In healthcare, it’s tough to know what’s “a lot” regarding cost. A lot in relation to…what? Can you quantify the cost of a colonoscopy with the cost of a TV as the baseline? What about veterinarian care, is that a closer approximation? The most common industry-wide method of benchmarking utilizes the Medicare Price Relativity Index. But what if you wanted a new way? Could there be alternative benchmarking approaches available? Can we ask more questions without answering them??

Thanks to price transparency data, and how we generally answer our own questions on the blog, the answer is yes. Your new options: conducting a percent of cohort analysis or using the Turquoise Health Price Relativity Index. Ohhh, ahhh.

But first, a moment for the OG

Typically, if you want to have a comparison point for how much an item or service will cost across different payers, you’d benchmark Medicaid or commercial rates against Medicare rates. This is referred to as percent of Medicare.

Say, for example, you’re looking at the cost of an (CPT 47652) in the state of Washington. If you did this in Search, you’d see that, for this code, it’s 186%.

Dreaming up some new ways

When we initially created Analyze, our data visualization module within the platform, we enabled users to generate reports based on IPPS and OPPS (Inpatient/Outpatient Prospective Payment System, ie, how Medicare prices claims) Medicare Reference Pricing.

However, Analyze itself had a few functionality limits we wanted to improve:

  1. You can only benchmark rates for CPTs/HCPCS and MS-DRGs
  2. Only hospitals are available for inclusion in reports
  3. IPPS and OPPS pricing only pertains to fee-for-service case rates or negotiated rate amounts making the benchmark of other reimbursement methodologies like percent of charge or per diem reimbursement very difficult

Since price transparency lets you establish baseline relativity between organizations at a high-level service category or individual code level, we started dreaming up new ways one might benchmark rates within the platform.

Medicare as the baseline

The graph below is an example percent of Medicare report. To generate this report, we took the payer-specific negotiated rate and the Medicare rate for each hospital and generated benchmarks as percent of Medicare.

Example percent of Medicare benchmark report.

For this analysis, the Medicare rate used isn’t a single rate for all providers, but instead one specific to each provider (that’s the “wage-adjusted” part). This helps control for differences in local employee costs, for example, but it also makes true comparison harder because both the numerator and the denominator are different for any two hospitals. As we said above, this also limits the codes or hospitals you can include as comparison points.

Percent of Cohort Analysis

Now, we allow for a “cohort average” baseline. To do this, we divide the negotiated rate for each hospital (aka the numerator) by the average rate across all hospitals selected for the comparison (aka the denominator). This makes the denominator the same for all hospitals and creates a new way to think about what’s high/low relative to whatever you might choose as your baseline. This new method also enables you to compare across different code types representing the same setting of care.

For example, MS-DRGs and APR-DRGs communicate care in an inpatient setting. There is inherently no crosswalk between these code sets; however,  two entities included in a report may be reimbursed by both DRG types.

Example percent of cohort average benchmark report

Additionally, you can now view the raw negotiated rates (i.e. the numerator used to generate the benchmark results in the report above) and thus compare to whatever your heart desires. That would look something like the graph below. What would you compare to? Your own rate, perhaps?

Use the raw negotiated rate to compare anything you like.

Or you could generate scores

As an alternative to benchmarking plans, rates, or contracts against Medicare or utilizing generalized discount databases (typically done when shopping for employee health plans), we have developed the Price Relativity Index (PRI) a Uniform Data System (UDS) alternative.

PRI is based on payer-reported price transparency data for commercial plans. Thanks to price transparency regulations, payers report their negotiated rates monthly, allowing the PRI to be up-to-date and provider-specific. As opposed to a traditional discount analysis, the relativity scores are driven by the unit costs of contract-specific services, allowing for a comparison focused on the true cost of care.

This creates an opportunity for us to compare the relative “strengths” of contracts (a.k.a. lowest rates). For every provider, each plan is assigned a relativity score, with the average of all scores being 1. Relativity scores, driven by unit costs of specified services, enable a more precise comparison focused on the true cost of care. Essentially, the higher the relativity index, the more expensive the services.

Imagine a hypothetical hospital: Turquoise Hospital (TQH) contracted with the Employer Solutions Health Plan (ESHP). Their contract only has negotiated rates for the following three outpatient services:

Service

Negotiated Rate

National Median Rate

Relative Score

Utilization

A

$1000

$1000

1.00

100

B

$750

$500

1.50

200

C

$200

$100

2.00

100

If we were to combine the relative scores for each rate, while factoring in utilization, we get a weighted average relative score of 1.50. In other words, outpatient ESHP rates at TQH would be assigned a Price Relativity Index of 1.50.

Let’s see this in action, shall we?

PRI in Chicago

Let’s focus on three academic flagship hospitals and one advocate system in Chicago for sake of comparison (inpatient only). We can see that Blue Advantage HMO consistently has the lowest rates of the three plans. Overall, it seems like Aetna HMO generally has the highest rates. Loyola has the most expensive care, with their Aetna HMO IPPS contract having the most expensive services overall. While Northshore has low rates for Aetna/BCBS, their United rates are quite high (almost as much as Loyola!)

PRI in Chicago

Depending on what you’re looking to do and who you are, you might choose Blue Advantage HMO given their high favorability or maybe your employee population is closest to Advocate Condell Medical Center, so you’d be happy to choose UHC given the favorability for that hospital. The data is your oyster!

PRI in Texas

PRI in TX

Looking at hospitals in Texas, Texas Health Arlington Memorial and USMD Hospital at Arlington are using the same Blue Choice PPO and UHC National PPO rates. Dallas Regional Medical Center has the most favorable outpatient rates with UHC National PPO of the cohort. The most unfavorable contract is Blue Choice PPO with William P. Clements Jr. Blue Choice PPO has the highest rates across the board, whereas Aetna HMO has the lowest Dallas regional, and overall, has the most favorable OP rates. As with our Chicago example, you’d have this benchmark analysis to drive decision-making no matter what items or services you value most.

With price transparency data, we can have multiple baselines

The drastically increased visibility into the cost of care offers us a gift: breadth of baselines. While Medicare rates are set annually by CMS, price transparency data offers monthly and yearly updated commercial payer data from both hospitals and payers. We are no longer stuck looking at the cost of healthcare through a lens primarily focused on Medicare if we don’t want to. If your organization wants a fuller picture of cost, why not go into negotiations having looked at the state of affairs from three different angles? Or pick a combination of the two you feel accounts for what factors matter most to your organization. And next time someone says, “How much do you think ____ costs?” You’ll have more than one way to figure it out.