Friday, February 13, 2009
To follow up on David's earlier posting about the final IRS Hospital report, I must say that I find the report on the whole a disappointment, particularly in its analysis of community benefit. Essentially, all of Part V. of the report (pages 39-92) is completely worthless. While the distributions and breakdowns all look very impressive, the simple fact is that all the charts and numbers are meaningless because the numbers that are reported by each hospital vary in their content and methodology. The IRS notes in the report (see page 2 of the report) that the reporting hospitals often varied significantly in their definition of "community benefit" - for example, some hospitals reported the difference between charges and private insurance reimbursements as "uncompensated care," which is absolutely ridiculous. As I've noted before, "customary charges" in the hospital context is an almost completely arbitrary number like the "rack rate" posted on the door of a hotel room. No one ever pays the "rack rate" for a hotel room, and no one except an uninformed, uninsured patient who doesn't know any better ever pays "customary charges" for hospital services. Put more concretely, if a hospital stated that it's customary charge for a single Tylenol was $100, and gave away 10,000 Tylenol to patients, then that hospital would have reported $1 million in "uncompensated care." Folks, that is just completely, absolutely nuts.
Other hospitals, according to the report, included Medicare "shortfalls" in uncompensated care (presumably, the difference between cost and Medicare reimbursements, though this was not entirely clear - it appears that some hospitals may be reporting the difference between customary charges and Medicare reimbursements), even though Medicare shortfalls are not included as "community benefits" in the final Schedule H to Form 990.
I suppose it was interesting to learn that certain categories of hospitals include bad debt in “uncompensated care” more than others, or that some categories include the difference between customary charges and private insurance payments as “uncompensated care” more than others but what this really means is that the numerical comparisons across categories or institutions are worthless. I hate to be a downer, but what good does it do to tell me, as Figure 73 does on page 89, that “uncompensated care” expenditures averaged 4.7% of revenues for hospitals having over $500 million in revenues, then tell me that well, that’s based on self-reported numbers that include all sorts of things that we don’t include as charity care on the Schedule H, and that other hospitals may or may not include these things, so that the 3.2% reported for hospitals with revenues between $25 million and $100 million might really be 4.7% if that group included all the things the other group did, or that maybe both numbers are really 2% once you got rid of the “slop” (e.g., reporting charge/reimbursement differences as uncompensated care; including Medicare shortfalls as uncompensated care, etc.). In short, you can take all those impressive charts and graphs in Part V. and burn them, because they offer nothing useful.
I was hoping that more of the report would focus on this fact – that is, that the IRS would concentrate on illustrating these differences in how hospitals report community benefit, and then show how the Schedule H has attacked these problems (or not). What really would have been useful is if the IRS had done something like "Group X of hospitals included the difference between customary charges and reimbursements as 'uncompensated care' and reported uncompensated care as XXX% of revenues; however, when we corrected the data to include on the difference between average costs and reimbursements, the uncompensated care rate fell to YYY%." The problem, of course, is that the IRS probably didn't actually have the information to do this useful thing because it didn't ask hospitals to report their exact methodology and to break out their numbers in this manner. The questionnaire only asked hospitals if particular items were included in their community benefit reporting (sometimes with a vague "please explain" plea at the end of the question, that I'm assuming was pretty much useless in terms of analyzing the reported numbers).
In addition, I was hoping that the IRS would use this data for some introspection on whether changes might be needed in Schedule H itself. For example, Schedule H does place some limits on community benefit reporting, by making hospitals report charity care separately from bad debt, and requiring charity care reporting at cost. But the Schedule H permits reporting organizations to choose its own accounting method in determining cost, which means that reported numbers almost certainly will not be consistent across reporting institutions. Moreover, the Schedule H reporting is geared to the entity that holds the EIN (taxpayer identification number), rather than requiring separate reporting for each hospital facility. In some cases, multiple hospitals will be covered by a single reporting entity (e.g., a single corporation operating multiple hospitals as "divisions" of that corporate entity), while in others, individual hospitals will have their own EIN and report separately. This means that it will be impossible to make the kinds of comparisons done in this report using Schedule H data because it will not be possible to separate out the reported Schedule H data for each underlying hospital. If a schedule H includes multiple hospital facilities of different sizes (in terms of patients served and revenues), or that span different geographic areas or population centers, for example, the IRS will not be able to use the Schedule H data to give us meaningful "breakouts" of community benefit expenditures by population size, geographic area, or even by the size/revenues of individual hospitals. In short, I was hoping that the IRS report would delve into these questions and think about how they might impact the future configuration of the Schedule H. But the report doesn't to that. The one sentence on page 168 that "The new Form 990 Schedule H reporting should reduce much of this variation" simply doesn't cut it.
I know the report was a lot of work, but frankly the section on community benefit expenditures was completely wasted work. I have great respect for the folks at the IRS who run TE/GE (the Tax-Exempt and Government Entities group), so why the IRS chose to give us 50+ pages of nonsense in this report, particularly in light of the new Schedule H, is a mystery to me.