Midwestern hospitals spread referrals to a greater variety of skilled nursing facilities and tap their favorite SNFs less often than do hospitals in other parts of the country, according to a recently published analysis of referral patterns.
In the Midwest, hospitals routinely referred patients to an average of about 14 SNFs in their vicinity as of 2008, the investigators determined. Roughly three-quarters of referrals were concentrated among a hospital’s five most-frequently utilized skilled facilities.
This is compared with an average referral network of 10.5 SNFs in the South, which had the narrowest networks. Nearly 85% of referrals in this region were concentrated among a hospital’s top five nursing homes.
Determining the causes of these regional discrepancies was outside the scope of this study, the investigators stated. It is possible that the variation in referral patterns is due in part to the number of post-acute facilities in a particular region, they wrote. Rural hospitals and not-for-profits tended to have smaller post-acute networks than urban hospitals and for-profits, they noted.
The findings are based on data from a variety of sources, including Medicare claims and and Institute of Medicine figures. The investigators examined skilled nursing facility admissions within 14 days of hospital discharge for Medicare beneficiaries at least 65 years old, and in either Part A or B.
The data came from 2008, prior to the ongoing push toward bundled payments that reward close coordination between acute and post-acute providers. But the findings seem to support payment systems that encourage hospitals to work with smaller networks of post-acute facilities.
“Hospitals in regions with lower post-acute care spending have smaller, more concentrated networks,” the study authors found. This type of preferential referring is a desired principle, according to bundled payment advocates. Bundled pay initiatives encourage hospitals to work closely with select post-acute partners, include nursing facilities and rehab centers, to provide needed services for a whole episode of care.
However, while the study seemed to support the lower-spending goals of bundled payment initiatives, it also revealed some challenges they face, the investigators noted.
For instance, hospitals face a “difficult tension” with regard to low-volume SNFs — those that receive 1% or less of a hospital’s annual referrals. Patients might be going to these SNFs for reasons that do not lower overall spending but are personally rational, such as proximity to home, the researchers noted. Furthermore, the government does not allow hospitals to limit patients to particular post-acute providers, so diffuse patient flow to low-volume SNFs likely will continue. Even though they account for only about one in 10 referrals, these facilities still could “undermine a hospital’s ability to save money under bundled payment” if they are inefficient, the authors stated.
The research team included members from a variety of institutions, including the University of California at Irvine, the University of Southern California, Harvard Medical School and the RAND Corporation. Full findings appear in the journal Healthcare: The Journal of Delivery Science and Innovation.