Last month we addressed the Hospice FY2019 Proposed Rule – Payment and Quality Reporting Program, we stated CMS is proposing to add Hospice Utilization and Payment Public Use File (Hospice PUF) data to Hospice Compare. The PUF data is derived primarily from hospice claims and the most current PUF data is from FY2015. The data would be in a segregated section of Compare as it contains information about hospice payments and utilization not quality measures. Other Compare sites such as Nursing Home Compare also contain respective provider PUF data. CMS believes the data serves as a resource for the healthcare community by providing information on hospice services provided to Medicare beneficiaries. It includes information for Medicare traditional fee-for-service beneficiaries as well as Medicare Advantage (MA) beneficiaries.

CMS also indicated in the proposed rule that other publicly available data could be used with/alongside the PUF data. CMS elaborates that the data may appear as shown in the PUF or after additional calculations. For example, the data could be averaged over multiple years, displayed as a percentage rather than the raw number so it has meaning to end-users, or other calculations in a given year or over multiple years. Supporting narrative may be added to make the data more understandable. By averaging or trending data over multiple years, CMS indicate it will make it fairer so that the data applies to hospices broadly regardless of size or location or other factors. Examples of how CMS could use the PUF/other publicly available data include:

  • Percent of days a hospice provided routine home care (RHC) to patients, averaged over multiple years
  • Percent of primary diagnosis of patients served by the hospice (cancer, dementia, circulatory/heart disease, stroke, respiratory disease) which would be a calculation of the total number of patients by diagnosis and dividing by the total number of patients that the hospice served
  • Site of service (long term care or non-skilled nursing facility, skilled nursing facility, inpatient hospital) with a notation of yes, based on whether the hospice serves patients in that facility type

Hospices, and the public, can view the most current PUF here. A hospice could view the current PUF data and compare its results to that of its peers. It can be viewed by provider number and by state. The Hospice PUF includes data for providers that had a valid identification number and submitted a Medicare Part A claim during the calendar year. The Hospice PUF provider table excludes providers with less than 11 hospice beneficiaries in calendar year. There are over 40 data content groupings including Average Age, Percent of Deaths in Hospice, Percent of Live Discharges, and diagnosis information. As stated above, it includes data for both Medicare fee-for-service and Medicare Advantage beneficiaries. That data is not a representation of quality. As such, the data are not risk adjusted and thus do not account for differences in the underlying severity of disease of patient populations treated by providers. The tables do not include information on length of stay, but rather days of care provided and hospice beneficiary counts. Hospice beneficiaries with very long hospice stays may be in hospice care for several years making it difficult to attribute long stays to a single calendar year.