Consumerism, disruption, innovation, retail health, transformation, big data: A new set of words is dominating the health-care lexicon. The vocabulary reflects the fact that smartphone-connected consumers have catapulted health-care delivery into the Internet-based, on-demand economy.
Success in this new economy driven, by a value-based business model, requires legacy health-care providers to target four goals: improved care experience, better population health, enhanced clinician satisfaction, and much lower costs.
Additionally, the competition has changed. Increasingly, innovative technology and retail companies with access to big data and powerful analytics are using digital connections to appeal and provide health-care services to consumers.
For example, Google’s Verily division is developing wearables and analytics for timely decision-making and care interventions. Walgreens’ partnership with telehealth provider MDLive offers consumers virtual services in 43 states, soon to be nationwide. IBM Watson Health is applying big data and cognitive computing to create personalized health and health-care protocols.
To compete in a price- and quality-sensitive healthcare environment, proactive health-care organizations are focusing big — and smarter — data and analytics efforts on three core business initiatives intended to transform, rather than simply manage, delivery and costs.
Health care faces a logistics challenge not unlike that of UPS in the sequencing of daily stops by its 110,000 vehicles along 55,000 routes, or the scheduling by American Airlines of its nearly 100,000 employees and 6,700 daily flights. Both companies use sophisticated workforce optimization and logistics modeling to schedule and deploy human and other resources.
U.S. hospital care, with a workforce of 1.4 million full-time registered nurses in the nation’s 3,200 hospitals and health systems, makes these numbers look small. Nurse staffing typically represents 25% of a hospital’s total operating costs, which in 2014 accounted for about 34% of the nation’s nearly $3 trillion health and supplies expenditures.
Given these numbers, human resources and scheduling practices must be highly cutting-edge and data-driven. Getting the right clinical staff to the right location at the right time is no small task. Moving health-care services to the most appropriate, lowest-cost setting requires assignment of staff within and across hospitals and to clinics, physician offices, nursing homes, and other settings.
A regional health system in the Midwest is using advanced optimization modeling to improve its nurse staffing and scheduling in 110 nursing units at 10 hospitals. Quantitative components include multiple objectives (e.g., staff preferences, maximizing coverage while minimizing costs), millions of variables (e.g., skill mix, demand fluctuation), plus lots of constraints (e.g., time-off requirements, staff availability).
According to an April 1 presentation at the 2017 American Organization of Nurse Executives, in the first year of the optimization initiative (2015-2016) the system: improved staff engagement by 38%; maintained an 87th percentile quality rating; effectively matched RN clinical resources to patient demand; implemented a highly flexible float pool across its hospitals to meet future staffing needs; and saved $16 million, representing 5% of its nursing labor spend.
Comprehensive benchmarking databases and advanced data analytics underpin the organizational ability to provide the high-value care required by health-care consumers and purchasers.
As in other industries, unwarranted or inappropriate variation in services is a significant source of suboptimal outcomes and unnecessarily high costs. Clinical variation reflects a gap between the desired “best clinical practice” and current performance. For example, not implementing a protocol for mobilization of patients’ knees following knee replacement is a suboptimal clinical practice that represents unwarranted variation and, in most cases, added cost.
Data-rich analytics can indicate when quality outcomes and/or costs differ among providers. For example, a 14-hospital system is using advanced analytics that integrate data from public and commercial sources to identify physicians who consistently outperform benchmarks. The focus is on key performance metrics, including readmissions and mortality rates for high-risk diagnoses such as heart failure.
Staffers compare metrics for top-performing physicians to those of the lowest performers to identify and address variances. For example, the best performer had a 0% mortality rate for heart failure patients, compared to 5.5% among the lowest performers. This analysis not only identifies quality issues, but also enables quantification and mitigation of cost implications.
Many clinicians who receive reliable and objective data indicating unwarranted variation in their own care — whether related to quality, outcomes, and/or cost — will immediately bring their practices in line with their colleagues. Thus, data analytics promises to raise the health system’s overall performance and lower costs.
Smarter data and analytics are providing organizations with a more comprehensive understanding of the “true costs” of delivering care. This is essential for successful participation in value-based payment programs.
Robust cost accounting analytics provide visibility for costs at a granular level. Costs incurred for chargeable and non-chargeable activities and those for items that may not directly generate revenue are both part of the total cost of patient care.
For example, the costs of hip implants may represent significant opportunities for cost reduction while improving quality. Cost accounting data supports answers to questions such as: What implants are our surgeons using (by surgeon), and what does each cost? What types of procedures are we doing with those implants? What is the mix of patients receiving those procedures? Analytics unearths variations in costs that are driven by vendor selection, physician preferences, and implant procedures performed on high-risk patients.
As payment structures move to bundle payments (i.e., one payment covers all aspects of care related to a diagnosis, such as knee replacement), hospitals must be able to quantify the total cost of an episode of care, not just a procedure. Use of best-practice cost accounting is key to understanding and managing expenditures before admission (e.g., “prehab”), during hospitalization (e.g., physician and nursing care, imaging, pharmacy), and after discharge (e.g., a nursing home).
As such, a hospital’s cost accounting system must be flexible enough to integrate various types of data from electronic health records, hospital departmental systems, and a wide range of providers. Additionally, the analysis of that data to encompass all preventive, acute, and post-hospital costs is vital.
Complex industries, such as transportation and retail trade, have long used rich databases and mathematical modeling to drive improved workforce optimization, business performance, and financial performance. Health care’s size, complexity, cost, and societal importance make it ripe for innovative solutions using big, smart data and analytics. The three core initiatives described here represent a good start for finance leaders in U.S. hospitals and health systems.
Jason H. Sussman ([email protected]) is a managing director of Kaufman, Hall & Associates, a provider of management consulting, enterprise performance management software, and benchmark data and analytics for hospital boards and management teams. He directs capital planning and allocation advisory services in the firm’s strategic and financial planning practice.