In this post, we will begin a transition from analyzing cost report data to analyzing another CMS data source, the data collected through the Home Health Quality Reporting Program (HH QRP).
This data is the basis for an upcoming modification to the CMS home health revenue cycle process referred to as value-based payments. More importantly, the KPIs from these data sources can be used to evaluate the relative performance of HHAs when choices are made by patients, providers and health plans to engage with them.
This quality reporting data has been collected by CMS for the last 5-10 years for all healthcare provider types. It is part of an overall strategy developed by their own internal business intelligence organization they refer to as the CMS Innovation Center. This department has driven many new advancements within the CMS revenue cycle process for all providers. They are responsible for developing new strategies to transform the healthcare system to the degree that it is influenced by CMS. Historically, these new payment models and strategies eventually influence the revenue cycle processes of other health plans, including the Medicaids.
In this post, we will begin to examine this HH QRP data as we did with cost report data in my previous posts. We will start with something simple, patient survey star ratings. CMS collects survey data from patients This data is published by CMS and is provided to consumers through their Care Compare website. Patients can look up any provider and view their reported quality KPIs before selecting them. These KPIs include star ratings intended to provide a more direct and familiar approach to potential patients interested in the relative quality of care delivered by available HHAs.
The survey results are also published by CMS as spreadsheets. These can be transferred to our Sisense business intelligence tool and linked to the existing cost report data. Like the cost report data, not all the data is usable. In order to qualify for a star rating, an agency must have at least 40 completed surveys during the reporting year. If they do not, they are still included in the data on Care Compare and our spreadsheet, but without a star rating.
The star ratings can be a value from 1 to 5. The data we will be using is the results of surveys for the 2022 calendar year:
In this data, we have 11,684 HHAs with some survey activity, but only 4653 (40%) with enough surveys to have a star rating. This data does not require the complex “cleaning” process we used with the cost reports. We will simply use only the records with a star rating.
In order to be able to calculate averages, I have converted the star values to numbers, 1 - 5. If we exclude the agencies without star ratings, this is how these ratings are distributed to all remaining HHAs:
We can see that the most common rating for an HHA is 4, about 46% of all agencies with a rating. Agencies with a rating of one star are only 2%, 106 agencies for 2022. There are 821 agencies with a five star rating, about 18% of all agencies reporting a rating. For all agencies with a star rating, the average rating is 3.67. This is similar to the distribution of star ratings for other service industries. Yelp reviews for restaurants have an average star rating of 3.77.
Just like the Yelp reviews, these ratings are most often used when a potential customer is interested in engaging with the HHA for the first time and is obtaining data regarding the expected quality of this potential experience. Like the dining experience, someone will probably avoid the HHA if they see a rating of one or two stars. Unlike the dining experience, a rating of 5 stars is ideal since price is not normally a patient or provider consideration when selecting a HHA.
Once someone begins working with an agency, a patient, health plan or a provider partner, they are more likely to judge the quality of the HHA by their own experience rather than the star rating, just like dining out.
In my opinion, this is the true value of these star ratings and the other HH QRP data elements. It is not their role in the calculation of value-based payments, but in their potential to repel and attract patients and partners exploring their available options.
Since this data is organized by the CCN as the agency identifier, like the cost report data, we can link the HH QRP data to the cost report data and see if we can develop insights on how this quality might be related to data in the cost reports like visits, size or operating efficiency, discussed in previous posts.
When we exclude cost reports without a star rating, we see a similar distribution of the star ratings to HHAs with cost reports. Here are the ratings for all HHAs with 2022 cost reports:
The average rating is 3.57. Using our previous measures of agencies by T-shirt size, we can see if these ratings might be related to agency size by total annual census. Here are the average star ratings by agency size:
These star ratings include patients with Medicare Part A and Medicare Advantage so let’s look at net visits (Medicare Part A and MA combined) by census for the same size categories from our previous post:
The cost report data and the star ratings are collected for the same year, but they are collected through two distinct processes unrelated to each other. We can see that there appears to be a direct relationship between the quantity of visits per census and star ratings. In fact, the two bar charts are nearly identical to each other when viewing the relative values of each metric by agency size.
For someone that works with data and business intelligence, these types of insights are very comforting. They are not shocking or unexpected like the relative reimbursement of MA, they support conclusions that many of us might predict without data, if you visit patients more, they will be more satisfied with their care.
Since these two data sources are independent of each other and yet seem to provide very similar and expected results, they each support the validity of both data sources and how the data has been compiled and presented by CMS and then by me through Sisense. This lends credibility to other data compiled through both data sources, in particular, the KPIs I have presented so far from the cost reports.
Visits cost money. Since the beginning of PDGM, HHAs have reduced visits overall until 2022. This appears to be at the expense of patient satisfaction. Is this true? If visits are a factor, are they the only factor or even the primary factor in determining star ratings regarding patient satisfaction?
To explore this, let’s look at our enterprise categories for the 2022 cost report data and the KPIs we explored previously when viewing agencies as the collective agencies they perceive themselves to be rather than individual CCNs assigned by CMS.
Here are the star ratings for the enterprise categories:
Now let’s compare these results to the visits per census for these enterprise categories with MA and Medicare Part A combined:
Now we are seeing some insights with greater value. Although visits appear to influence star ratings when we look at HHAs by size, it is outweighed by the organizational structure of the HHAs. Our top 10 enterprises, leaders in operational efficiency and profit margins, are also leaders when it comes to patient satisfaction, by a wide margin. This is a major myth buster.
Many people believe that visit reductions reduce the quality of care. Our data by agency size supports this position, but these charts show that any negative impact to patient satisfaction from reduced visits can be overcome by a well run organization. In fact, these disadvantages can not only be overcome, these operationally efficient agencies excel at delivering care to satisfied patients. They are like well run restaurants with good food and excellent service at a fair price.
Comments