Using home health cost report data and business intelligence, this post will explore how some agencies are successfully dealing with decreasing profit margins.
In previous blog posts, we have looked at HHA cost report data as CMS does, as an aggregate of all agencies. In this post, we will begin to break these HHAs into groups that reveal portions of these agencies that are particularly vulnerable to decreasing profit margins or have implemented successful strategies for dealing with the issue.
Enterprise HHAs
We last discussed the creation of a new way to identify connected HHAs using cost report balance sheet items. I designated these entities as “enterprise” HHAs in that these individual agencies are all connected by the same management structure even though they submit individual cost reports by CCN. The goal of this effort was to create identities for the HHAs consistent with how they are viewed by themselves and other HHAs in the industry.
Using this definition of HHAs, there are 249 such enterprise organizations I discovered for 2022. Each has multiple cost reports for agencies connected by the criteria I established in the previous post. These enterprise organizations represent 30% of all HHAs by CCN and 44% of the total census for 2022. The remainder are agencies that I could not link with other agencies using this criteria.
In addition, I divided the enterprise organizations into two groups. The first group is the top 10 enterprise organizations by the collective census of all the connected agencies. The second group is the remaining 239 enterprise organizations. It turns out that this organization of the data creates three groups of significant size that can be measured separately as we did for the entire industry. Here is the distribution of the HHA census for 2022 by these categories:
Using the cost report data to calculate profit margins under Medicare Part A, MA and net Medicare, we established averages for all agencies in the industry for 2022. Now, we are going to explore these same KPIs for these enterprise groups, starting with Medicare Part A profit margins. This is 7.77% on average for all agencies collectively using Medicare Part A revenue and total operating expenses. Here is how it looks when we break it down by enterprise categories:
If you are looking for insights on what is going on inside the home health industry, this is the kind of business intelligence results you are looking for. We see significant differences in the 7.77% margins when we view the data through the prism of the enterprise. Inside these numbers may be insights on how to deal with decreasing profit margins.
The two questions that leap out to me are:
How do the top 10 enterprise organizations do so well?
Why do the remaining enterprise organizations do so poorly?
One of the myths I hear regularly about these larger HHA organizations, perpetuated by CMS, is that they have better margins by driving down visits. Many times, this statement is intended to portray these organizations as sacrificing quality of care for greater profits. So let’s start by taking a look at Medicare Part A visits by census for each of these enterprise categories for 2022:
What we see is that all three groups match the industry average for Medicare visits per census. This makes this chart even more interesting than the last one. How are the top 10 groups so much more profitable when their visits per census are the same as the other enterprise groups? Why aren’t the other groups equally profitable?
Finding the Missing Key
At this point, exploring this data is similar to looking for your missing car keys. First, you look in the most obvious spots, then, before you spend a lot of fruitless time looking, you try to limit the search area by thinking about the places they would most likely be and those locations where they are unlikely to be found.
A profit margin is the gap between revenue and expenses. In order for this value to change, either revenue, expenses or both must change. I have done quite a bit of previous research on the revenue side of PDGM. When PDGM started, the original CMS behavioral adjustment calculated for 2020 in the proposed rule was based on CMS expectations that HHAs would alter their behavior to increase revenue. In particular, by decreasing LUPAs and altering diagnosis code sequences to associate claims with higher paying PDGM groups. I remember when I first read this in the proposed rule, I found the projections very hard to believe. CMS estimated that these behavior changes would increase payments by 8.7% and like each year since, they reduced this in half when the 2020 final rule was implemented.
At the time (2019), I had created a model with Sisense that could simulate PDGM payments with existing claim data. I participated in an experiment using hundreds of coded charts trying to see how much I could increase the CMW by re-sequencing DX codes as the primary DX. Even taking every opportunity to increase reimbursement, I could only get something around a 5% increase in revenue through this method. In the description of the 2020 proposed rule, this original BA from CMS assumed that nearly all agencies would participate in these actions and that they would take these opportunities to alter reimbursement a significant portion of the time.
I can’t remember the exact ratios, but I remember thinking that whoever made these projections had very little understanding of how HHAs actually conduct billing and coding and the degree that agencies fear the certain audits that these processes would generate when detected through simple claim analysis by the RAC auditors.
Looking back, these revenue enhancing outcomes never happened. In fact, LUPA rates increased and CMW did not. I believe that this predictive analytics error by CMS is directly responsible for the switch from this original BA to the BA we see today, defined by the comparison of PPS and PDGM payment rates using claims from the current PDGM era.
By changing the definition of the BA, CMS was able to eliminate the financial liability they had created through their first false assumption and participate in the savings created by agencies when visits began to decline in 2020 and 2021.
One thing to understand about PDGM and the other cost-based payment models used by CMS, they are designed to allow CMS to regulate the increase in spending to providers when healthcare costs are increasing. They do this by using cost report data from two years before the proposed rule and then they control the estimated difference between those expenses and the estimated expenses for the proposed year. It is through this estimate that they are able to control spending while maintaining the survivability, in theory, of these provider organizations by the healthcare sector. Although these models work great to control increasing costs, they do not work as well when costs are declining. Whatever you think about the new behavioral adjustment as an HHA, it was a stroke of creative genius from the perspective of CMS.
In one move, it wiped out the previous erroneous behavioral adjustment making it irrelevant. When the 2020 PDGM claims were reduced by the original BA, this was included in the comparison between the revenue from PDGM compared to PPS for 2020. The industry visit reduction savings (PPS payments compared to PDGM) more than made up for the money CMS withheld for the original BA. This liability that would have to have been paid back to HHAs was wiped clean and replaced with the permanent and temporary adjustments calculated retroactively for 2020 and then again for each subsequent year of PDGM.
Another feature of PDGM that was designed in by CMS was the ability to narrow the profit pipeline for agencies making those HHAs with larger profits get less money while increasing the revenue for lower margin agencies. I was able to confirm this with my model in 2019. At the time, I found that PDGM would reduce profits for the top 25% highest margin agencies who had optimized their revenue cycle under PPS and distribute this revenue to the lower 75%. I referred to this as the robin hood effect.
CMS accomplishes this by making the PDGM payment formula nearly invulnerable to revenue increasing tactics. Even the two they believed would happen in 2020, never came to pass.
This long explanation is to back up my current assumption that the increase in profit margins for the top 10 enterprises is probably not related to their ability to increase their PDGM revenue, but has to be associated with a lower ratio of expenses to this revenue. In other words, I don’t have to look for my car keys in the revenue room because they probably aren’t there. However, if I can’t find them anywhere else, I will have to come back and look.
The operating costs can be divided into two subgroups, the direct costs associated with visiting patients in the home and all other business operating costs. If we can do this, we can search these areas separately for the data that will give us the answer we are looking for.
In the 2024 final rule, CMS discusses “margins” that they build from the direct costs of patient care from the cost reports. I discussed these margins and the faulty assumptions they were built on in a previous post. CMS goes into detailed explanations regarding the expenses that are included in the direct costs. They include the salaries of the clinicians performing visits, transportation costs, medical supplies, and even liability insurance.
I need to do a similar drill down on operating expenses, which include these direct costs, in order to create separate search areas for my missing data. For my purpose, I do not need to measure these costs the same way that CMS does, I just need to separate the costs associated with patient care in the home by these clinicians and the total operating costs to create a ratio comparing these direct costs with all operating costs reported by the HHA in the cost report.
At this point, I am going to make another assumption based on basic economics. These HHAs have to pay their staff somewhat similar wages as their local competitors in order to keep them. This means that it is unlikely that our top 10 group could have achieved margins under Medicare over four times greater than the industry average by paying their clinicians significantly less than they could make working elsewhere.
To calculate direct costs, I will add up all clinician salaries and include medical supplies. I will use this value to create a new KPI (Key Performance Indicator) I will call “Operational Efficiency”. Operational Efficiency is the ratio of these direct costs to total operating expenses. This KPI is intended to show not the relative wages paid to clinicians by HHAs, but how much is spent on all other operating expenses in comparison. This is what this looks like for 2022 for our enterprise groups:
It looks like we have found our car keys. It appears that the success of these groups is directly related not to reducing visits made to patients, but to their efforts to reduce all operational costs not directly related to patient care.
These costs are essential to running an HHA, they include management, billing, coding, facilities, software, utilities, advertising, etc. All the expenses you must have to be a business. It appears that being connected does not necessarily guarantee success (Other Enterprises), but that some organizations are significantly more operationally efficient than others.
It is possible to drill down further and look at individual expenses related to the total to see what might separate these groups from each other, but I suspect that these specific strategies might vary with each organization, but the overall tactic of focusing on general expense reduction instead of visit reduction seems to be very successful. Not only does it help with Medicare, but all profit margins from all health plans.
Here are the 2022 profit margins for the enterprise groups under Other insurance in the cost reports. As in previous posts, I will refer to this as MA since it is nearly all of it.
Here are the visits per census for these groups:
This provides other interesting insights. First, the visit reductions by these groups for the top 10 enterprises and other enterprises are identical, but because of extreme operating efficiency differences, the impact is very different. For HHAs not connected (No Enterprise), we see visits about 20% higher than the national average, but better MA margins than the national average. This is a clue toward another interesting KPI specific to MA that I will explore in the next post.
Now let’s look at Medicare Part A and MA (Part C) combined or Net Medicare.
What we see is that the top 10 enterprises are the only profitable organizations, as a group, under Medicare Part A and Part C combined. Other enterprises perform significantly worse than the rest of the industry.
I find this chart very interesting because we can assume that above average performance by the top 10 is not limited to just the top 10. Number 11 should show similar results, but somewhere down the list these connected agencies’ operational efficiency deteriorates. It is almost as if many of these connected HHAs were not actually trying to make a profit since the unconnected agencies have a profit margin 150% better than these agencies. Again, this is not related to visits. Here are the net visits for these groups:
Using this KPI our Other Enterprise group is the best performer, but operational inefficiency wipes out these gains. The most interesting thing about these last two charts is that the group with the highest net visits is the “No Enterprise” group, yet their net profit margins outperform the national average. We will explore this in my next post.
Many times when I share this kind of business intelligence data with HHA leaders, I get the comment that “this is interesting, but how does this benefit me? What actions can I take based on this information?”
Many times they are correct, but in this case, I think the KPI of operational efficiency can be the key to survival in the hard times we are experiencing now and the even harder times to come. We can take a lesson from these efficient top 10 organizations and learn how to improve our own. This can be done with a simple exercise, take your own current financials and calculate your direct costs as I did, total clinician salaries and medical supplies. Take your total operating expenses and divide this into the direct costs.
The gold standard for this KPI is 40%. If you can get there, you can not only survive, but you should make a double digit profit including all operating expenses. Even if you fall short, you can substantially improve your financial position.
These cost reductions will not be easy. I can tell you from personal experience. In 2010, the bottom fell out in the real estate industry affecting the entire economy. As a software company, it impacted me as well and I did not see it coming. My banks called in my loans and I had to make these hard decisions to survive. Many of the costs I felt were absolutely necessary to survive turned out not to be. In the end, these painful choices I made then improved my organization and when the market improved, our margins were better than ever. As my organization grew again, I made better decisions on how to spend my money.
Healthcare cannot survive without care in the home. There is no way to predict how CMS and MA plans will adjust to these realities, but they will have to when the extinction of many of these agencies becomes a reality. Another KPI I have referenced in previous posts is the number of agencies reporting a negative net income. Here is what this KPI looks like by the enterprise group:
None of the groups are immune to this danger, even with positive net margins overall, our top 10 group has 25% of their individual agencies under water. The most interesting part of this chart is that the Other Enterprises match the industry average of negative agency margins even with their extremely poor operational efficiency.
Successful business intelligence data explorations, like this one, not only answer the questions you might be researching, but create indicators of other areas you might want to explore.
When looking for your car keys, you might find some money in the couch cushions. This might indicate that you might want to explore all of them in the home for additional change. In the next post, we will explore some of the questions created during this exercise with BI.
Comments