Healthcare spending in the United States continues to outstrip spending in all other member countries of the Organisation for Economic Co-operation and Development by a wide margin, accounting for 17.9% of gross domestic product in 2016, according to HealthAffairs.org.
In a bid to capitalize on new growth opportunities that could reduce costs and improve care delivery outcomes, the healthcare industry has shifted its focus away from the traditional, fee-for-service to a more sustainable, value-driven model fueled by real-world advancements and stronger consumer demand.
As a result, provider organizations are leaning heavily on unique data practices to help curtail rising medical costs, manage variation in care delivery, and achieve higher-level outcomes in a risk-based environment. For a provider organization and leaders, this work is incredibly resource-intensive because it requires part analytics, part culture change.
An analytical approach
Identifying patient and population trends has become vital in achieving long-term, value-based reimbursement success, leading to broader opportunities to ensure that care and outcomes are appropriately managed – whether that is decisions made over prescription medications, medical procedures or staffing, to name a few. All benefit from standardized measures and treatment based on the best evidence available on clinical effectiveness.
Having standardized levels of care that can be supported by data empowers providers to determine where unnecessary expenditures can be driven out of the system. However, such value-oriented reimbursement models present a fundamental challenge in the ways care gets delivered, because health systems and providers are now required to manage all points of patient care across the continuum—at levels of quality and cost that will guarantee reimbursement.
For providers to guarantee their claims get paid under value-based models, a strong data strategy is essential to reduce inefficiencies and standardize quality healthcare delivery. Intermountain Healthcare and Kaiser Permanente are two organizations embracing data to aggressively adopt value-based care. The caveat? Aligning with such an approach requires not only expertise but also an organizational structure that is open to changes in care delivery based on evidence-driven best practices and outcomes.
Case in point: Providence St. Joseph Health
During a recent webinar, Cutting Edge AI – Should You Take a Chance? speakers Ari Robiseck, M.D., Chief Medical Analytics Officer at Providence St. Joseph Health and Jean-Claude Saghbini, Chief Technology Officer at Wolters Kluwer Health, presented the role Artificial Intelligence (AI) will play in healthcare and what organizations need to do to build a robust AI strategy.
Dr. Robicsek discussed how the esteemed hospital has been able to harness underlying patient data to identify patterns in care delivery to drive better outcomes at the lowest costs possible. Providence St. Joseph is focused on serving the community including those unable to pay, making it increasingly important for the organization to identify how to efficiently achieve the best clinical outcomes.
Responsible for coordinating clinical analytics that aligned with the Catholic 50-hospital healthcare system, Dr. Robicsek leveraged the power of data analytics to better serve broader organizational queries while also providing a view into the best methods to optimize the use of resources within its facilities.
To determine how to best measure the value of care provided, Dr. Robicsek and his team leveraged a value-oriented architecture, which had the capabilities to support advanced analytical techniques. They used as a barometer an analysis of the average costs to the system for total knee replacement, finding that costs to the organization ranged from $5,600 to $10,000 per case—among the higher-volume surgeons.
By modeling the impact of costs per case with regard to higher-cost physicians, the team discovered it could reduce total costs—but not nearly enough to make the margin paid out by Medicare. Through a deeper analysis, the team also discovered an opportunity to reduce costs per case across the board, breaking down the costs into the following.
- Costs per case of implants, the biggest driver of costs
- Costs per case of time
- Costs per case of room and board
- Costs per case of pharmaceuticals, such as pain medication and hematological agents.
However, analyzing costs only got them half way there because clinical outcomes are highly valued within the organization and a motivator for behavior change in clinicians. Drawing on complex data sets and insights from clinical and administrative systems, Dr. Robicsek’s team used advanced analytical techniques and created a value-oriented schematic that enabled physicians to identify where they varied in outcomes and costs per case compared to peers. This benchmarking data combined with modeled scenarios that identified potential areas in which physicians could reduce their costs per case, helped physicians in the system migrate toward those benchmarks.
Best data practices
Value-based care holds great promise, but getting it right requires the industry to adopt a unique data approach. The following improvements can help providers and other healthcare practitioners leverage information technology to achieve two key aims: restructuring care delivery and measuring results.
Build a case
Determine whether and how your specific institution could support a data strategy to champion for care delivery improvements.
Prioritize which needles to move
Establish common data and precise language definitions to improve effective outcomes achievement.
Data as a driver
Aggregate structured data to encompass the full care cycle, specific to each medical condition or treatment management option.
Build in compliance
All healthcare stakeholders—from providers to payers, to vendors—share responsibility for the authenticity and integrity of data exchange.
Adopt interoperability standards enable data mapping, normalization, and communication between different providers and payer organizations as part of a larger restructuring effort.
Clinical decision support is that much more important in the value-based model because it eliminates waste and identifies areas where practices can be standardized relative to the best care and outcomes, and data has proved to be an invaluable asset in the identification of patterns.