*2019 E&M changes will impact all areas of practice management.*

Far-reaching impacts to work relative value unit (RVU)-based compensation models that will pay providers in full or part are expected if the Centers for Medicare & Medicaid Services’ (CMS’s) proposed evaluation and maintenance (E&M) changes become a final rule.

In my last article, I talked about the financial impact, which will be complicated by the Modifier 25 reduction proposal, as well as the new G codes. Work RVU-based compensation plans will also be affected by the latter two factors, but for a different reason.

For the past 10 years, we have seen significant growth in the number of physicians being compensated, at least in part, based on their volume of work RVUs. These so-called productivity models normally combine some guaranteed base salary and then add some percentage of a total compensation capacity using other metrics. For example, some percentage may be tied to timely documentation or highly compliant coding. But the largest component is normally tied to some increase in work RVUs above some expected minimum range. For some practices, FTE metrics are defined by work RVUs, and under the 2019 proposed E&M model, everything changes.

Calculating the work RVU differences was more difficult than calculating the financial impact, because in the tables provided by CMS, the new RVUs, which combine work and practice expense, were fixed across four different categories. For example, 99201 had an effective RVU of 1.18, while codes 99202 through 99205 had an effective RVU of 3.59. We saw the same thing for established office visits. 99211 had an RVU of 0.66, while 99212 through 99215 shared the same RVU value of 2.47. Unfortunately, CMS did not publish the individual work RVU component, and as such, I had to calculate this by imputing the values from existing ratios.

**Table 1: Comparison and impact of work RVUs for 2018 versus 2019**

In Table 1, I created the work RVUs by first looking at the ratio of work to work plus practice expenses, and then extending that to the 2019 effective RVU value. For example, for 2018, for code 99203, the work RVU is 1.42 and the non-facility practice expense RVU is 1.48, which gives us an effective RVU of 2.90. If I divide the work RVU by the effective RVU, I get a ratio of 0.4897, which means that the work RVU is approximately 48.97 percent of the effective RVU. Next, I multiply that ratio by the 2019 effective RVU for 99203 (3.59), and I get an estimated work RVU of 1.76. This is an estimate, because I don’t know whether CMS is going to stick to a single work RVU for each of the codes in a given category or whether they are going to be based on the time recommendations, as they have been for the past 26 years. If not, then this will represent a statutory departure for creation and modification of the work RVU.

I did this for each code, and arguably, this may not be the method that CMS uses, as discussed above. For example, they may choose to use a fixed value that is the average of the work RVU values for the group. If this were the case, then we might see a work RVU value for 99202-99205 of 1.85 and an average work RVU value for 99212-99215 of 1.19. For this example, I chose to use the imputed values I calculated in the method described above. Finally, I calculated the difference between the 2019 and 2018 values to come up with a delta value for each code.

The next step was similar to what I did with the financial impact; using the 2016 Medicare frequency data, I multiplied the work RVU delta by the total frequency of paid procedures by specialty for codes 99201-99205 and 99211-99215, which gave me the total frequency-adjusted difference. I also added a column for the total paid work RVUs for each of those codes in order to come up with a percentage effect by specialty and for all specialties. Taking 10 specialties in alphabetical order, the table looked like this:

**Table 2: Impact of 2019 proposed changes on work RVUs by Specialty**

In this table, the total work RVU is calculated by multiplying the 2018 work RVU by the total 2016 frequency. The total impact is calculated by multiplying the work RVU delta by that same frequency. And the percentage column is calculated by dividing the total wRVU impact by the total wRVU. In this case, say, for cardiology, we see that if this model were in effect for 2018, assuming the same volume as 2016, all cardiologists would have realized a 10.46 percent reduction in the number of work RVUs reported. For those physicians who are being compensated even in part by work RVUs, this is going to have a huge impact across the board. On average, had 2018 volumes been the same (or similar) to 2016 volumes, we would have seen an overall reduction of 3.6 million work RVUs, or 1.12 percent.

As with the financial impact, there are going to be winners and losers here. Some in this category maintained the same rank, such as podiatrists – who, if CMS was not going to change the entire RVU model for them, would have been the clear winner, while neurology (and not internal medicine, as was the case with the financial impact analysis) would have been the clear loser, with a reduction in work RVUs of 15.78 percent. The two tables below show the top 10 winners and the top 10 losers for work RVUs.

**Table 3: Top 10 work RVU winners under the 2019 proposed E&M changes**

**Table 4: Top 10 work RVU losers under the 2019 proposed E&M changes**

Calculating the total loss to providers is a bit more difficult, since each organization and each contract potentially has a different payment rate per work RVU. For example, let’s say that cardiologists are compensated, on average, at the rate of $57 per work RVU. And let’s say that 15 percent of their compensation is tied to this. Based on my analysis, cardiologists would have seen a reduction of 10.46 percent in work RVUs.

Let’s say that the typical cardiologist reports 6,500 work RVUs. If 15 percent of this is tied to compensation, then we could see that the average cardiologist sees a reduction in their work RVU-based compensation of around $37,000. Here’s the calculation: we take 15 percent of the 6,500 work RVUs, which give us 975 work RVUs tied to compensation, multiplied by $57 per work RVU. This equals a total wRVU payment to the cardiologist of $55,575. If the reduction of 10.46 percent holds true, then this would result in a net payment reduction of $8,114. The more that is tied to work RVUs, the greater the overall impact.

Just like the calculations above, then, you can create an analysis for your own practice. You would start with Table 1, which shows the difference between the 2018 and the proposed estimated 2019 work RVUs. Then, multiply the current frequency (for any period) by the difference between the two values and you will get the total difference in work RVUs for that period. Let’s look at an example for an internal medicine physician.

**Table 5: Sample work RVU impact analysis**

In this example, we can see the volume of each code during the given period of time let’s say it’s the first quarter of 2018. I multiply the unit work RVU by the frequency to get the total work RVU. I do the same with the impact; multiply the unit impact by the frequency to get the total work RVU impact. At the bottom, I can see, for this example, that the provider would have reported almost 550 fewer work RVUs for that period, which equates to a reduction of 27.41 percent. Let’s say that an internal medicine physician gets compensated at $50 per work RVU – and with our prior example, has 15 percent of compensation at risk. In this case, the 545.09 reduction in work RVUs equates to $4,118 in compensation for that period.

The bottom line is that this change is going to have far-reaching impacts, with tentacles reaching to all areas of practice management and the revenue cycle. I really don’t believe that CMS has given enough thought or created enough simulations in all of the difference areas to intelligently roll out these changes on Jan. 1, 2019.

Per my prior articles and my comments to CMS, I encourage them to use 2019 to test this model on some smaller sample of physicians, which will give everyone a chance to better evaluate the outcomes and consequences. If CMS really wants to eliminate confusion, contribute to administrative simplification, and calm the chaos in our industry, they will hold off on full implementation until appropriate testing can be completed.

And that’s the world according to Frank.