We show how to analyze machine-readable files for a competitive edge.

What if everyone played Texas Hold ’em with all their cards showing? This would seem to even the odds—but not if the cards were very, very difficult to read. That’s the scenario set up by the price transparency mandate from the U.S.’s Centers for Medicare & Medicaid Services (CMS). Through our work with industry-leading businesses and extensive analysis of MRF data, we’ve developed a series of best practices to tackle this thorny challenge.

Transparency requires commercial health insurers to publish a machine-readable file (MRF) of their negotiated network prices every month. The data is available to CMS, other payers, employers, providers, the media, academia, the public, etc. And there’s a lot of it to analyze, hundreds of terabytes (TB) a month, revealing prices for all covered items and services between the health plan issuer and in-network providers, plus allowed amounts and billed charges from out-of-network providers.

MRF data is filled with business insights if payers know what to look for and where. Approaching MRF data with a clear business strategy will help payers identify the data that will provide competitive insights to use in assembling networks and negotiating contracts; developing strategies for mitigating the potential public relations impact of MRF data; and making MRF compliance more cost effective.

Building a winning hand

Payers’ main challenges with analyzing competitors’ MRF data are its volume and complexity. Based on our work for a client, a month’s worth of MRFs for just five states and 16 payers generated 1.5 TB of information. Further, two completely compliant MRFs nonetheless may be organized differently and require extensive normalization. Business objectives should guide analysis priorities.

A Cognizant client in highly competitive markets wanted to understand its competitors’ pricing and networks in target states. We analyzed data for our client’s (and top competitors’) largest markets and highest volume services in those states. This approach narrowed the scope, keeping the initial analysis manageable. While the MRF can provide very granular data for individual National Provider Identifiers (NPI), payers need to focus on insights likely to have the greatest impact on their business. Those typically come from high-volume, high-cost services and from understanding how competitors have structured rates and networks.

Within those categories, payers should prioritize capturing insights in these areas:

1.    Pricing/rate variations

MRF data analysis offers unprecedented insight into market positioning and outliers. It’s now possible to examine rates for a single CPT code by county and/or NPI. We recommended that one client focus price optimization efforts on a large county in which 20% of its NPIs were contracted for a high value code. State-wide, our analysis found the client’s average reimbursement rate for a procedure was 28% higher than that of its lowest competitor. Drilling deeper into a single county in that same state, we identified that one specific provider contract, which was 21% higher on average, was largely driving this price difference. Analyzing this provider contract, we identified that the pricing difference was driven by much higher nurse practitioner rates. Insights like these will help our client optimize its pricing for that procedure.

2.    Provider contract complexity

In our analysis of MRF data across 13 states, we are finding vast disparities in rates for the same procedure, sometimes with the same physician, with no clear rationale for the variances. Our analysis of one CPT code showed our client had nearly six times as many rates and more than nine times as many NPIs as a much larger competitor. These findings raised the issue of whether our client’s contracts were overly complex; each rate requires business rules, adjudication, etc., all of which increase operations costs.

3.    MRF compliance issues

Comparing and analyzing MRF files can reveal possible problems with how a payer is generating MRF data. In one MRF, we found the same provider had 92 rates across four of our client’s health plans. Another MRF contained 2,000 providers, each of which had 35 rates for a single CPT code. These variances could be the result of overly complex contracts. Another possibility is that our client’s MRF processes were generating incorrect file data. Payers can investigate such anomalies and optimize their MRF generation processes.

4.    Network insights

Payers may use MRF data to compare their networks’ adequacy to those of competitors, evaluating such factors as whether they are contracted with enough providers for the populations served. The data can also reveal differences in how payers create networks. MRF data showed one client’s competitor reimbursed providers 20% less for a specific code. Further investigation showed the competitor exclusively contracted with nurse practitioners for the code, whereas our client uses a mix of providers. Correlating claims volume and utilization data with the code will help our client determine how and whether to change its contracting for that procedure.

5.    Payment methodologies

MRF data should bring insights around capitated per-member-per-month fees and bundled services. Few of the MRFs we have analyzed have included this data so far, despite the adoption of bundling and capitation arrangements in those markets, suggesting the industry is just starting down the path toward price transparency maturity. As this data becomes available, it will provide intelligence about how payers structure value-based reimbursements.

6.    Potential PR issues

Published commercial rates are trending at 150% to 300% of Medicare rates. This information, and cost variances for the same procedures at different hospitals in the same area, tends to generate increased scrutiny of the industry’s costs and rate disparities. MRF analysis reveals how a payer’s contracted rates compare to Medicare reimbursements and may also show influences on those rates, such as geographic location or procedure type. Payers should be alert to and have explanations for data that employers, the media, providers and others may question.

When to place bets

MRF compliance is in flux. Payers are still interpreting guidelines and making changes to how they construct their data files. Our experience shows that MRF data quality is improving, though the MRF schema permits a variety of formats that payers should reconcile before making comparisons. Before making business decisions based on MRF data, payers should take the following steps to ensure they are reading their competitors’ cards correctly:

• Understand the trustworthiness of data. We examined rates for an in-patient bladder procedure from several payers in a northeastern state. One large payer had a much lower rate than its four competitors. Digging into the MRF data showed the large payer’s MRF file reported telehealth and other codes, suggesting this was an in-home or remote procedure. The other four showed in-patient procedure codes. The large payer’s file was erroneous. When unusual discrepancies occur, be skeptical about accepting the initial finding.
  
  
• Tap additional sources to generate insights from raw MRF data. Applying national cost benchmarks and internal data, such as claims volume and utilization, can help payers identify codes, NPIs, markets and plans to analyze. NPIs with disparate rates that only generate a few claims against a specific code may not be worth addressing, at least initially.
  
  
• Ensure compliance without oversharing. Some payers may reveal more than the transparency rule mandates. We found many payers organized data by their health plans while one of our clients had included data segmented by its contracts. Both approaches are compliant, yet our client was potentially revealing more data than necessary.
  
  
• Establish capabilities to ingest and analyze MRF data. The massive volume of MRF data makes it difficult for desktop programs to analyze. A variety of vendors have software platforms for ingesting and normalizing MRFs; this may be the prudent near-term option to gain immediate insights while MRF compliance matures. Custom solutions designed to parse MRF data will need continual adjustment as MRFs mature and payer approaches change.

Many payers are launching competitive analytics initiatives. Payers slow to analyze how their data compares to competitors’ and benchmarks will find themselves ill-equipped to answer employer, provider and public questions about their rates and practices. They also will lose opportunities to make proactive changes and improve their competitive position in key markets. By focusing MRF analysis on key markets and competitors plus internal data, payers can quickly gain and apply useful insights, in effect dealing themselves new cards and a better hand.

To learn more, visit the Healthcare section of our website or contact us.

This article was written by Derek Spearing, Senior Director of Consulting in Cognizant’s Health Sciences practice.