21
Moneyball and Analytics Modern Healthcare's Guide to Predictive Analytics.

[E book] Moneyball and Analytics - Modern Healthcare's Guide to Predictive Analytics

Embed Size (px)

Citation preview

Page 1: [E book] Moneyball and Analytics  - Modern Healthcare's Guide to Predictive Analytics

Moneyball andAnalytics

Modern Healthcare's Guide to Predictive Analytics.

Page 2: [E book] Moneyball and Analytics  - Modern Healthcare's Guide to Predictive Analytics

PART ISUMMARYYou can’t miss the hype surrounding big data and analytics. Whether baseball or healthcare, we’re fed a steady stream of how analytics is changing our digital economy.

Data is described as the new oil, soil, the next big thing and the force behind management and technology revolutions.

If information is the oil of the 21st century, no industry enjoys this valuable resource more than healthcare.

For too long the U.S. healthcare system fostered an interesting dichotomy: constant innovation, yet no real change. Major economic, social, demographic, technological, and regulatory changes are transforming this.

Innovative health care models are shifting the focus to outcomes and cost savings rather than just pure volumes.

Patients expect and deserve excellent care from their health care providers. Hospitals are on the front lines of advancing this mission.

Predictive analytics is quickly becoming healthcare’s competitive differentiator. Organizations are turning to big data to transform health information into health intelligence. By leveraging the power of prediction, healthcare providers are improving patient care, reducing costs, and serving their communities in new ways.

Healthcare Transformation and Innovation

Value-based care, consumerism, and personalized care are becoming the norm. To keep pace, organizational dependence on data and analytics is steadily building.

PART II

How To Build A Predictive Analytics Strategy

There’s no one-size-fits all approach. Success begins with a solid business case, leading to carefully planned pilots and operationalization across the enterprise.

PART III

Proof Positive: Predictive Analytics Use Cases

5 analytics use cases that capture value and seize opportunities.

Page 3: [E book] Moneyball and Analytics  - Modern Healthcare's Guide to Predictive Analytics

Healthcare 2.001 Innovation in healthcare

02 Trends driving transformation

03 Moneyball: for healthcare

04 Seeing 3D: Data-driven decisions

graymatteranalytics.com

“Success in investing is not a function of what you buy. It’s a function of what you pay.”

Howard Marks

WE CAN NOW SAY THE SAME FOR HEALTHCARE

Healthcare Transformation.

PART I

Page 4: [E book] Moneyball and Analytics  - Modern Healthcare's Guide to Predictive Analytics

01

New models of healthcare delivery are emerging. Providers, clinicians, payers and patients all share in the transition to value-based care.

The healthcare industry is clearly shifting to payment for value (VBC), as fee-for-service reimbursement models are being phased out.

Ambitious focus areas include: • Reducing the cost of care • Enhancing the patient experience, and • Improving overall health outcomes

graymatteranalytics.com

“America’s health care system is neither healthy, caring, nor a system”

Walter Cronkite - 1993

Change won’t be easy, but it will be quick.

[1] The Digital Universe Driving Data Growth in Healthcare. 2014 Vertical Industry Brief. EMC2

Healthcare Transformation.

“ T r i p l eA i m ”

As the industry transforms to a value-based system, redefining value and success in healthcare starts with creating more sustainable healthcare systems.

Requiring institutions to quickly innovate by:

• Cultivating transformational initiatives, capabilities, new care models and consumer-centric services

• Managing regulation, risk and governance, despite competing mandates

• Using data to enhance outcomes and deliver business value

• Securing enterprise information

Page 5: [E book] Moneyball and Analytics  - Modern Healthcare's Guide to Predictive Analytics

02Trends Driving Transformation.Major economic, social, demographic, technological, and regulatory changes are reshaping the healthcare industry.

graymatteranalytics.com

Image: Building A Big Data Strategy In Five Easy Steps. Platfora

Preventive Care

• Digital health and IoT bringing together allstakeholders (consumer, physician, payer, etc.) to put the major users of the healthcare system in control of managing disease and health management

Value-Based Care Delivery

• Reduce cost, increase quality/value• Transparency and accountability• Reimbursement rates tied to performance

Consumerism

• Patients playing a greater role in care as their knowledge grows

• Emphasis on patient engagement, member retention, competitive differentiation and customer service

• “Retailing” of healthcare includes minute clinics and other alternative options

Personalized Medicine

• Care delivery is increasingly tailored at the individual level. One size fits none.

• The Internet of Things (IoT) bridges the digital and physical worlds to modify physician and patient interaction

• Digital health leveraging technology to enable remote monitoring, telehealth and individualized behavior modification pathways

Page 6: [E book] Moneyball and Analytics  - Modern Healthcare's Guide to Predictive Analytics

03MoneyballMindset.To keep pace with transformative trends, organizational dependence on data and analytics is steadily building.

Widespread adoption of electronic health records is driving hospitals and health systems to rethink their approaches to data management and technological competencies.

Pressured by business, clinical and regulatory demands, innovative organizations are expanding their analytical capabilities by turning data into actionable insights.

graymatteranalytics.com

Paul DePodesta, who helped transform the Oakland Athletics recently joined the Scripps Translational Science Institute as an assistant professor. In healthcare, like baseball, his analytical capabilities will help manage barriers to growth, including; deeply held traditions, monolithic organizational and operational structures and psychological resistance to change. 2

Trending.

“Analytics is quickly becoming viewed as a competitive differentiator and can add value to a variety of other organizational goals, including consumer experience, growth initiatives, and cost reduction.”3

Image: Moneyball. Columbia Pictures. 2011[1] Baseball statistician Billy James, as quoted in Michael Lewis’ book Moneyball: The Art of Winning an Unfair Game.[2] Moneyball Mindset. Hospitals and Health Networks. Karyn Hede. April 2016.[3] Health system analytics – The missing key to unlock value-based care. Deloitte Center for Health Solutions. Based on 2015 US Hospital and Health System Analytics Survey.

Page 7: [E book] Moneyball and Analytics  - Modern Healthcare's Guide to Predictive Analytics

04Data-Driven Decision Making Fueling Predictive Analytics.

New models of healthcare delivery require better analytics tools.

The total amount of medical information in the world doubles every five years. Making matters worse. The half-life of facts grows ever shorter. Imagine this: Medical knowledge about cirrhosis or hepatitis takes about forty-five years for half of it to be disproven or become out-of-date.2

Data is transforming industries and society at unprecedented rates. To keep pace while meeting value-based and business goals, new data models, algorithms and machine-learning techniques can help find patterns and insights where no one thought to look.

With the infusion of technology and business intelligence, organizations can take advantage of cloud, analytics, mobile and social technologies to seize opportunity while navigating challenges.

graymatteranalytics.com

Of C-level executives,

74%would like to achieve a deeper

understanding of the technologies underlying big data tools 3

Triple Aim, Value-Based, and Beyond

Figure 1: The value of analytics in healthcare. Interviews with healthcare executives. IBM Institute for Business Value analytics.[2] The Half-life of Facts: Why Everything We Know Has an Expiration Data. Samuel Arbesman. Penguin Group.[3] Economist Intelligence Unit.

FIG.1

Page 8: [E book] Moneyball and Analytics  - Modern Healthcare's Guide to Predictive Analytics

Building a Predictive Analytics Strategy.

Healthcare’s digital transformation is in many ways a giant IT project. But on a deeper level, as data increasingly flows from silos to unified platforms, we’re seeing digital transformations are an organization wide process. Making communication and collaboration arguably the most important drivers of success.

While no two approaches are identical, here are the 5 most common building blocks to predicative analytics success.

graymatteranalytics.com

“Predictive analytics is not reinventing the wheel. It’s applying what doctors have been doing on a larger scale. What’s changed is our ability to better measure, aggregate, and make sense of previously hard-to-obtain or non-existent behavioral, psychosocial, and biometric data.

Combining these new datasets with the existing sciences of epidemiology and clinical medicine allows us to accelerate progress in understanding the relationships between external factors and human biology—ultimately resulting in enhanced reengineering of clinical pathways and truly personalized care.”

Vinnie RameshChief Technology OfficerCo-founder of Wellframe

05 Build a business case

06 Examine fundamental capabilities

07 Operationalize strategy

08 Use pilot program to demonstrate analytics value

09 Put your data to work

PART II

Page 9: [E book] Moneyball and Analytics  - Modern Healthcare's Guide to Predictive Analytics

05Building a Predictive Analytics Strategy.

Step 1BUILD YOURBUSINESS CASE

Given our digital, analytics driven world, our natural tendency is to build a strategic priority around big data and the IoT, but the solution isn’t to just build a digital strategy just because everyone else is.

Transformation into a digitally-driven organization requires total organizational commitment. Moving past the hype takes a measure of resolve that few organizations demonstrate. A 2015 survey by MIT Sloan Management Review and the SAS Institute revealed the inconvenient truth about the unglamorous but necessary actions required to improve decision making with analytics.

As with any good strategy, the process of embedding predictive analytics starts with specific objectives tied to helping the organization achieve a sustainable competitive advantage.

Aligning an IT model with specific business goals enables big data to find new opportunities and help increase productivity, efficiency, and business processes.

graymatteranalytics.com

Image: Bloomberg. Photographer: Christopher Payne[1] Beyond the Hype: The Hard Work Behind Analytics Success. MIT Sloan Research Report. Spring 2016.

COMPETITIVE ADVANTAGE FROM ANALYTICS IS DECLINING The percentage of organizations gaining competitive advantage from analytics declined significantly in 2015.1

“Analytics used to be a competitive advantage, but now it’s becoming table stakes,”says Steve Allan, head of analytics for Silicon Valley Bank.

Page 10: [E book] Moneyball and Analytics  - Modern Healthcare's Guide to Predictive Analytics

06After clear business objectives have been established, current analytics capabilities can be taken to a desired future state through a series of detailed assessments.

To bridge strategy, design and implementation, a modular program begins with the below assessments. Followed by transformative analysis of people, process, and technology. Successful programs include continuous monitoring.

graymatteranalytics.com

Building a Predictive Analytics Strategy.

Step 2EXAMINE FUNDAMENTAL CAPABILITIES

CurrentAnalyticsEnvironmentIs HereYOU WANT TO

BE HERE

Anal

ytic

s M

atur

ity

Maturity Assessment:Through an analytics maturity assessment, a current maturity state is compared to desired maturity levels to help determine how quickly and cost effectively analytics can help seize opportunities.

Data Analytics Capability Assessment:A capability assessment recommends organizational structure, data governance models and resource needs required to establish analytics as a competitive differentiator.

Solution Design:For a successful analytics implementation, a prioritized roadmap includes a comprehensive solution design; including approaches to governance, adoption, change management and a technology framework.

ANALYTICS

BUSINESSINTELLEGENCE

DATAMANAGEMENT

SAMPLE ANALYTICS MATURITY / ADOPTION MODEL

Page 11: [E book] Moneyball and Analytics  - Modern Healthcare's Guide to Predictive Analytics

07

graymatteranalytics.com

Building a Predictive Analytics Strategy.

Step 3OPERATIONALIZEANALYTICS STRATEGY

Once a business case for operational, clinical and financial analytics is set, next steps include establishing a formal strategy with the following essential components.

Identify and DescribeA data and analytics roadmap adds the tools, methods and necessary skills to collect, manage, publish business intelligence.

Provision and ShareTo provision and share data, determine the best packaging (files, transactions, data streams, etc.), content, format, and methods to enable data availability and delivery.

Stage and StoreBest practices for enterprise data storage and sharing should include master/reference data, business event details, and processing demands (apps, reports, dashboards).

Govern and ManageLastly, stakeholders establish data governance best practices; including methods, processes, access policies and conflict resolution steps.

Integrate and MoveData infrastructure best practices ensure bulk data movement, processing and applications are available during source data movement/transformation.

Page 12: [E book] Moneyball and Analytics  - Modern Healthcare's Guide to Predictive Analytics

08

graymatteranalytics.com

Despite being awash in information, many organizations are still starved of analytics meaning. The modern decision maker might be more data-driven, but organizationally, culture feels the same.

To speed digital transformation and analytics adoption, pilot programs or project specific proof-of-concept demonstrations can help teams truly appreciate and understand the value of analytics. Thereby, managing natural frictions that come in any change management process.

With buy-in from all levels, organizations unlock the full potential of data-driven intelligence. Creating a “data culture”, allows insights to come from anyone, anywhere, at any time.

“Culture trumps data, I don’t care how good your model is. If you don’t understand the culture…you’re

not going to succeed with analytics and deliver success for the business”1

Jim SpriggDirector of Database Marketing and Analytics

InterContinental Hotels Group

Building a Predictive Analytics Strategy.

Step 4USE A PILOT PROGRAM TO PROVE VALUE OF ANALYTICS The primacy of people in the digital age.

Interesting research out of MIT shows the real challenge is recognizing that using big data and analytics to better solve problems and/or make decisions obscures the organizational reality that new analytics often requires new behaviors from the workforce.2

Accenture, the consultancy, shows the success of any company going through fundamental digital transformation is understanding that it’s first and foremost a people issue.3

Both examples highlight an unrelenting fact: Before a business can truly enjoy the benefits of data-driven decision making, an organization and its people must first be aligned to a common purpose. Common purpose starts and ends with culture.

[1] Beyond the Hype: The Hard Work Behind Analytics Success. MIT Sloan Research Report. Spring 2016.[2] Why Your Analytics are Failing You. Harvard Business Review. Michael Schrage.[3] People First: The Primacy of People in a Digital Age. Accenture Technology Vision 2016

Page 13: [E book] Moneyball and Analytics  - Modern Healthcare's Guide to Predictive Analytics

09

graymatteranalytics.com

Building a Predictive Analytics Strategy.

Step 5PUT YOUR DATATO WORK

The race to equip hospitals and health systems with analytics is a crowded space. Selecting the best strategy to universal, enterprise level predicative analytics is no easy task.

Unfortunately, choosing the best analytics platform comes down to more than a cost/benefit analysis of build-it, buy-it, or lease-it.

To put your data to work, the best vendor/partner will offer an integrated portfolio of healthcare related software, services, and solutions. Through an integrated portfolio, business analytics will offer the breadth and depth to fuel better business and health outcomes.

57of healthcare data is useful if tagged and analyzed (diagnoses, research, analysis)1

but only 3.1 of all healthcare data provides the highest value

THERE’S VALUE HIDING IN YOUR DATA, ARE YOU LOOKING IN THE RIGHT SPOTS?

[1] The Digital Universe Driving Data Growth in Healthcare. 2014 Vertical Industry Brief. EMC2

%

%

Page 14: [E book] Moneyball and Analytics  - Modern Healthcare's Guide to Predictive Analytics

Use Cases.

5 innovative ways providers are using predictive analytics to improve routine health care.

graymatteranalytics.com

10 Predict preventable heart failure

11 Improved quality of care at the U.S.’s largest health system

12 Less waste through better clinical decisions

13 Better workflows, not just more technology

14 Context drives tech adoption and accurate diagnosis

PART III

Page 15: [E book] Moneyball and Analytics  - Modern Healthcare's Guide to Predictive Analytics

10Improving Routine Care Through Predictive Analytics.

USE CASE:Reduce preventable heart failure hospital readmission rates

According to the American Heart Association, about 5.7 million adults in the United States suffer from heart failure, with the number expected to rise to 8 million by 2030. Statistics show that each year about 870,000 new cases are diagnosed and about 50 percent of those diagnosed will die within five years.

However, many heart failure patients can lead a full, enjoyable life when their condition is managed with proper medications or devices and with healthy lifestyle changes.To expand on these preventative steps, innovative healthcare providers are increasingly using sophisticated analytics to sort through terabytes of clinical data to uncover opportunities to improve quality and boost efficiency.

Clinical Intervention for Specific At-Risk Populations

Parkland Health and Hospital System in Dallas, Texas, has developed a validated EHR-based algorithm to predict readmission risk in patients with heart failure. Patients deemed at high risk for readmission receive evidence-based interventions, including education by a multidisciplinary team, follow-up telephone support within two days of discharge to ensure medication adherence, an outpatient follow-up appointment within seven days, and a non-urgent primary-care appointment. In a prospective study, the algorithm-based intervention reduced readmissions by 26%.1,2

graymatteranalytics.com

“It takes all these pieces of data from the EHR, and it has an algorithm, and tells us which patient is at higher risk for heart failure” 3

Dr. Susann LandChief Medical Officer

Texas Health Harris Methodist Hospital Hurst-Euless-Bedford

[1, 3] Making Predictive Analytics a Routine Part of Patient Care. Harvard Business Review. Ravi B. Parikh[2] Allocating scarce resources in real-time to reduce heart failure readmissions: a prospective, controlled study. View study here.

Page 16: [E book] Moneyball and Analytics  - Modern Healthcare's Guide to Predictive Analytics

11Improving Routine Care Through Predictive Analytics.

USE CASE:Integrating EHR with a data warehouse to improve quality of care

The Veteran’s Health Administration (VHA), the largest health system in the United States, has collected electronic data from its patients for over three decades. Beginning in 2006, the VHA built a corporate data warehouse as a repository for patient-level data across its national sites.

The sheer amount of inpatient and outpatient data has allowed the VHA to create comprehensive algorithms that reliably predict meaningful outcomes such as risk of death and hospitalization.

Better clinical workflows, better outcomes

VHA's evolution toward "big data," defined as the rapid evolution of applying advanced tools and approaches to large, complex, and rapidly changing data sets helps illustrate how advanced analysis is already supporting the VHA's activities. Ranging from routine clinical care of individual patients--for example, monitoring medication administration and predicting risk of adverse outcomes--to evaluating a system wide initiative to bring the principles of the patient-centered medical home to all veterans. 1

The VHA’s investment in an integrated EHR and data repository — 5% of its total health spending —is substantial. However, the ability to reliably predict outcomes to improve quality of care may explain why the VHA’s net return on EHR investment is over $3 billion.

graymatteranalytics.com

[1] Making Predictive Analytics a Routine Part of Patient Care. Harvard Business Review. Ravi B. Parikh

Page 17: [E book] Moneyball and Analytics  - Modern Healthcare's Guide to Predictive Analytics

12Improving Routine Care Through Predictive Analytics.

USE CASE:Improve clinical decisions; eliminating unnecessary procedures, saving money and preventing overtreatment

Uncertainty over a clinical decision often leads physicians to overtreat or undertreat patients. Predictive analytics can allow clinicians to steer high-cost interventions to those high-risk patients who actually need them.

Specifically targeted data can help reduce unnecessary interventions and waste. When uncertainty exists, clinicians will treat more patients with preventative measures. However, if data narrows down the truly at-risk patients, fewer will be exposed to unnecessary procedures.

Focus on low-value decision points

Consider the use of antibiotics to treat newborns. While less than 0.05% of all newborns have infection confirmed by blood culture, 11% of them receive antibiotics. Kaiser Permanente of Northern California has used predictive analytics to reduce this overuse.

Its researchers have developed an algorithm to accurately predict the risk of severe neonatal infection based on a mother’s clinical data and the baby’s condition immediately after birth. Using this algorithm OB/GYNs can better determine which babies need antibiotics, sparing up to 250,000 American newborns each year from receiving unnecessary antibiotics.

This could reduce medication costs and side-effects among vulnerable newborns. 1

graymatteranalytics.com

[1] Making Predictive Analytics a Routine Part of Patient Care. Harvard Business Review. Ravi B. Parikh

Page 18: [E book] Moneyball and Analytics  - Modern Healthcare's Guide to Predictive Analytics

13Improving Routine Care Through Predictive Analytics.

USE CASE:Integrate data-driven insight, not technology, into clinical workflows

Don't overload clinicians with unimportant data; they might start to ignore it. Data that doesn't fit into the clinical workflows doesn't need to be in front of them. Decision-support tools can help practicing physicians spot the right data for their patients.

To increase adoption, make workflow integration easy

Physicians see hundreds of numbers (vital signs, laboratory values, etc.) each day. So there’s a danger that an algorithm’s output may just be another number that physicians ignore if it does not fit well into a daily workflow.

As an example. Decision-support tools with predictive outputs can be iterative enough to be useful for a practicing physician.

A critically ill patient’s risk of deteriorating varies constantly in the hospital. Thus, static risk percentages would be of little use to physicians, who need real-time information to make real-time decisions. 1

With high-value data from the EHR and custom algorithms, decision-support outputs can quickly identify a patient’s current clinical status across organ systems. Offering a great example of just-in-time analytics improving provider’s existing workflows.

graymatteranalytics.com

[1] Making Predictive Analytics a Routine Part of Patient Care. Harvard Business Review. Ravi B. Parikh

Page 19: [E book] Moneyball and Analytics  - Modern Healthcare's Guide to Predictive Analytics

14Improving Routine Care Through Predictive Analytics.

USE CASE:Create context by aggregating patient data at the point of care

graymatteranalytics.com

[1] The Future of Personalized Healthcare: Predictive Analytics. Rock Health. View online

Lacking appropriate context, clinical indicators—including vital signs—can generate false positives or negatives in alert systems.

Using textbook definitions, 14% to 38% of heart rate observations and 15% to 30% of respiratory rate observations would have resulted in false alarms.

The importance of having context around an individual’s historical data is so crucial, especially in healthcare where we are still learning what “healthy” or “normal” is. Textbook guidelines often provide ranges that are integrated into clinical alert systems without any context for each patient. For instance, data collected from the Cincinnati Children’s Hospital Medical Center and Children’s Hospital of Philadelphia showed that 14 to 38% of heart rate observations and 15% to 30% of respiratory rate observations would have resulted in false alerts based on textbook definitions. Lucile Packard Children’s Hospital Stanford has implemented these findings around their early warning algorithm systems. 1

Page 20: [E book] Moneyball and Analytics  - Modern Healthcare's Guide to Predictive Analytics

ACCELERATE YOUR ANALYTICS

With Gray Matter Analytics CoreTechs™, our Analytics as a Service cloud platform manages healthcare's digital transformation with:

• Rapid analytics deployment and scalability• Advanced IT capabilities - machine learning

and predictive analytics• Minimal IT cost• Enhanced data privacy and security

Your Cloud Platform To Power Healthcare 2.0

Combining the on-demand benefits of cloud computing with the knowledge empowerment of data analytics, our CoreTechs platform enables healthcare

organizations to quickly structure their data around their evolving business needs.

Contact us today to learn more or visit us at online.

Learn More →

Page 21: [E book] Moneyball and Analytics  - Modern Healthcare's Guide to Predictive Analytics

ABOUT GRAY MATTER ANALYTICS

Founded in 2013, Gray Matter Analytics specializes in data strategy and advanced analytics capabilities. We help healthcare enterprises excel in the digital economy.

Through advisory services and educational workshops, we add the technology, intellectual capital, and processes to deliver on the promise of data-driven decision making.

Whether deploying a technology infrastructure, building a data science practice, or enacting robust data governance, our team understands healthcare’s unique goals and challenges.

Flagship services and solutions include; our Mind Assessmentmanagement solution methodology, GMA Genius® solution prioritization, and our hosted, Analytics as a Service (AaaS), CoreTechs™.

We have offices in Chicago and San Francisco.

www.graymatteranalytics.com