Since its first publication in 2005, the Milliman Medical Index (MMI) has proven a valuable measure of average healthcare costs and changes in those costs for a hypothetical "typical American family of four." In prior years we have defined that family as a male age 47, a female age 37, a child age 4, and a child under age 1. In reality, family compositions vary, and families can have extremely different levels of healthcare expenses. This variation results from differences in family size, the family members’ ages and genders, where they live, their income levels, their unique health conditions, and a host of other variables. The MMI interactive tool allows you to build your own family based on several demographic options and better understand the associated healthcare costs.

The interactive tool allows the user to understand how costs are expected to vary for different types of families with respect to size, age and gender mix, and different cost levels representative of variation by area. While this does not capture all cost variations by individual, such as those driven by individual health conditions, it captures significant features that drive expected average costs.

In the interactive tool, the user can set the following inputs:

• Geographic area: Choose from among all metropolitan statistical areas (MSAs) in the country. Each MSA has been classified as either low-cost, mid-cost, or high-cost, and the healthcare costs displayed will reflect those levels. Healthcare costs in the low-cost MSAs are set at approximately the 20th percentile of all MSAs, mid-cost areas are the average, and high-cost areas are at the 80th percentile.
• Choose the “MMI family” or a custom family
  • If you choose the MMI family and a mid-cost area like Lincoln, Nebraska, you will see the same resulting healthcare costs as in the MMI report, with the total dollars displayed by type of service in total.
  • Custom family: You can create your own family with up to six family members, and see the same outputs as for the MMI family. Demographic specifications include:
    • Average person
    • Average adult
    • Average child
    • Child age 0-1
    • Child age 2 and older
    • Adult female or adult male age 19-25
    • Adult female or adult male age 26-44
    • Adult female or adult male age 45-64

Technical appendix

The MMI is made possible through Milliman’s ongoing research on healthcare costs. It is derived from Milliman’s flagship health cost research tool, the Health Cost Guidelines™, as well as a variety of other Milliman and industry data sources, including Milliman’s MidMarket Survey.

The MMI portrays the projected total cost of medical care for an average person, and for a hypothetical family of four (two adults and two children), covered under an average employer-sponsored PPO health benefit program. The MMI reflects the following:

• Nationwide average provider fee levels negotiated by insurance companies and preferred provider networks
• Average PPO benefit levels offered under employer-sponsored health benefit programs
• Utilization levels representative of the average for people covered by large employer group health benefit plans in the United States

The ACA introduced the concept of “metallic tiers” for benefit plans starting in 2014. Individual and small group policies must have a metallic tier level of “bronze” or higher (silver, gold, and platinum). Bronze implies that, on average, the plan will pay 60% of the costs for the essential health benefits (EHBs) that must be provided by the benefit plan. To help avoid penalties, larger employers must provide plans that, on average, pay at least 60% of the cost of covered services, a threshold deemed “minimum value.” The MMI plan has an actuarial value of approximately 84.0% in 2021.

This year, we changed our area factor methodology from a population composition approach to a morbidity-adjusted approach. Estimated costs in the interactive tool are for a population with an "average" risk, which allows the user to adjust the cost based on the health status level. Comparisons to prior year tool outputs may not be appropriate as a result of this new methodology.

Variation in costs

While the MMI measures costs for an average person, and for a hypothetical family of four, any particular family or individual could have significantly different costs. Variables that affect costs include:

• Age and gender. There is wide variation in costs by age, with older people generally having higher average costs than younger people. Variation also exists by gender. Our MMI-illustrated family of four consists of a male age 47, a female age 37, a child age 4, and a child under age 1. This mix allows for demonstration of the range of services utilized by adult men, adult women, and children. Average utilization and costs of specific services will be different for other demographic groups.
• Individual health status. Tremendous variation also results from health status differences. People with severe or chronic conditions are likely to have much higher average healthcare costs than people without these conditions.
• Geographic area. Significant variation exists among healthcare costs by geographic area because of differences in healthcare provider practice patterns and average costs for the same services. For example, the relative cost of living affects healthcare costs, as labor costs (e.g., nurses and technicians) tend to be higher in areas where the cost of living is higher. Access to advanced technology also affects the utilization of services by geographic area.
• Provider variation. The cost of healthcare depends on the specific providers used. Even in the same city, costs for the same service can vary dramatically from one provider to another. The cost variation results from differences in billed charge levels, discounted payment rates that payers have negotiated, and implementation of payment methodologies that may influence utilization rates, such as capitation or case rates.
• Insurance coverage. The presence of insurance coverage and the amount of required out-of-pocket cost sharing also affects healthcare spending. With all other variables being equal, richer benefit plans usually have higher utilization rates and costs than leaner plans.

As described earlier, the interactive tool captures some of these variations, namely those driven by age and gender, as well as ranges representative of geographic cost differences.

Acknowledgments

Thanks to Phil Ellenberg and Kelsie Gosser for their work building the MMI interactive tool.