A Critique of the Office of the Inspector General’s Report on Swing Beds in Critical Access Hospitals

Return to search

Purpose

In March 2015, the Office of the Inspector General (OIG) issued a report on Medicare’s expenditures on swing beds in Critical Access Hospitals (“Medicare Could Have Saved Billions At Critical Access Hospitals If Swing Bed Services Were Reimbursed Using The Skilled Nursing Facility Prospective Payment System Rates.” Department of Health and Human Services, Office of Inspector General, March 2015, A-05-12-00046, hereafter “OIG report”). The purpose of this Policy Brief is to draw from our body of work in this topic to evaluate the methods and data of the OIG report.

METHODS AND DATA USED IN THE OIG REPORT

The March 2015 report is the latest in a series of OIG reports on payment for rural health care services. The report estimates the Medicare savings if swing bed days in Critical Access Hospitals (CAHs) were paid using Skilled Nursing Facility (SNF) Prospective Payment System (PPS) Rates instead of the current method of cost-based reimbursement. Below, we discuss what we believe are three important limitations of the methods and data that should be considered when interpreting the OIG report’s findings. The OIG report over-estimates the potential Medicare savings by removing swing beds from cost-based reimbursement. In a 2013 article,1 we show that the formula used by Medicare for reimbursement of swing bed care is complex and requires methods that recognize this complexity. Empirical estimates of the marginal cost per day to the hospital and the implied Medicare expenditure per day were carefully constructed, accounting for the increase in perunit fixed cost2 allocation when days are reduced. In contrast, the OIG3 used the following method to estimate the savings to Medicare from paying CAHs at a SNF rate:

The simple per diem method used by OIG fails to account for fixed cost transfers among services, over-estimating the Medicare savings of removing swing beds from cost-based reimbursement. Why? A typical CAH has different types of inpatient days: acute (Medicare and non-Medicare), swing (Medicare and non-Medicare), and observation. Medicare pays for a share of a CAH’s fixed costs based on the proportion of total inpatient days that were for acute and swing Medicare inpatients. Under the current cost-based reimbursement methodology, if swing bed days are removed from cost-based reimbursement, the CAH’s fixed costs would be allocated between acute Medicare and acute non-Medicare inpatient days only, a smaller number of days. Spreading the CAH’s fixed costs over a smaller number of inpatient days increases the fixed costs per day of the acute Medicare patients, and Medicare would be required to pay these additional costs because of the unique nature of cost-based reimbursement cost accounting. This fixed cost transfer thus offsets a portion of the OIG suggested savings achieved by removing swing bed days from cost-based reimbursement.4 In our 2013 article, we estimated that the average per diem cost of a swing bed day to Medicare ($581) is less than half of the average CAH total inpatient per diem cost ($1,302). We agree with the OIG that the average per diem cost of a swing bed day to Medicare is higher than the average SNF PPS per diem rate; however, the OIG estimate of the Medicare savings is unequivocally too large. The net effect of this difference between the $581 (actual reimbursement) and $1,302 (per-diem reimbursement) is considerable. The net savings (solely from changing payment methodologies) using the per-diem method is ($1,302 – $263.66)*829,104 days or $861m. The net savings using the correct method is ($581 – $263.66)*829,104 days or $263m. Thus, using 2009 data we estimate using the incorrect method yields estimates of potential “cost savings” that are over three times too large.

Read more