Skilled Nursing Facilities: Too many beds

March 27, 2018
Rebecca Laes-Kushner, MA, MPA
Formerly an Associate with the Center for Health Law and Economics

More than 15,500 skilled nursing facilities (SNFs) provide care to more than 1.35 million people in the United States who need assistance with their Activities of Daily Living (ADLs), including going to the toilet, getting out of bed, getting dressed, feeding themselves, and showering, or who have cognitive difficulties, such as from dementia.1 SNFs also provide post-acute skilled nursing care and rehabilitative services. Medicaid is the primary payer for long-term nursing needs of elders, while Medicare covers SNF rehabilitative stays of no more than 100 days. In 2015, Medicaid paid for care for 61.6% of nursing facility residents, while Medicare covered 14.2% of residents and 24.2% of residents primarily paid for their care themselves or through private insurance.2

Nationally, SNF use has declined as people live longer and choose home and community-based services (HCBS) over institutional care. From 2004 to 2014, the percentage of people age 65 and older in nursing homes dropped from 3.6% to 2.5%, a decrease of 24.5%. In addition, the number of people age 65 and older in nursing homes decreased 11.7% during the same time period, from 1.32 million to 1.16 million.3,4 By comparison, the number of nursing facility beds has only decreased 3.9% during the same time period.5 This disparity leads to overbedding, which, as I explain below, drives up Medicaid costs unnecessarily. 

Another factor influencing occupancy levels has been the increase in short-term rehabilitative stays at SNFs. Medicare rehabilitative rates ($436) are higher than Medicaid long-term care rates ($247)6, making rehabilitative services a more lucrative component of facilities’ service offerings despite the more intense need for staff resources. In New York state, long-term stays hovered at around 140,000 between 2000 and 2010, while very short stays, of 30 days or less, more than doubled from approximately 62,000 to 130,000 in the same time period.7 This increased turnover from rehabilitative residents leads to more unused bed days, i.e., a lower occupancy rate.

The problem with low occupancy rates

SNF bed use is measured by the occupancy rate, calculated as the number of resident days divided by the number of available bed days (the number of beds x days in the year). The occupancy rate shows what percentage of the beds are filled at a facility over the course of the year. Across the country, occupancy rates have been decreasing since the 1970s (see Figure 1)8, meaning there are more beds than needed.

When a SNF has a low occupancy rate, it is sitting on unused beds. This becomes a problem because per diem rates, which are calculated as the total cost divided by the number of bed days, can include costs for those unused beds. Similarly, capital costs are usually based on either actual costs (such as mortgage and real estate taxes) or an independent assessment of the market value of the facility (known as Fair Rental Value (FRV)); Medicaid could therefore end up paying for a 100-bed mortgage or the value of a 100-bed facility, even though typically only 80 beds are used. 

States have several policy levers they can employ to reduce unnecessary Medicaid spending that would otherwise be caused by low occupancy levels. This blog focuses on two such methods.

Lever 1:  Occupancy standards

Many states set minimum thresholds, or occupancy standards, for one or more of the cost centers (such as direct care, operations, and capital costs) used to set their SNF rates.  They pay the facility based on the higher of the standard or the facility’s actual occupancy level.  Suppose the occupancy standard is 90%.  If a facility’s actual rate is 85%, then the standard of 90% is used to calculate the per diem rate.  If the facility’s actual rate is 95%, then that higher rate is used to calculate payments.  See Figure 2


Massachusetts has the highest occupancy standard in the country, at 96%. Some states use state averages to set their standards; for example, South Dakota uses the greater of the facility’s actual occupancy rate or 3% less than the state average, while Rhode Island uses the higher of the actual rate or 98% of the statewide average.9

South Carolina’s occupancy regulations address the sparser populations in rural areas that may not be able to support high occupancy rates at existing facilities. The minimum occupancy level in the state is 92%, with two exceptions: In counties where the county-wide occupancy rate is less than 90%, then the state will calculate Medicaid payments based on the greater of the SNF’s actual occupancy rate or the average rate for the county that the facility is located in. In addition, in counties with only one SNF, the state will calculate Medicaid rates based on the greater of the SNF’s actual occupancy or 85%.10

Finally, in Tennessee, facilities are not eligible for incentive bonuses if their occupancy rates are less than 80%.11

Lever 2:  Bed Buybacks

A handful of states, including Louisiana and Oregon, have run bed buyback programs to encourage facilities to surrender their licenses for unneeded beds and thus increase the occupancy rate in the state. These programs, which are voluntary, enable the state to retire beds from circulation by offering an enhanced per diem rate for several years as a financial incentive for a SNF to “buy”—and then immediately retire—beds from another facility.

In Oregon, the state occupancy rate in 2013 was 60.1%, twenty points lower than the national average of 80.8%.12  The state began a bed buyback program in 2013, with the goal of reducing SNF beds by 1,500, or approximately 12.3% of the beds in the state. Oregon House Bill (HB) 2216 authorized the Department of Human Services (DHS) to work with SNFs to reduce statewide bed capacity, “in order to align financial incentives with the goals of achieving better patient care and improved health status while restraining growth in the per capita cost of health care.”13 

The bed buyback program allowed a SNF to purchase another facility’s entire bed capacity, with the seller surrendering its SNF license. In return for buying the beds, the purchaser received a daily augmented rate of $9.75 per resident day for up to four years. 

In 2014, 222 beds (14.8% of the 1,500 bed target) were reduced, with another 35 beds expected to be eliminated. However, all except a 78-bed SNF closed due to market or other conditions and not due to the bed reduction initiative.14 By 2015, the occupancy rate rose to 64%, now 18 points lower than the national average of 82%15 (see Figure 3), in part likely due to facility closings and voluntary bed surrenders that occurred outside the buyback program. 

The program faced challenges, however. In order to launch the program, the state’s Legislative Assembly had to exempt the purchasing and selling facilities from anti-trust laws, including federal laws, provided the state approved the transaction.  The buyback was extended by six months due to low participation.  Finally, one report commented that “potential [buyback] purchases are complex in part due to existing financing mechanisms whereby the land is held by another entity than the owner of the facility.”16   

Any bed buyback program will face similar risks, given that success is dependent on many external forces.  Therefore, other policy options may be better choices for a state looking to reduce costs caused by overbedding. 

Other policy levers

States employ other policy levers to reduce overbedding. A common technique is to limit the opening of new facilities through rigorous Determination of Need (DON) or Certification of Need (CON) regulations.  For existing facilities, setting rates at a percentage of median costs (120% of median) or at a percentile (90th percentile), can also help curb payments to facilities that have higher than average expenses in cost centers affected by excess beds (such as capital).    


What is the “right” number of beds?  That’s hard to say. 

Nursing facility use is decreasing, but eventually will level off. Baby Boomers are beginning to need more help with their activities of daily living (ADLs) such as dressing, bathing, going to the toilet, and moving about the home, and many prefer to get that care at home rather than in a nursing facility. But the highest need elders will likely continue to require care at nursing facilities. The huge capital startup cost to open a new facility means that beds taken offline are not likely to be replaced, in particular when an entire facility closes. Finally, we do not know the medical and technological breakthroughs coming in the next few decades.

Despite the murkiness of the future, state Medicaid agencies can act now to maximize policies, such as occupancy standards, that foster elimination of unneeded nursing facility beds and lower Medicaid costs.

1 Harrington, C. et al.  Kaiser Family Foundation.  “Nursing Facilities, Staffing, Residents and Facility Deficiencies, 2009 Through 2015”.  (July 2017).

2 Ibid.

3 U.S. Department of Health and Human Services, Centers for Disease Control and Prevention (CDC).  Health, United States, 2016: With Chartbook on Long-term Trends in Health.  (2017)

4 Note, this decrease occurred despite a 28.3% growth in the population age 65+.  Source:  U.S. Census Bureau.  Current Population Survey, Annual Social and Economic Supplement, 2004 and 2014.

5 U.S. Department of Health and Human Services, Centers for Disease Control and Prevention (CDC).  Health, United States, 2016: With Chartbook on Long-term Trends in Health.  (2017)

6 Dennison, T.  Medicaid Institute at United Hospital Fund.  New York’s Nursing Homes:  Shifting Roles and New Challenges.  (Aug 2013).

7 Ibid.

8 U.S. Centers for Disease Control and Prevention (CDC).  National Center for Health Statistics (NCHS). National Nursing Home Survey (NNHS) for 1977, 1985, 1995, and 2004 data; and National Study of Long-Term Care Providers (NSLTCP) for 2014.

9 Medicaid and CHIP Payment and Access Commission (MACPAC).  “States' Medicaid Fee-for-Service Nursing Facility Payment Policies”.  2014.

10 State of South Carolina.  “Plan under Title XIX of the Social Security Act.”  Attachment 4.19-D.  Approved 6/19/2013.

11 Medicaid and CHIP Payment and Access Commission (MACPAC).  “States' Medicaid Fee-for-Service Nursing Facility Payment Policies”.  2014.

12 Centers for Disease Control (CDC).  Table 101.  Nursing homes, beds, residents, and occupancy rates, by state:  United States, selected years 1995-2013

13 77th Oregon Legislative Assembly—2013 Regular Session.  “House Bill 2216”.  Section 15.  Passed by House 5/14/2013.

14 Oregon Department of Human Services.  Letter to State Emergency Board.  Subject:  Update on Nursing Facility Capacity Reduction Activities.  10/23/2014.

15 Oregon Department of Human Services.  “The State of Nursing Facilities in Oregon, 2015.” (April 2016)

16 Leading Age Oregon.  Legislative Brief.  2/13/2014.