Spending, Prescription Histories Identify Future High Cost Members

October 10, 2018 - An individual’s healthcare spending and prescription drug histories are among the most accurate predictors of future high costs, according to a new Society of Actuaries (SOA) report.

SOA used claims data collected by the Health Care Cost Institute (HCCI) to determine which beneficiary characteristics best predict whether or not members will become high cost claimants.

SOA believes using these data elements can help payers and other healthcare organizations predict potential healthcare costs more effectively than traditional risk adjustment methods.

SOA estimates that 17 percent of the most expensive healthcare users account for 75 percent of all US healthcare spending, creating a financial and clinical imperative to identify high spenders as quickly and accurately as possible.

“Due to the size of the contribution to costs of this small segment of the population, there is considerable interest in understanding what portion of it can be predicted,” SOA said.

“While risk adjustment models are good at predicting average costs of care for a specific category of people, they are still not effective at identifying individuals who may be at risk for a very high claim in the near future.”

SOA found that patients with an expensive claims history are likely to experience even higher healthcare care costs in the near future.

Individuals with annual claims totaling $100,000 in 2011 were likely to incur $1 million in spending by 2015. Enrollees with significant previous care costs also had a greater likelihood of exceeding $1 million in healthcare claims than did other beneficiary groups.

“Member cost history has the biggest impact on the probability of an individual being high-cost,” SOA said. “When looking at the cost history, if members were high-cost last year, they are more likely to be high-cost this year. The impact increases as the prior year costs increase.”

Perhaps unsurprisingly, the advancing age of a health plan member was also correlated with increased spending.

“The age pattern is relatively consistent across years,” the team explained. “For the $100,000, $250,000 and $500,000 claims categories, members are more likely to be high-cost as they age.”

An individual’s prescription history can also help to predict a potentially high-cost enrollee, the team found. However, members with a history of prescription drug use are less likely than the other observed groups to incur claims beyond the $500,000 or $1 million range.

Gender and individual health plan enrollment are also helpful predictors of potential high claims costs, according to the report.

Male beneficiaries are likely to incur slightly higher costs than female beneficiaries. Individual health plan enrollees are also more likely to have higher claims than private insurance enrollees, the team added.

SOA believes that cost history data may be a highly effective tool for payers to determine future care costs and improve spending predictions.

Beneficiary cost history, prescription drug coverage, and other types of cost predictors may help actuarial organizations evaluate programs designed to contain costs or the impact of federal policies on healthcare spending.

“There are many potential avenues for future work,” SOA concluded.

“With the HCCI data, it would be very interesting to explore the many relationships among the members (spatially, temporally and hierarchically) in more depth. It would also be interesting to try to quantify the impact of wellness programs or to explore further the impact of the Affordable Care Act.”