Baseball and Big Data: "Winning" Through Insights and Execution

In 2003, Michael Lewis published Moneyball, a treatise on the importance of data, analytics, insights and execution for the success of the Oakland Athletics, a small market major league baseball team.¹ The author noted that this team had a better baseball record over an extended period of time than teams that spent huge sums of money acquiring talent, including franchises in Los Angeles, New York, Texas and Baltimore.

As of mid-2014, the New York Yankees have a payroll of approximately $209.4 million and a record of 39-35, compared with the Oakland A’s payroll of $74.7 million and a record of 47-29. This disparity exists more than ten years after the widespread adoption of data-driven decision making. Why?

It involves generating insights from a vast repository of information and superior execution. It’s about doing things differently. The same statement can be applied to healthcare, where huge investments are being made in information technology (IT). Enterprise-wide Electronic Medical Records (EMR) costing $50-200 million are not uncommon and analytical tools are proliferating. Despite regional information networks being created, healthcare delivery remains ineffective and inefficient.

This article highlights the importance of targeted analytics to population health management, provider network management, health system and employer risk management, and patient medical management. A&M believes the critical success factors are still insight and execution, and not “excessive” front-end investment in a technology infrastructure.

High IT Investment Reflects High Expectations

The Gartner Technology Hype Cycle is a research tool that provides a graphical view of an application’s development and productivity over time. It describes the technology life cycle from introduction through inflated expectations and the trough of disillusionment (based on failure to deliver), and if successful, to the slope of enlightenment (next generation product enhancement) and the plateau of productivity (acceptance).²

Gartner Technology Hype Cycle

The passage of the Health Information Technology for Economic and Clinical Health (HITECH) Act provided $27 billion in financial incentives beginning 2011, for healthcare providers to implement electronic medical records. Incentives are paid based on the achievement of meaningful use, divided into three stages: 1) data capture and sharing, 2) clinical process enhancement and 3) outcomes improvement. Payment for Stage III implementation is targeted for 2016.

Healthcare providers have made substantial investments in electronic medical record systems, inclusive of clinical, administrative, financial and operational functionality. However, challenges remain regarding interoperability with legacy and other IT systems, implementation to hospital-employed and affiliated physicians, data extraction and physician acceptance. EMRs were touted as tools to enhance physician productivity, and yet their design appears more oriented towards administrative processes, ancillary staff, redundancies and billing. Future system enhancements, combined with remote connectivity via tablets and smart phones, would ideally lead to goal attainment (i.e., enhanced outcomes at lower cost).

The promise of longitudinal, patient-centric health information (HIE) exchanges across an entire ecosystem – hospitals, physicians, ancillary providers, reference labs and home health agencies – remains unfulfilled.

Population Health Management

According to Mathematica Policy Research, “population health management (PHM) programs are a set of interventions designed to maintain and improve people’s health across the full continuum of care – from low-risk, healthy individuals to high-risk individuals with one or more chronic conditions.”³ Large employers are most interested in PHM programs as opportunities exist to promote healthy lifestyles, reduce health risks and focus medical management efforts on the 5-10% of the employee population accounting for 49-63% of costs.⁴ Data is essential not only for population risk stratification, but also to identify the drivers of healthcare expenditures (medical, pharmacy; inpatient, outpatient), gaps in care and clinical outcomes.

Centers for Medicare & Medicaid Services (CMS) and employers have increasingly recognized that hospital systems are nearly always higher priced than free-standing physician practices for a wide variety of outpatient procedures, diagnostic tests and drug infusions. Hospital outpatient departments typically cost one and a half to three times more than ambulatory surgery centers.⁵ Diagnostic imaging premiums range from 40% to 400% depending on the type of test: X-Ray, CT, MRI or ultrasound.⁶ Specialty-drug infusions, those covered under the medical benefit, are often far higher in hospital outpatient facilities.⁷ Data-driven opportunities exist to identify high-cost site of care without necessarily affecting the physician-patient relationship.

Health systems are also using data to identify variation in physician utilization of health services (i.e., length of stay, specialist consultations, tests) based on risk-adjusted condition parameters, as well as to assess the cost and quality of post-acute care. The shift from a fee-for-service to shared risk business model driven by Accountable Care Organizations (ACO) is fundamentally altering the value proposition of specific providers. Data-driven opportunities exist to stratify physicians, nursing homes, home care agencies and others based on their ability to deliver lower cost care without materially affecting outcomes.

Risk Management

About one percent of Americans account for 20% of healthcare costs, and less than one-half of those affected have had a history of high costs in the prior year.⁴ Organ transplantation, major trauma, cancer and rare conditions often contribute to excessively high costs. Stop-loss insurance, combined with Centers of Excellence programs, are often used to reduce financial risk and optimize patient outcomes.

The Patient Protection & Accountable Care Act (PPACA) prohibits health plans from limiting medical coverage based on annual or lifetime spending caps. As a result, hospitals, skilled nursing facilities and drug manufacturers are no longer incentivized to manage exceedingly high-cost patients. The risk to self-insured employers of not having stop-loss insurance has significantly increased.

Historically, the frequency and severity of large claims exceeding $1 million in any given year has varied from year-to-year. Since the passage of the PPACA, the incidence of large claims and their aggregate value has increased 40-50% in 2009-2011.⁸ Claims of $3-5 million, once exceedingly rare, are now being seen more frequently by employer stop-loss carriers. Actuarial and predictive models are increasingly essential given the PPACA mandate and the random occurrence of exceedingly high cost claims.

Patient Medical Management

The value proposition associated with electronic medical records includes the possession of data (i.e., timely access to a patient’s medical history and physical, as well as diagnostic, laboratory and imaging results), and clinical decision support, inclusive of alerts and reminders.⁹ The ability to use queries and access notes (via natural language processing) is somewhat limited, at least at present.

EMR data is potentially superior to claims data for risk stratification, care-gap detection, pharmacy reconciliation, predictive analytics, discharge planning and more due to its timeliness, quantification (i.e., the level of Hemoglobin A1c, rather than whether an HgBA1c test was done), and potential for immediate action.

Leading healthcare systems such as Geisinger are also using EMR-based decision support systems to refine the roles of ancillary support staff, nurses and physicians to maximize the productivity of primary care physicians – always in short supply.

However, most healthcare systems have yet to fully integrate the use of EMR data across the continuum of care, from inpatient hospital to post-acute care and ambulatory services. The advent of Medicare Value-Based Purchasing penalties have focused attention on care processes to reduce 30-day admissions (and associated penalties) and not the need for chronic care management, inclusive of primary and secondary prevention activities.

Conclusion

The implementation of EMR systems and analytic tools alone does not result in more efficient and effective care delivery. Data-driven insights leading to changes in the process of care and staff behaviors are also required. The resulting benefits would include enhanced clinical decision-making, fewer gaps in care and preventable errors, reduced practice variation among providers, less unnecessary and redundant services, increased compliance with evidence-based guidelines, a greater focus on preventive activities and lower costs. Choosing the right IT systems and analytic platforms remains essential, given the related costs.

--------------------------------------------------------------------------------------

¹Lewis M. Moneyball. W.W. Norton & Company, 2003

² Wikipedia: Hype Cycle.

³ Felt-Lisk S. and Higgins T. Exploring the Promise of Population Health Management Programs to Improve Health. Mathematica Policy Research, Aug. 2011.

⁴ U.S. Department of Health and Human Services, Agency for Healthcare Research and Quality, Medical Expenditure Panel Survey (MEPS), 2008

⁵ Stegman M. Hospital Outpatient Surgery Versus Freestanding ASC Costs and Reimbursement. Journal of Health Care Compliance; Mar.-Apr. 2005.

⁶ Area Hospital Imaging Costs vs. Intermountain Medical Imaging.

⁷ Einodshofer M, Duren L. Cost Management through Care Management, Part 2: The Importance of Managing Specialty Drug Utilization in the Medical Benefit; Sep. / Oct. 2012 Vol 5 (6)

⁸ D.W.Van Dyke & Co., Inc. Survey of large employer stop loss carriers, 2013 (n=37)

⁹ Kern L, Wilcox A, Shapiro J, Dhopeshwarkar RV, Kaushal R. Which Components of Health Information Technology Will Drive Financial Value? Am J Manag Care. 2012;18(8):438-445

Key Contacts:

Guy Sansone
Managing Director
+1 212 328 8513

Steven Bussey
Managing Director
+1 212 763 9662

David Gruber, MD
Managing Director
Director of Research
+1 212 763 9801

For more information:
Healthcare