In this episode, Maria Miranda, Director of Emerging Payment Models and Ely Labovitz, Senior Technology Developer at BESLER discuss technology and data that can help you optimize your performance under episode payment model programs.
Mike Passanante: Hi, this is Mike Passanante. Welcome back to the Hospital Finance Podcast. Today, I am joined by Maria Miranda and Ely Labovitz of Besler. And we are going to talk with you today about utilizing data and technology to manage your EPA programs. Maria and Ely, welcome to the program.
Maria Miranda: Thank you, Mike.
Ely Labovitz: Thank you, Mike.
Mike: So, first off, why don’t you tell us about the data that CMS is providing hospitals to help them contain their episode payment model costs?
Maria: Sure, Mike. So, CMS provides hospitals with claims data for every type of provider that sees the patient during the anchor admission and for the 90 days post-discharge. The information includes patient demographics such as age and gender. But it also contains specific identifying information such as the patient’s account number and the Medicare HIC number.
Additionally, CMS files include detailed information about physicians and about the facilities that treat those patients during the anchor stay and all subsequent care received within the 90-day period after discharge. This allows the provider to track the patient costs, not only in their own facility, but also by post-acute facilities and physicians that treat their patients.
Mike: Now, it seems that CMS is providing a lot of meaningful data to providers around their EPM programs. Can a hospital paint a complete picture of these episodes with only the CMS data?
Ely: So, Mike, at the end of the day, the hospital is missing three components. First and foremost, the data CMS provides is somewhat complex. And you really need a good analytics tool to organize the data, so the results can be meaningful and trends can be established.
Additionally, while CMS does give the provider some comparative regional information, it’s really only in summary form. The provider can see how each service type compares to other facilities in the region—for example, in-patient skilled nursing rehab physician. However, CMS does not provide the claims level detail needed to perform any type of complex analytics.
Let me elaborate on this for a minute, Mike. Using data provided by CMS, a provider may see a trend with a certain post-acute SNF such as a longer than usual average length of stay. It will be useful to see if other providers in the region are seeing the same trend with this nursing facility or other nursing facilities in the area.
Unfortunately, because CMS doesn’t provide this level of detail, it would be impossible for the provider to know if this was an anomaly for their patients, or if it was a normal trend across all patients in this region.
And the final key component missing is data that will allow a provider to have better advanced planning. CMS releases provider data. But until they do, a provider has no way of knowing how far their costs deviate from their target numbers. And once the numbers are released, it may take several years to reverse a negative trend.
Mike: So, what can a hospital do to better prepare for additional mandatory bundles?
Ely: The answer to these problems would be technology solutions to better analyze the data they’re given and obtain data that they just don’t have.
Mike: Okay! So Maria, let me turn to you. What are some of the technology solutions that Besler could recommend to a hospital to help them analyze their CMS data?
Maria: Sure! We utilize Tableau which is a great analytic tool for slicing and dicing data. And there are a lot of different programs out there. We happen to use that. And we really do like that program. This type of tool can really filter your data and graphs by your DRG fracture combination, by year, by physician, by post-acute provider, and even by discharge status. And you can see your average length of stay, your total cost for these categories as well.
To illustrate the power of the tool, and specific to the CJR program, for example, in the CMS-provided dataset, the provider is given the total cost for each type of service—so for in-patient SNF, rehab, et cetera.
Using Tableau, you can see what percentage of your total cost goes into each of these categories. When we bring all these files into Tableau, we can replicate this type of analysis for every operating physician.
We can then filter the data by just one DRG fracture combination if, for example, you’ve identified one of these DRG fracture combinations to be problematic for your facility. I can take the report and draw a line through the bars that then represents the target amount. And it can clearly show you how each physician fares compared to that target, and what percentage of each physician’s total cost per episode goes into each service type.
That’s not something you can do with the data that CMS gives you.
Tableau also allows a user to import other datasets from external sources. Using this type of tool, you can bring in external data that perhaps CMS doesn’t give you, like the star ratings, for example, which can easily be downloaded from HospitalCompare.com.
Once imported, you can then link the new data to the existing CMS data sources and perform further analysis.
Mike: Okay. So, a tool like Tableau and easily obtained external data such as facility star ratings, for instance, help paint a more detailed picture. But that really does not address two really key issues, namely detailed regional comparisons and obtaining data prior to being subjected to penalties.
So, can you tell us what Besler is doing to help better equip providers with the data they need today?
Ely: Sure, Mike. Excellent question!
In the past year, Besler purchased the national dataset of benchmarking data that provides every in-patient, out-patient, SNF, home health, and doctor’s office visit paid by Medicare.
Although this data is identified at the patient level—meaning there’s no names, there’s no addresses—it does contain enough demographic information to allow us to create some meaningful trends. We created regional benchmarks for all acute, post-acute, and out-patient services at the regional and national levels. This allows us to tick the provider-specific EPM data provided to our customers and analyze it at the regional and national levels.
Currently, Besler is using this data in three different ways. We take each metric provided by CMS in the provider-specific files, and compare them at the national regional levels. These metrics are provided to our customers as part of our standard service engagement and provide a detailed view as to how a provider compares to other facilities within the region.
Secondly, we created a score card for every acute facility in America that lists key metrics and will help a provider understand their risk when it comes to Medicare’s EPM programs.
The score card is an excellent predictor of a facility’s total spend in each of the four mandatory EPM programs. It also provides a facility with the quality scores in four distinct areas. And lastly, it provides the data on their total at-risk dollars under the four programs.
Finally, with this data, Besler have developed a nationwide and regional CMI comparison at both the overall facility and DRG levels.
Mike: Maria, can you tell us a little bit more about CMI and how that affects the hospital’s bottom line.
Maria: Sure! So, CMI assigns a unique weight to each DRG, Mike. The weight reflects the average level of resources for an average Medicare patient in that DRG relative to the average level of resources for all Medicare patients.
The weights are intended to account for cost variations between different types of treatments. So more costly conditions are assigned a higher DRG rates, while less costly procedures have a lower weight. A hospital cost mix index or CMI is a relative average of all cases assigned the DRG for a given year, and is a good metric to determine how CMS adjusts DRG payments that to that facility.
Ely: So, Mike, as a follow-up to that, what we’re doing with this CMI data is we’ve created these national benchmark data to calculate the CMI for each DRG combination. So we’re not only doing it at the full facility level, which is a metric that Medicare actually would provide to the hospital, but we’re providing additional data that you can’t get from Medicare, which is a CMI for each DRG combination.
This allows our analysts to determine if a specific diagnosis is being reimbursed at a reasonable rate compared to other facilities within the comparison region. A higher CMI score for a particular group of DRG’s would generally mean that the patients at this facility have a higher level of complication than in other facilities. And of course, Mike, the converse would also be true, resulting in a low reimbursement rate.
Our analysts can help determine whether or not the coding practices at the facility (which is our client) is receiving the optimal level of reimbursement at the DRG level.
Mike: I’ve seen the analyses that your teams have created. And it’s actually pretty fascinating. You can uncover some really interesting things. Can you give us some examples of what you’ve identified using these tools?
Maria: Absolutely! So, basically, when you’re evaluating total costs for these types of programs, it’s really important to seek out variation because removing that variation is an important element in controlling the cost.
So, in one instance, we identified very large variation in post-acute care cost for one of our hospital clients. We noticed that the cost for [discharge status 61] was exceptionally high.
So, it turns out that this facility had some CJR patients going to a critical access hospital after discharge. This critical access hospital offered swing bed services for which they receive Medicare cost reimbursement as opposed to per diem reimbursement. So whenever patients went to that facility, the episodes almost always exceeded the target rate.
In another case, we noticed that readmission rates vary greatly from one year to the next. And by drilling down to the physician level, we identified that one specific physician had a rate of readmission that far exceeded all the other physicians. This physician also had one of the highest volumes for the program. So this one physician had a huge impact on the hospital’s overall program variance.
And then, finally, another example using our graphs by post-acute care provider and bringing in the star ratings as Ely mentioned before, we were able to assure one of our hospital clients that the average length of stay and total cost per episode for SNF services on average was lower when their patients went to SNF facilities that had a higher star rating.
And these are things that you would want to know if your facility was considering which SNF facilities to collaborate with for the CJR program.
Mike: So, this is great stuff. And as you can see, I think we’ve got some smart people and some great technology here at Besler that can provide a lot of clarity and insight around EPM programs.
So, if your hospital is in the CJR program or if you’re in involved in any of the other EPM programs that are mandatory from CMS, we could certainly help you with that and provide some insight.
And right now, we’re also offering a free score card to hospitals that are affected by these programs. And what’s in that score card again? Can you remind me?
Ely: Sure, Mike. The score card will really give a hospital some insight as to where they fall against the target pricing in their region. The score cards are broken up into four different reports.
We provide an overall benchmark as to where the facility rates compare to regional pricing for the bundled programs.
We also provide quality scores based on data that we got from Hospital Compare which actually shows their quality scores within the four EPM programs mandated by Medicare.
The third area of the score card actually shows them the Medicare dollars at risk in their facility per EPM program.
And lastly, we’ll give them a view of what we would call the heat map, which are the individual CMI scores for the DRG’s that make up those four individual EPM programs.
Mike: So, tremendous amount of insight at no cost right upfront. We’d love to talk with you if you’re in one of these programs and would like some assistance and some insights.
And you’d like a score card, please go ahead and email Maria directly at mmiranda@Besler.com. And she’ll be happy to begin that discussion with you.
Ely and Maria, thank you for joining me today on the Hospital Finance Podcast.
Maria: Thank you.
Ely: Thanks, Mike, for having us.