The Kaiser Family Foundation Employer Benefits survey underlying data, trended from 2018, with 2020 data + proprietary employer data.

The version and most recent effective date are always listed below the percentile gauges.

Once per year for Kaiser data; employer data is loaded periodically. Latest version is effective April 2021.

2020 updates are summarized here

Plan value approximates actuarial value, or, on average, the percentage of in-network claims paid by insurance. Values range between 60% to over 90% and are often referred to by metal levels. Bronze is 60%, Silver 70%, Gold 80%, and Platinum 90% +/- 2%. Healthier people will have lower claims and pay a greater percentage, while those with higher claims will pay a lower percentage of the value.

A multiple regression model run using over 100 distinct plan designs and the output actuarial value from the CMS actuarial value calculator–it was found that ~85% of the actuarial value of a plan is determined by 3 factors: deductible, coinsurance, and out-of-pocket maximum. Actual results are within 1-2% for most plan structures. 

Plan value, or metal value, under the “summary of benefits” area is an absolute number, and an estimate of what the plan pays on average. The plan value gauge below the chart on the right is a value relative to the cohort selected. For example, 58% means the firm or group in question has a more generous plan design than 58% of comparison firms. 

An atypical plan design like $0 deductible and $7,000 out-of-pocket max may have a skewed plan value number. If you use outlier values like $15,000 deductibles you will not get an error. Results should end up being between 60-90% for virtually all comparable plans. 

The ranges in the charts are the dollar values that reflect the middle 50% (the interquartile range, or the 25th to 75th percentiles) as well as the top and bottom 10%. If you hover you’ll over each chart you’ll see dollars and percentiles as well as the comparison firm in question. 

The color-coded gauges reflect the percentile ranking of the modeled firm relative to the cohort selected. Green means the top third most attractive (most competitive) or 67th to 100th percentiles; yellow is the middle third or 34th to 66th percentile; and red means the one-third least competitive or 0 to 33rd percentile. 

The data for individual firms reflect the average numbers if there are multiple plans–so if half the employees are enrolled in a $1,000 deductible plan and half are in a $3,000 deductible plan, the number represented for comparison purposes in the charts is $2,000. Firms typically offer 1-3 plans to their employees. Future versions of the tool may have a breakdown by plan types and/or metal levels.

Since single coverage is the most common type of coverage used when running the benchmarking exercise, those values are used. Family deductibles and OOP max are most often 2x single.

Yes, it reflects an aggregate deductible, which is the case for 90% of plans. 

Any firm contributions to HSAs or similar accounts, for modeling, should be netted out of what an employee pays via payroll “Net annual employee payroll.” For example, if an employee with single coverage pays $2,000 per year for health insurance coverage (or $83 per pay period), but is given $500 in HSA contributions from the firm, the net amount to input would be $1,500. Employee contributions to HSA accounts should not be used in the modeling exercise as no net dollars are impacted.

There is clustering around certain attributes, such as coinsurance. Many plans have 20% coinsurance.

Some healthcare plans have a mix of copays and coinsurance. A plan with a $25 office visit copay is approximately a 20% copay given typical charges. If there are scenarios where office visits are free and hospitalizations are 20%, you can use 20% as the coinsurance or 15-19%, since the bulk of actuarial value hinges on higher dollar categories 

The four regions used in the tool are West, South, Midwest, and Northeast, and are based on the BLS classifications of regions.