitioners can leverage predicative analytics to identify patients at-risk for 30-day readmissions, same day readmissions, and value-based contract readmissions.  One of the first operational areas to implement artificial intelligence, the science allows an organization to optimize Medicare payment incentives and avoid penalties.

  1. MACRA / MIPS

According to the Weill Cornell Medical College and Medical Group Management Association, the average physician spends 15.1 hours per week processing quality metrics. Data science and artificial intelligence can build custom tools to manage the pressure coming from regulatory agencies, meaningful use criteria, ever increasing costs, risk contracts, and the need to improve safety and quality.

For any questions regarding these trends, or for any other healthcare or operational matters, please contact Advis at (708) 478-7030.