Editor’s Note: For state health programs, payers, providers and life sciences companies, data and analytics have become essential to facilitating efficient and effective healthcare delivery. The right data assets and analytics expertise are core requirements for ensuring access to care, optimizing population health initiatives, and achieving quality and financial goals.

A Little Chaos Makes It Exciting

There is a quote by science fiction writer Amelia Atwater-Rhodes that is the ideal lead-in to any discussion about healthcare data and analytics: “Life is nothing without a little chaos to make it interesting.” By that definition, life is very interesting these days for those who are dealing with healthcare data. Healthcare leaders have to navigate a complex array of databases and technologies to make the right decisions about the optimal tools to invest in for their organizations. They have to make those decisions within often stringent time and budget constraints—and balance them against other priorities.

The technology sector has seen an opportunity in this struggle to apply its expertise in data aggregation, predictive analytics, data mining and software to healthcare. Technology companies are continuing to move into the healthcare space, bringing new ideas and options—but also critical challenges.

The Three Big Challenges in Data-Driven Healthcare

Healthcare is facing three big challenges around data and analytics:

  1. The purpose of data has changed. In the past, data was important for tasks such as paying claims, seeking Food and Drug Administration (FDA) approval for a new drug or device, and supporting financial and clinical transactions. In the past 10 or 15 years, however, the questions being asked of data sources are very different than the ones those sources were built to answer. We’ve gone from needing data only to answer very siloed questions to needing data that cuts across various segments and facets of healthcare. The challenge is figuring out how to use real-world data to advance us toward high-value, “frictionless” healthcare.
  2. The development of tools for making data impactful has lagged behind the development of IT infrastructure. In many ways, IT has moved ahead faster than the tools needed to put it to work supporting clinical practice, shared decision making and care management. There are many new entrants creating software and tools to make data assets more usable, valuable and streamlined for end users. But we are not there yet.
  3. There is a lot of technology innovation, but it is difficult to identify the highest-value advances and integrate them into an organization’s workflows. It is important that organizations find ways to discern the best options from the ever-expanding array of tools to support their goal of being high-performing healthcare systems. Questions remain around how decision makers can use the power of big data in ways that are of the greatest value to themselves, their organizations and the healthcare system as a whole.

It is critical that healthcare leaders overcome these challenges and begin to incorporate data and analytics into their organizational work streams to inform decisions, improve communication among care team members, and help achieve quality and financial objectives. What’s driving the crucial need for data and analytics? There are nine megatrends that are key to realizing the promise of data-driven healthcare.

Trend 1: Stakeholders Search for Meaningful Information in a Data-Abundant World.

Data plays a critical role in the modern healthcare payment and delivery system. With more data than ever available for use, organizations must become more sophisticated data consumers, carefully assessing the usefulness of data to inform their program and policy goals.

A Stanford University report noted that by 2020, we will be generating 2,314 exobytes of healthcare data annually—enough to fill the typical PC about 100 billion times. That data holds the promise of bringing significant improvements to our healthcare system. But the promise can only be realized when the data is relevant to our policy and program objectives, easily accessible and usable, and able to be connected to actionable workflows.

For all of its volume, velocity and variety, most healthcare data is only accessible to a subset of users at a time. It is often unrefined and unstructured—and lacking standard definitions and proven use cases. To help realize the promise of data-driven healthcare, organizations are increasingly developing and implementing strategies that will define their data and analytic needs—and connect the right data sources with the right users at the right time.

As our healthcare system expects more from us—to manage population health, cut costs, increase efficiency and mitigate the risks inherent in value-based purchasing—the need for longer-term data strategies is growing. The role of commercial vendors is also growing to fill that need. We are seeing an increasing number of commercial vendors specializing in enterprise resource planning, performance management, business intelligence and data aggregation. They are expanding their capabilities, methods and reporting tools as the market evolves to ensure they are ready to meet emerging requirements.

Public data agencies are also calibrating their roles in this new landscape. Consider, for example, that the Centers for Medicare & Medicaid Services (CMS) is the steward of some of the most valuable healthcare data sets in the country. As we move forward into a new data-driven world, public entities are uniquely positioned to fill the public’s healthcare needs by allocating resources toward higher-value activities.

Trend 2: New Analytics Have the Capacity to Reveal Healthcare Cost Trends.

Data and analytics are enabling stakeholders to understand their cost drivers and develop strategies to address them. Payers and providers, for example, are using data to monitor spending by subpopulations and service lines, develop more efficient clinical practices, test new payment methodologies, and strengthen their referral patterns and networks. They are also using their enhanced data capacity to control administrative spending, streamline claims processing and conduct artificial intelligence (AI)-powered fraud analyses. One payer recently developed its own internal data pipeline and customized AI algorithms to flag potentially fraudulent claims early, saving millions of dollars annually.

Public agencies, such as CMS, are using data to reduce fraud, promote price transparency, respond to cost concerns and reveal price variations for clinical services. In addition, statewide total cost of care measures are allowing policymakers to understand market cost drivers and develop targeted policy solutions to address them.

Employers also are using data to address healthcare costs and understand the impact of those costs on their bottom lines. Self-insured employers particularly—who cover approximately 60% of the U.S. workforce—are implementing new analytic and technological solutions to monitor their employees’ well-being, encourage healthier lifestyles among their employees and predict those at risk of disease to promote early intervention.

In addition, large companies are using data to bypass traditional insurer networks and directly negotiate and contract with providers. A growing number of major employers are also working with technology companies to develop collaborative solutions that reduce costs.

Finally, consumers are more frequently engaging with new applications to assess their plan options and engage in service shopping. With more than a quarter of consumers now in high deductible health plans and many facing even higher coinsurance liabilities, we anticipate there will be an even greater willingness to engage with pricing tools.

Trend 3: Consumers Are Generating and Using Health Data, but Clinical Connections Remain Limited.

Consumers, particularly younger generations at home with the Internet of Things (IoT), are increasingly comfortable using devices to monitor their health and well-being. Whether this data will be clinically useful remains an open question.

According to one study, global wearable shipments exceeded 100 million units for the first time last year. During the same time period, U.S. residents downloaded more than 200 million health applications across a wide range of areas, from treatment and prescription compliance to fitness.

Providers are trying to use technology to increase patient engagement by opening up personal health records for patient access. According to a recent study, in 2015, just under 70% of the nation’s hospitals offered the ability for patients to view, download and transmit their personal health data, while 63% allowed patients to securely message their providers.

Similarly, payers have tried to create transparency tools to steer patients toward value. Patient engagement, however, remains low. A Health Affairs study of 3,000 non-elderly adults shows that only 13% of those who had out-of-pocket spending sought out cost information, and only 3% compared costs across providers.

While there are many new tools built around consumer-generated data, they are not well connected to clinical practice yet. Successes have been limited to localized cases involving targeted health coaching around specific disease states, such as diabetes. As the market evolves, however, health systems, vendors and consumers all will be looking for applications that bring clear value, drive behavioral change and create more seamless care delivery.

Trend 4: Population Health Is an Intriguing Paradigm, but Data Development Is Slow.

Population health focuses on health outcomes, cost and quality for groups of individuals. In general, population health programs have yet to show a broad-based return on investment.

Some delivery systems, in the context of their value-based payment arrangements, are reaching beyond their four walls to manage the populations they treat. In the case of low-income populations, that may include addressing social factors that affect healthcare outcomes.

Data, methods and infrastructure development to support population health have lagged for a range of reasons, including technical complexity and privacy issues. Addressing these deficits is critical for increasing the value of population health programs moving forward.

Even when delivery systems are committed to making changes to clinical practice to support population health, they face significant challenges. A key hurdle is that funding for infrastructure and data is often inadequate. In addition, while data science can provide the capability to micro-target specific patient subpopulations and identify gaps in care, standardization of that data, as well as interoperability across providers and, where needed, social service agencies, isn’t well developed.

Technology companies recognize that there are complex delivery system data needs. Traditional healthcare information exchange and electronic health record (EHR) vendors are working to modify their tools to standardize new fields and enable data integration. Cloud service providers are promoting environments for collecting, sharing, analyzing and reporting data for population health models that require collaboration between healthcare stakeholders and social service agencies.

Both traditional vendors and cloud service providers are working to bring in more data sources, provide advanced predictive analytics, and support the dissemination of more meaningful information to providers and care managers through dashboards and reporting platforms. All of these factors are foundational for building effective population health initiatives.

Trend 5: Sharing Data Is Not Widely Accepted as Good Policy or Business.

Healthcare data has existed in silos for a long time. Changes in policy and care delivery, however, are now requiring stakeholders to share data about patients so treatment can be managed across the care continuum.

Sharing data between and among different entities is extremely complex. Data sharing carries business risk, and efforts to pool data are hindered by legal and regulatory frameworks.

For both providers and plans, data is an extremely important competitive asset. Market incentives still exist for keeping patient panels, demographic information, and utilization and cost data in-house. But a push from Medicare, as well as state reforms that are happening in Medicaid, are providing both incentives and penalties to encourage data holders to share information. To be willing to share their data, providers and plans need to know that what they gain is more valuable than what they might lose in terms of perceived competitive advantage.

Even when stakeholders are willing to share data, there’s a patchwork of federal and state laws that can present challenges. For example, the federal substance abuse confidentiality regulations almost always require patient consent for data to be shared—and the consent form requirement sometimes means that a single patient has to provide consent on multiple occasions.

State laws related to mental health, HIV and other types of sensitive health information vary. They also can often prohibit data sharing without patient consent or limit the circumstances under which data can be shared.

Trend 6: Analytic Methods Are Proliferating. AI Is Emerging as the New Frontier.

The rapid progression of methods and predictive analytics in the healthcare industry has generated new and exciting tools, as well as new ways of using and thinking about healthcare data. In the near future, the market will begin to focus on high-value use cases around the types of analytics that organizations are incorporating into their performance management and clinical activities.

AI will play a big role. There is promising early evidence around the potential impact that AI may have on clinical practice, performance management and delivery systems. There also, however, are significant challenges to broad-based implementation, including data quality issues and technical infrastructure needs. In the future, however, we believe AI applications will be developed that allow more organizations to apply and acquire AI capabilities. AI just may not be the immediate-term disruptor that it has been framed as in other discussions.

There is no “one size fits all” answer in terms of the right analytics for an organization. Local data strategies are required for healthcare industry stakeholders to integrate data effectively and make the optimal analytic investments. Most critically, organizations must focus on strategies that link their investments to clearly defined outcomes and/or mission goals.

Trend 7: Data and Analytics Enable Clinical Transformation on the Front Lines of Both Care Delivery and Research.

Many delivery systems are seeking to identify opportunities to leverage investments in health information technology (HIT) and private and public health information exchange capabilities. They are recognizing the chance to improve information sharing among providers and other players, as well as to integrate with clinical and care team workflows. New technologies also offer the opportunity to share information between providers and payers to support more effective population health management.

One of the most exciting areas of growth is the integration of data and predictive analytics from new platforms with existing clinical protocols to drive increasingly sophisticated clinical decision support. Historically, there have been challenges around the proliferation of data that isn’t as well-targeted or integrated as it could be, making it difficult for providers to engage and take action. To overcome this hurdle, clinical decision support systems are focusing on dashboard reporting and message-based tools that provide actionable information while reducing provider and administrative burden.

Finally, we can’t discuss clinical transformation without speaking about precision medicine. Like AI, precision medicine is a newly emerging area. Unlike AI, however, it already has generated a significant number of real-world cases to prove its value. Life sciences companies, providers, and clinical researchers are increasingly leveraging new genomic, biomarker, and molecular data to drive more accurate diagnoses and more targeted treatments.

Trend 8: Providers and Payers Are Increasingly Using and Sharing Data to Manage Performance and Demonstrate Value.

Payment and delivery transformation is not just about cost management. It encompasses the larger concepts of value-based payment and alternative payment models, as well as of targeting resources within the healthcare system to improve outcomes, quality and value. There is a tremendous amount of innovation and interest around value-based transformation across both the public and private healthcare sectors:

  • Providers: Many leading health systems are building data and analytic capabilities to monitor and manage performance, reduce adverse events and readmissions, and improve quality. The trajectory of value-based performance remains unclear, however, which could be a hurdle to broad adoption of data-driven performance management.
  • Public sector: CMS has implemented a range of alternative payment models and has been a driver behind both sharing and using data to help providers focus on achieving quality and performance management goals. Some states, particularly those pursuing Medicaid innovation and other waivers, are required to measure and report performance data to demonstrate delivery system performance improvements.
  • Private payers: Private payers are continuing to innovate using claims and other data to target high-risk patients, as well as to engage care managers and in-network providers in performance management initiatives.
  • Life sciences companies: Some life sciences companies are beginning to explore value-based contracting with payers. This is a very new approach but demonstrates that value-based thinking is extending into the life sciences sector.

Trend 9: Key Data Gaps and Limitations Inhibit Payment and Delivery System Strategic Planning.

Data gaps are more of a mega challenge than a megatrend. Many organizations struggle with significant data gaps in a host of areas, including the universe of providers and the relationships with systems, clinicians and service locations, as well as around payers and enrollment.

One of the most critical issues is the lack of clear, shared data sources that define the relationships within delivery systems. The delivery system has been changing rapidly in recent years—and that will continue as disruptive delivery models and payment relationships proliferate. Traditional data sources that were used in a volume-focused context don’t help us understand our new value-based environment. Shared data sources don’t currently exist. All stakeholders in the market would benefit if there were shared capabilities to manage and link provider, payer and patient information while at the same time reducing administrative burden.

The public sector is most likely to develop a shared “source of truth.” State agencies have the ability to develop, leverage, and support shared capabilities and systems that can serve as a common data source across payers and providers. Some private-sector organizations and data vendors have sought to enter this space in limited ways, but they are unlikely to meet the need for a truly shared source—though there is the possibility for a public-private collaboration to create an effective solution.


In developing these nine critical megatrends around data-driven healthcare, an array of key questions arose that don’t yet have answers: Will organizations be able to keep up with a continually shifting technology landscape? What about those left behind? Will a data divide emerge? How much of data’s potential to improve healthcare is real—and how much is hype? How much of our new data analytic capacity will generate long-term change? Can the industry develop well-calibrated solutions that bring real value while minimizing provider burden?

The long-term prospects for data-driven healthcare will be determined largely by whether key stakeholders can demonstrate value from the data-to-workflow connections they establish. At Manatt, we will continue to monitor the rapidly changing data landscape in healthcare—and help all of the players overcome current challenges to integrate effective data solutions into their organizations.