Not a Member?  Become One Today!

How Medical Claims Data Drives Chronic Disease Management
 

By Joseph Berardo Jr.  12/16/2013
 

The U.S. health care system is shifting toward a patient-centric, outcomes-based structure, and employers are keeping pace with new health plan strategies of their own. A 2013 study by McKinsey & Co. evaluated a range of health care initiatives and found that their potential impact could mean $300 billion to $450 billion in reduced health care spending, or 12 percent to 17 percent of the $2.6 trillion baseline in U.S. health care costs.

For the most part, employers want more value beyond access to care, with customizable care plans based on an individual’s risk profile and needs. Targeting health issues, rather than simply implementing a one-size-fits-all health and wellness program, is critical for long-term sustainability.

One increasingly popular strategy involves employers entering into partnerships with health care service companies and provider groups to take advantage of deep discounts and give employees greater access to coordinated care. Within this model, health care data analytics plays an important role, providing information relevant to population health management, such as determining the chances of a relapse, the likelihood of noncompliance, and the progression of chronic disease.

Some plans are designed exclusively around chronic disease and include educational materials, one-on-one counseling, transportation to a hospital or doctor’s office, and assistance in coordinating care among providers/physicians. Health claims and other medical data are used to identify members with chronic conditions and provide them with the tools and support they need to better manage their health.

Take, for example, diabetic plan participants who need assistance to better manage their disease. The right disease management solution can potentially prevent costly future emergency-room visits or hospitalizations. This is significant given that the total estimated cost of diagnosed diabetes in 2012 was $245 billion, including $176 billion in direct medical costs and $69 billion in reduced productivity.

The key features of a chronic disease management plan are data analytics, predictive modeling and intervention.

Data Analytics

An innovative health care services provider should have the ability to integrate data and disease management to categorize and focus on members who will benefit the most from outreach and tools aimed at improving care management. The solution should be able to aggregate member health data, including past claims and medical records, as well as processes.

Individuals who are at the highest risk for serious health issues are identified and may receive tailored interventions to help them overcome their unique challenges. Health and wellness plans, along with incentives that encourage members to be proactive about their health, can be developed based on a given population’s needs. In general, data analytics provides an opportunity to keep costs low and the quality of plan members’ health care at its peak.

Effective Approaches

For maximum efficiency, a data analytics solution should consist of high-level predictive modeling combined with individualized outreach programs and the latest health analysis technology that confidentially assesses the health risks of every member.

Specifically, an effective data analytics solution provider:

    • Identifies population health care issues.
    • Finds patients that may benefit from disease management.
    • Establishes a baseline cost for the member and overall health plan and provides ongoing medical utilization and costs reports.
    • Contacts the identified member to provide information about a helpful disease management program.
    • Provides assistance in navigating the health care system, as well as methods for improving lifestyle.
    • Contacts the provider if there are compliance issues.
    • Refers the patient to a consulting specialist when additional care is needed.
    • Uses contracted nurse practitioners when a participant appears to be in a difficult position and needs additional attention that does not involve a costly emergency-room visit.
    • Addresses all associated co-morbid conditions in the plan member.

Data can be stratified and analyzed through various parameters, including:

    • Pharmaceutical utilization.
    • Lab results.
    • Inpatient/outpatient days.
    • Doctor visits.
    • Disability.
    • Workers’ comp.
    • Absenteeism and presenteeism.

Patients who benefit from intervention include, for example, diabetics who:

    • Have poor control.
    • Do not comply with medication regimens.
    • Have abnormal lab values.
    • Are repeatedly hospitalized or visit the emergency room often.

Predictive Modeling

Predictive modeling involves providing a risk assessment and adjustment process to the employee population to determine if the workforce is susceptible to particular illnesses or disease states. This enables businesses to more accurately forecast medical costs and customize health and wellness programs.

Predictive modeling, which uses algorithms that analyze hundreds of data points to make a diagnosis or a prediction of risk, can:

Based on the results, an average member population may be divided into risk categories and their interventions for each group.

With predictive modeling, the information is more accurate and can provide timely insights into potential risk for members, both current and future. The result is premium rates that better balance pricing pressures with the bottom line, as well as benefit designs that are geared more closely toward the risk of a particular population.

The predictive modeling process usually involves four steps:

    • Collecting, evaluating and integrating member enrollment and claims data.
    • Applying grouping and risk marker identification algorithms—ideally, grouping occurs by episodes of care and the algorithms include categories of pharmaceutical treatments.
    • Summarizing the presence of clinical risk markers to create a risk profile.
    • Assessing overall risk by adding assigned risk weights for all the identified markers.

The five most essential qualities of a predictive modeling program are accuracy, transparency, interoperability, support for diverse and changing operational needs, and industry credibility.

Intervention

Health data must be used to constantly finesse the value proposition and ensure that protocols for a particular disease are producing optimal results. Even small interventions can have an enormous impact. A 2010 study published in the journal Health Affairs found that disease prevention and wellness programs led to a drop in medical costs of about $3.27 for every dollar spent on wellness programs and that absenteeism costs fell by about $2.73 for every dollar spent.

Organizations that focus solely on medical and pharmacy costs when creating employee health benefit strategies risk misidentifying the health conditions that have the biggest impact on productivity. For example, a study by the nonprofit National Business Coalition on Health found that cancer and coronary heart disease were consistently among the top five conditions driving overall benefit costs, but the chronic health conditions that were most important in driving costs related to lost productivity were depression, obesity, arthritis, back/neck pain and anxiety. Addressing health risks through workplace interventions can reduce, or at least slow, rising costs that result from preventable health risks.

Rising health care costs threaten the financial sustainability of companies and the nation. To protect the bottom line and create healthier, more productive employees, companies are focusing on chronic care management and discovering innovative strategies. Data analytics is the key to unlocking the full potential for intervention and positive change in the workplace.

Joseph Berardo Jr. is CEO and president of MagnaCare, an administrator of self-insured health plans for employers in New York and New Jersey.

Related SHRM Articles:

Three Health Plan Data Points to Examine with Brokers, SHRM Online Benefits, January 2014

Groups Offer Guidance for Biometric Screenings, SHRM Online Benefits, October 2013

Avoiding the Chronic Disease Cost Avalanche, SHRM Online Benefits, September 2011

Treat Depression Along with Chronic Illness for Costs Benefits, SHRM Online Benefits Discipline, November 2011

Lessons from First-Generation Disease Management Programs, SHRM Online Benefits Discipline, September 2011

Predictive Modeling: Finding the High-Cost Employee, SHRM Online Benefits Discipline, April 2005 (reviewed October 2013)

Quick Links:

SHRM Online Benefits page

SHRM Online Health Care Reform Resource Page

SHRM Online Wellness Programs Resource Page
Copyright Image Obtain reuse/copying permission