Spotlight on High-Cost Claims in Q3 2024 Trends

This issue of Trends also includes a reminder about our handy summary charts, the ACA Dollar Amounts and Percentages, which we updated as new information becomes available. The most recent updates were in May.

Key statistics

Medical HCCs

Rx HCCs

HCC Risk Factors

* 2023 medical claims are incomplete due to lack of runout data.

Source: Segal’s SHAPE data warehouse

Mitigating high-cost claims

As illustrated, over the past five years, top 1 percent of medical claims have consistently represented approximately one-third of all medical expenses with one exception: 2020. Rx HHCs represent almost half of all expenses. Consequently, effective strategies for predicting, preventing and managing HCCs have become increasingly important for plan sponsors.

The Pareto Principle and leveraged trend

Medical claims closely follow the Pareto Principle, which states that 80 percent of the consequences come from 20 percent of the causes. In fact, the top 20 percent of medical claims account for approximately 85 percent of all medical costs. For prescription drug costs, the top 20 percent of claims account for a stunning 95 percent of all prescription drug costs. To manage healthcare costs, it is imperative for plan sponsors to get a better understanding of underlying drivers and identify proven intervention strategies to mitigate future HCC risk and improve population health.

HCCs can also be defined by a dollar threshold (e.g., $250,000). However, when doing so it is easy to mistakenly conclude that HCCs are trending faster than overall medical claims. This phenomenon is actually attributable to leveraging, where trend increases as the fixed dollar threshold gets higher. For example, if overall medical trend is 6 percent, we’d expect claims greater than $250,000 to increase 13 percent and claims above $500,000 to increase by more than 15 percent. However, HCCs are trending similarly to other medical claims.

Improvements in identifying HCCs

Reviewing insurers’ HCC reports often leaves plan sponsors feeling helpless. The diagnosis descriptions provide little insight into true causes of HCCs and potential intervention strategies. To identify how a high-cost claim may be prevented, one must also look at the claims history and identify the risk factors associated with the unfavorable outcomes.

Early identification of risk factors associated with a future high-cost event allows for proactive interventions and care management strategies to be implemented, potentially leading to improved health outcomes and significant cost savings.

What are the risk factors?

It is relatively rare for claims to be in the top 1 percent of costs without any prior indication of risk. Although there are dozens of chronic conditions affecting the general population, a small subset are responsible for a disproportionately large amount of illness and death. Over 70 percent of the population in the top 1 percent of medical costs have one or more of the six chronic conditions shown in the third graph.

Plan participants with CHF are 15.7 times more likely to have an HCC than the general population. However, less than 1 percent of the population is diagnosed with this condition and plan sponsors will likely be able to make a bigger impact by focusing on managing other conditions to prevent disease progression that results in CHF. Participants with CHF have on average 2.5 other chronic conditions, as illustrated in the third graph, with hypertension (93 percent), CAD (66 percent) and diabetes (58 percent) being the main ones.

The prevalence of these chronic conditions continues to increase amongst high-cost claims. In 2019, 70 percent of the top 1 percent of medical claims had one of those chronic conditions, and by 2023 it increased to 72 percent, with increases being exhibited each year. Without comprehensive population health management programs in place that address these conditions holistically, plan sponsors will see little progress in mitigating these high-cost events.

Predictive modeling

Advancements in predictive models have shown promise in recent years in a variety of areas in the healthcare delivery system, including risk stratification, optimizing resource allocation, reducing hospital readmission rates, predicting adverse events and predicting disease progression. Predictive models use advanced machine learning algorithms to identify complex relationships among predictor variables and their target (e.g., medical costs).

Although administrative claims data does not contain the level of detail required to accurately predict all HCCs, it can provide valuable insights into them and the factors driving that risk. Furthermore, it allows plan sponsors to stratify their populations to focus on the those most at-risk of adverse events to target for disease management programs.

HCC management strategies

Viable HCC management strategies should start with a data-driven approach that is customized depending on each plan’s needs. Common approaches include:

• Preventing and managing chronic diseases. A comprehensive wellness program can help plan participants get engaged in their health and prevent onset of chronic conditions. For those who do develop chronic conditions, plans should have resources in place to help them manage their conditions effectively. This could be anything from a point solution that specializes in certain conditions to providing benefits websites that summarize effective exercises and/or recipes relevant to each condition (e.g., keto-friendly recipes to manage diabetes).
• Implement and monitor case management. Plans should discuss how HCC cases are identified and managed with the medical vendor. Vendors may need to lower identification thresholds and/or change how they conduct patient outreach depending on the population.
• Implement proactive claim reviews. Plan sponsors should implement a support structure for large claim reviews to detect the billing errors and abuses that can occur on very large, complex cases.
• Promote preventive cancer screenings. Cancer is a common cause of HCCs. Over half of cancers experienced in most populations have screening guidelines depending on age and gender. Additionally, the recommendations are constantly changing. Plans should educate and promote these screenings regularly.
• Consider a stop-loss policy. Some of the highest cost claims include conditions such as burns, leukemia and premature newborns, which often have few intervention opportunities available. Plans should be aware of these possibilities and plan for them accordingly. If a plan isn’t adequately funded to cover these events, stop-loss policies should be considered based on risk tolerance.

Compliance reminder

Stay on top of ACA amounts

Get our handy summary charts, the ACA Dollar Amounts and Percentages, which we updated in May.