Healthcare under the Delegated Risk Model in California: Lower cost, high quality
16 July 2018
Health insurance is increasingly difficult to afford. As reported in the 2018 Milliman Medical Index (MMI), the typical American family of four covered by an average employer-sponsored preferred provider organization (PPO) plan will have annual healthcare expenditures totaling approximately $28,166. Californians are not exempt from this trend, also paying increasingly high costs for their healthcare. According to the 2013 Berkeley Forum report, employer-sponsored health insurance premium rates were projected to nearly double from 2011 to 2022, ultimately reaching $31,728 for family coverage in 2022. Those premium increases will be borne by both employers and employees. According to the MMI, on average premiums are funded approximately two-thirds by employers and one-third by employees through payroll deduction.
Some good news for Californians is that they would likely be paying a lot more without managed care plans that use the delegated model. In brief, the term "delegated model" describes a health insurance plan where financial risk for healthcare services is transferred from an insurance company to healthcare providers (e.g., physicians or hospitals). Most commonly this involves the insurance company paying a fixed, per capita dollar amount (a capitation rate) to a group of physicians, and the physicians assume financial responsibility to provide all professional services for each health plan member. They may also have full or partial risk for hospital services provided to those same members. In California, capitation can only be used in health maintenance organization (HMO) plans. Other common types of plans, PPO-style plans and other fee-for-service (FFS) plans, cannot use capitation.
Measuring the impact of the delegated model on healthcare expenditures is tricky for at least two reasons. First, the average person who enrolls in an HMO plan might have a different health status from the average PPO/FFS plan enrollee. For example, they might be younger, or just healthier than average. Second, per capita healthcare costs vary by geographic area, for a variety of reasons. HMOs tend to be concentrated in urban areas, while PPO/FFS plans are prevalent in all areas of the state.
IHA Atlas data quantifies savings
Fortunately, data published by the Integrated Healthcare Association (IHA) allows us to compare per capita healthcare expenditures for HMO versus PPO/FFS plans, adjusted for differences in the mix of members by health status and by geographic area. Results indicate that for commercial health insurance plans (i.e., non-Medicare, non-Medicaid), total healthcare expenditures per capita are lower under HMO plans than under PPO/FFS plans, as shown in the graph below. They were 5% lower in 2013 and 7% lower in 2015.*
As previously mentioned, providers can also take on varying degrees of financial risk. For example, they might assume risk for just professional services, which is the most common type of capitation arrangement. They might also have a shared risk arrangement for hospital expenses, sharing with the health plan in any differences between actual and budgeted hospital expenses. At the extreme, providers might accept risk for all healthcare services, under what is often called global or full-risk capitation.
Many industry experts, including a substantial number of providers in California, feel that assuming a meaningful degree of financial responsibility for healthcare purchasing decisions is key to ensuring those decisions strike the right balance between fiscal responsibility and providing high-quality care. In fact, one of two major goals stated by the Berkeley Forum was to reduce "the share of healthcare expenditures paid for via fee-for-service from the current 78% to 50%." The delegated model is also consistent with Medicare's goal of shifting more members from fee-for-service payments to value-based payments.
The IHA data includes splits for HMO members covered by plans using professional-only capitation versus plans using global capitation, adjusted for differences in average member risk scores. In aggregate, without geographic area mix adjustment, the data indicates that total healthcare expenditures per capita were 6% lower under HMO plans using global capitation than under HMO plans only using professional services capitation. However, the data volume is relatively low for global capitation members, and heavily skewed toward Southern California. The data did not seem robust enough to provide a reliable comparison of costs under the two plan types after adjusting for differences in the geographic mix of members.
Moving along the spectrum of managed care with global capitation at one end, we find more loosely managed plans at the other end. Such plans do not use capitation, although they often incorporate certain managed care activities, such as large case management and disease management programs. Accountable care organization (ACO) plans tend to fall at this end of the spectrum. On a risk-adjusted basis, the IHA data indicates that total healthcare expenditures per capita were lower for members in ACOs than for members not in ACOs. They were 6% lower in 2015, the only year for which IHA has published this data, on risk-adjusted and area-mix-adjusted bases, suggesting that even less aggressive forms of managed care can yield savings. However, the IHA data did not allow for a direct comparison of per capita costs between ACO and HMO plans, on risk-adjusted and area-mix-adjusted bases.
Patient out-of-pocket expenses and quality
Lower costs are nice, of course, but only if costs are not simply shifted to patients, and only if the quality of care remains high. The IHA data suggests that managed care plans may be achieving all of these outcomes.
The IHA data measures how much of healthcare expenditures are paid by health plans versus patients. The plan-paid percentage of expenditures is higher under HMO plans than under PPO/FFS plans. In 2015, the only year for which this data is available, HMO plans paid 92% of healthcare expenses and members paid 8%. In contrast, PPO/FFS plans paid only 82% and members paid 18%.
To help measure quality, the IHA researchers collected 10 Healthcare Effectiveness Data and Information Set (HEDIS) clinical quality measures for specific health conditions, and created an aggregate measure, called the Clinical Quality Composite. The Composite measure suggests that the quality of care in HMO plans is higher than the quality of care in PPO/FFS plans. That conclusion is also supported by differences in the risk-adjusted readmission rates, which are slightly lower for HMOs in 2015, the first year that IHA published this statistic.
Conclusions
While managed care plans might not be the perfect solution for every person and in every area of the state, they are a valuable part of California's long-term solution to providing high-quality care at affordable prices. Using the 2015 IHA data, if managed care plans (as represented by HMOs) had not existed, the per capita healthcare expenditures among commercial health plan enrollees would have been approximately 5% higher in 2015, totaling approximately $3 billion more in statewide healthcare expenditures. And that is just for commercial plan members, whose costs comprise approximately one-half of the statewide total healthcare expenditures. The IHA data for the Medicare population suggests that their costs would also be higher without managed care plans. Comparable data is not available yet from IHA on the Medicaid population, but it would likely tell a similar story.
Some good news for Californians is that they would likely be paying a lot more without managed care plans that use the delegated model. In brief, the term "delegated model" describes a health insurance plan where financial risk for healthcare services is transferred from an insurance company to healthcare providers (e.g., physicians or hospitals). Most commonly this involves the insurance company paying a fixed, per capita dollar amount (a capitation rate) to a group of physicians, and the physicians assume financial responsibility to provide all professional services for each health plan member. They may also have full or partial risk for hospital services provided to those same members. In California, capitation can only be used in health maintenance organization (HMO) plans. Other common types of plans, PPO-style plans and other fee-for-service (FFS) plans, cannot use capitation.
Measuring the impact of the delegated model on healthcare expenditures is tricky for at least two reasons. First, the average person who enrolls in an HMO plan might have a different health status from the average PPO/FFS plan enrollee. For example, they might be younger, or just healthier than average. Second, per capita healthcare costs vary by geographic area, for a variety of reasons. HMOs tend to be concentrated in urban areas, while PPO/FFS plans are prevalent in all areas of the state.
IHA Atlas data quantifies savings
Fortunately, data published by the Integrated Healthcare Association (IHA) allows us to compare per capita healthcare expenditures for HMO versus PPO/FFS plans, adjusted for differences in the mix of members by health status and by geographic area. Results indicate that for commercial health insurance plans (i.e., non-Medicare, non-Medicaid), total healthcare expenditures per capita are lower under HMO plans than under PPO/FFS plans, as shown in the graph below. They were 5% lower in 2013 and 7% lower in 2015.*
As previously mentioned, providers can also take on varying degrees of financial risk. For example, they might assume risk for just professional services, which is the most common type of capitation arrangement. They might also have a shared risk arrangement for hospital expenses, sharing with the health plan in any differences between actual and budgeted hospital expenses. At the extreme, providers might accept risk for all healthcare services, under what is often called global or full-risk capitation.
Many industry experts, including a substantial number of providers in California, feel that assuming a meaningful degree of financial responsibility for healthcare purchasing decisions is key to ensuring those decisions strike the right balance between fiscal responsibility and providing high-quality care. In fact, one of two major goals stated by the Berkeley Forum was to reduce "the share of healthcare expenditures paid for via fee-for-service from the current 78% to 50%." The delegated model is also consistent with Medicare's goal of shifting more members from fee-for-service payments to value-based payments.
The IHA data includes splits for HMO members covered by plans using professional-only capitation versus plans using global capitation, adjusted for differences in average member risk scores. In aggregate, without geographic area mix adjustment, the data indicates that total healthcare expenditures per capita were 6% lower under HMO plans using global capitation than under HMO plans only using professional services capitation. However, the data volume is relatively low for global capitation members, and heavily skewed toward Southern California. The data did not seem robust enough to provide a reliable comparison of costs under the two plan types after adjusting for differences in the geographic mix of members.
Moving along the spectrum of managed care with global capitation at one end, we find more loosely managed plans at the other end. Such plans do not use capitation, although they often incorporate certain managed care activities, such as large case management and disease management programs. Accountable care organization (ACO) plans tend to fall at this end of the spectrum. On a risk-adjusted basis, the IHA data indicates that total healthcare expenditures per capita were lower for members in ACOs than for members not in ACOs. They were 6% lower in 2015, the only year for which IHA has published this data, on risk-adjusted and area-mix-adjusted bases, suggesting that even less aggressive forms of managed care can yield savings. However, the IHA data did not allow for a direct comparison of per capita costs between ACO and HMO plans, on risk-adjusted and area-mix-adjusted bases.
Patient out-of-pocket expenses and quality
Lower costs are nice, of course, but only if costs are not simply shifted to patients, and only if the quality of care remains high. The IHA data suggests that managed care plans may be achieving all of these outcomes.
The IHA data measures how much of healthcare expenditures are paid by health plans versus patients. The plan-paid percentage of expenditures is higher under HMO plans than under PPO/FFS plans. In 2015, the only year for which this data is available, HMO plans paid 92% of healthcare expenses and members paid 8%. In contrast, PPO/FFS plans paid only 82% and members paid 18%.
To help measure quality, the IHA researchers collected 10 Healthcare Effectiveness Data and Information Set (HEDIS) clinical quality measures for specific health conditions, and created an aggregate measure, called the Clinical Quality Composite. The Composite measure suggests that the quality of care in HMO plans is higher than the quality of care in PPO/FFS plans. That conclusion is also supported by differences in the risk-adjusted readmission rates, which are slightly lower for HMOs in 2015, the first year that IHA published this statistic.
Conclusions
While managed care plans might not be the perfect solution for every person and in every area of the state, they are a valuable part of California's long-term solution to providing high-quality care at affordable prices. Using the 2015 IHA data, if managed care plans (as represented by HMOs) had not existed, the per capita healthcare expenditures among commercial health plan enrollees would have been approximately 5% higher in 2015, totaling approximately $3 billion more in statewide healthcare expenditures. And that is just for commercial plan members, whose costs comprise approximately one-half of the statewide total healthcare expenditures. The IHA data for the Medicare population suggests that their costs would also be higher without managed care plans. Comparable data is not available yet from IHA on the Medicaid population, but it would likely tell a similar story.
*The IHA's March 1, 2018, press release posted on their website cited a 9% difference, which may have not been adjusted for the differences in the mix of members by geographic area. After making that mix adjustment, we calculated a slightly lower difference, at 7%.