Home  |   Subscribe  |   Resources  |   Reprints  |   Writers' Guidelines

Web Exclusive

How to Craft an Analytics Strategy
By Jimmy Liu

Analytics is being used everywhere today: sports teams use data points for drafting and signing the players that will work best with their strategy; banking uses it to predict future valuation; and retail stores bring in analytics to send customers the most appropriate ad or coupon. Each day, our world depends more on the ability to analyze large amounts of data in complex ways.

Health care organizations are now leveraging data analysis for business purposes—and it is proving to have profound results. Risk-adjusted data analysis can have a powerful effect on improving patient outcomes, reducing emergent care, and increasing the revenue of health plans and health care organizations. If done correctly, analytics in health care, like in sports, banking, or retail, can help organizations gain a competitive edge. For example, analytics can create disease-based audits or help health plans recover millions of dollars through retrospective chart reviews.

While these potential benefits make a good case for crafting a full-bodied analytics strategy, doing so is not as easy as it sounds. Health plans tend to be siloed in their approach due to a lack of resources—both financial and staff-related. Hospitals and other health care organizations also lack funding for analytics projects and staff with the appropriate skill set to analyze large data sets. Should a health care organization find the means to hire a qualified analyst to extract meaningful data, there is often a training gap as it relates to analytics around risk adjustment to evaluate risk score gaps, among other things. It also is important to develop and retain such talent.

If a health care organization is working with vendors in creating an analytics strategy, it still will require internal competencies to determine whether the strategy offered for a particular outreach project is feasible, executable, and financially responsible. In other words, an organization needs a process to determine that vendors are delivering what they say they are.

Best Practices for Creating an Analytics Strategy
Given that these challenges seem daunting, here are four best practices an organization should consider when developing an analytics strategy.

1. Assess in-house talent. Before seeking outside help, look across the organization and identify potential individuals—such as those with a proven background or those with both technical and business experience—who may be able to handle an analytics project. It helps to make a list of the skill sets that would be required for the project and ask, “To what level do I need in-house capabilities to either do some of these analytics, check the results of projects, or improve on them?” Finally, an organization must assess whether it’s possible to allocate such individuals partially or fully for the purpose of health data analysis.

2. Evaluate supplemental needs. After taking an inventory of in-house talent, an organization may want to take a closer look at its vendor partnerships. If, for example, the organization recognizes the importance of analytics but does not have the capabilities in house, whether staff or IT resources, there are vendors that recognize the value and complexity of analytics and have the manpower required for the project. A careful review of vendor competencies and industry reputation helps identify the best fit to meet an organization’s specific needs.

3. Allocate a specific, realistic budget. It costs money to make money. That’s why it’s important to create a realistic budget to present to management that includes financial requirements needed for a risk-adjustment analytics team, training costs, and technology. In creating a budget, an organization must ask itself, “How much should I invest in a risk adjustment plan?” as well as “How many dollars do I have at risk?”

Budgets must reflect the magnitude of the revenue at risk rather than an organization’s size and available talent pool. Sometimes the size required for an analytics team will vary depending on the complexity of the project. Health plans may have eight people on an analytics team working on a particular project while a different project may utilize only three people. In addition, if vendors are being leveraged, the scope of capabilities and team required in-house has the potential to change.

4. Focus on a core competency or specific area. Analytics can be defined by what actions an organization is willing to take to mitigate risks. When setting a goal for an analytics project, an organization should ask itself, “What are my areas of improvement?” “Should I go bigger or smaller?” and “Should I be more refined in my data mining before I make those decisions?” Also, before shifting gears from one analytics project to the next, it’s important to make sure every effort to improve—for example, a quality indicator score—has been exhausted through previous data-mining efforts.

Refining data analytics and algorithms based on prior results is the best way to make incremental improvements over time, but understanding the key metrics to review and what refinements to make must be the focus of the assembled analytics team.

Ongoing Data Analytics Efforts
After building an analytics strategy, one of the most important things to keep in mind is that the project is part of an ongoing effort. Be prepared for the long haul. Most health care executives take a one- to two-year approach to projects and may not want to commit to anything beyond that.

With any risk-based analytics project, it takes time to train people and mine data. Even with the right talent in place, it takes a while to build capabilities and changes will not be achieved overnight. Setting a strong foundation of internal and external resources and making adjustments over time to improve performance will set the organization down the path of crafting a sound analytics strategy.

— Jimmy Liu is vice president of risk analytics services for Altegra Health. An expert in risk analytics and risk scoring methodologies, Liu leads the risk analytics team and also provides sales support to the Altegra Health sales and account management teams.