September 2021 | Point of View

Why health payers need insight democratization to unlock the value of data

To meet today’s challenges, it’s crucial that payers decentralize data and make it accessible enterprise-wide

By Dawn Kramer, Sumitro Sarkar, and Gina Lusch

Leaders at health plan organizations know that many of the internal and external pressures they face can be alleviated by a data-driven culture that enables them to unlock the value of enterprise data, manage KPIs more effectively, and support better decision-making.

The potential power of data is clear, whether it’s insights to transform decision support around medical loss ratios (MLR), claim settlement cycles and denial rates, provider contracting, first contact resolution rates, or other critical cost containment areas.

Adopting a hybrid insight development model, then, is the most effective way to democratize data for decision support.

Repeatable functions around data ingestion, curation, and storage when standardized through consistent processes will reduce fragmentation and eventually drive down fixed costs.

The key challenge is time to market—getting the enterprise data to a state that is reliable, accessible, consistent, and robust to support insight building.

Enterprises spend entirely too much time enabling these foundational capabilities, to the point that they lose time that should have been spent deriving insights from the data to support tactical and transformative decisions.

This situation is only magnified by the numerous external and internal pressures (from the insight demand side). A value-driven, platform-based analytics structure is the path to responding to—and even thriving within—this confluence of challenges.

Increasing competitive pressures, like trend of payers bundling benefits and enhancing care models in pursuit of market differentiation.
Uncertain risk profiles, particularly because senior populations deferred medical care during the pandemic.
The emergence of new treatment protocols, namely the rise in virtual and remote treatment amid COVID-19, which also sows uncertainty for payers.
Increased government influence is ratcheting up the pressure on payers; enrollment in Medicare Advantage plans, for instance, grew 9% in 2020.
Increased adoption of virtual care with a slower reimbursement strategy from CMS and others

Increased administrative costs — insurers now spend $2,500 per member.
Increased medical costs. As a result of aging populations, the prevalence of chronic conditions, and utilization trends, the cost of care has been rising for years. In 2019, it was predicted that overall healthcare costs would rise by an average of 5.5% per year over the next decade — growing from $3.5 trillion in 2017 to $6 trillion by 2027.
Limited product innovation. Many health plans still use inefficient processes and monolithic systems that increase tech debt while sophisticated visions for digital transformation lag behind. Facing new competition, heightened consumer expectations, and internal concerns about cost and the need to quickly solve immediate issues, decision-makers often choose patches and short-term fixes over the bigger strategic picture. One result is the difficulty in accessing and aggregating data across enterprises.
Rising technology debt and a lack of vision for digital transformation. Payers often have a lack of consensus and coordination around strategic goals and the decisions needed to achieve them. These organizational barriers are exacerbated by heuristic cultures and complex processes. Meanwhile, technological barriers make it difficult to access and aggregate data across the enterprise.
Lack of insight-driven action. Gaps in data, technology and analytical skills, as well as challenges in data access and quality, lead to sub-optimal insights for decision support.

Decision-making needs to be enabled across the enterprise and aligned to an organization’s most pressing business problems. Standardization of data-driven decision support is the right idea, but not at the expense of equitability and accessibility across an organization’s various functions, tools, models, and incentives.

In other words, seeking the lowest common denominator for enterprise analytics can be unintentionally reductive, limiting leaders across the enterprise to a state of linear dependence on a central analytics organization and disempowering them from generating the insights they need on-demand.

The solution is a prefabricated data analytics platform built specifically for an approach we call “insight democratization.”

This approach is founded on several key conditions that prevail in an enterprise:

Datasets are decentralized (i.e., no centralized System of Records for all data)
Multiple versions of datasets leading to questionable accuracy and consistency
Limited accessibility (i.e., lack of process and a source for specific datasets)
Lack of centralized management capability for key foundational capabilities such as member/patient segmentation, provider attribution, risk adjustment etc.
Lack of a well-formed and automated “innovation-to-test” sandbox for citizen data scientists

Fortunately, most health plans already have the necessary capabilities to put this in place. The key is the time to enable foundational data capabilities to even start leveraging the data assets and adopting and activating an ROI-based perspective that focuses on accurately measuring the impact of insight-driven actions in a process of continuous improvement.

We see leaders struggling to identify a “starting point” and to bring this to life. Yes, the data is available; yes, an analytics organization is in place; yes, core metrics and KPIs have been identified—but how do you bring it all together without imposing costly limitations on your various business units or losing sight of the need for a continuous feedback loop that will truly allow for growth and improvement?

With a value-based operational mindset, any healthcare payer organization can implement an insight democratization approach built to overcome the challenges they face today.

The fundamental principles of this approach include:

A platform-based DataOps and DevOps approach to reduce the “data to insights” lifecycle
Establishing capabilities to persist data that can be trusted as accurate, consistent, and easily accessible
A set of reusable capabilities that allow automation of repeatable insights
A data model and structure that aligns to the business functions, and is scalable
An operating model that allows subject matter experts to seamlessly derive insights from relevant data
The ability to support decision-making across the enterprise in a way that is standardized, equitable, and accessible—without reducing such functions to a sub-optimal decision support model driven by consensus, cost-efficacy, and the ability to deliver “average” answers to specific problems

Yet, the familiar limitations of siloed and/or centralized data architecture and analytics models make it difficult, if not impossible, for this mode of insight generation to achieve results.

What’s left is a “linear" demand-supply approach to insight development: multiple backlogs of workflow, stopping points to learn new tools, and decision-making based on resource availability.

This impedes the productivity of the value-creation entities—staff and teams—in a quagmire of operations. Even with new data platforms, many payers continue to struggle to align analytics to real business decisions.

Insight democratization to the rescue: A new kind of data platform

Insight democratization all but eliminates these challenges and inefficiencies, empowering functional leaders and their teams to easily access the data they need to support the decisions they need to make, when they need to make them, without being stuck in a silo or having to go through an intermediary organization.

Moreover, with a data platform that accelerates and scales an insight democratization approach, leaders can be assured of desired data accuracy, acceleration, and consistency to reach the optimal decision support throughput:

Insights-driven advisory where the front-end insight becomes the decision
Tested hypotheses (not starting from scratch) that support quicker decisions
Scalability supported by the consistencies of a platform-based approach

The platform also should be designed to foster a community of producers and consumers of insights who build reusable components that can be leveraged to accelerate insight-building—uniting “affinity groups” of producers and consumers of data who “govern" data accuracy to instill and maintain trust.

For instance, in their book The Platform Revolution, Geoffrey Parker, Marshall W Van Alstyne, and Sangeet Paul Choudary describe how a “network" model has allowed more optimal management of demand supply where insight (content) producers can also be insight consumers, and vice versa. We are taking this notion a step further and building it directly into the DNA of a healthy payer analytics organization.

That said, it’s important to not let the hype around new technologies—like artificial intelligence and machine learning—distract from the identification of the vital business problems first. Adopting a hyper-focus on value means determining what business questions and KPIs can be addressed by data and analytics, and how they will help fulfill an organization’s overall vision.

As part of this focus, it’s important to ensure that programs and initiatives are having their intended impacts. By connecting actions to measures (and aligning people, solutions, and measures to overall performance), leaders and functional teams can decide on subsequent actions with more intention and purpose. Platform-based analytics within an insight democratization approach is the way to achieve this.

Seize the opportunity to enable insights—not just generate them

In any number of other industries, success stories abound when employees across an enterprise have access to data to support decision-making.

There’s no reason why health insurers can’t do the same. With an insight democratization approach, payers can improve MLR, increase efficiencies around claim processing (including settlement cycles and denial rates), and enhance provider lifecycle management—among other significant value additions.

Implementing a new data platform is no easy feat—but we also know that the role of an analytics organization should increasingly be about enabling insights instead of simply generating insights.

With this approach, payers can contain costs and create industry-leading, data-driven operations that are better equipped to serve staff, physicians, and ultimately the members they serve.