In 2014, Medicare spent more than $15 billion on hospice care for 1.3 million of its beneficiaries. More precise breakdowns of where those dollars went, however, have not always been publicly available. A new data set the CMS released Thursday is likely to change that, even as some raise concerns about the accessibility and reliability of other spending data provided by the agency.
The Hospice Utilization and Payment Public Use File contains privacy-protected information on the services of 4,015 hospice providers to Medicare beneficiaries. It covers details such as days spent in hospice care, regional variations in spending, and patient demographics and diagnoses. The data are drawn from Medicare claims and are scrubbed of information that could identify individual patients, the CMS said.
“CMS believes that greater data transparency leads to a more effectively functioning health care system, which leads to better care and smarter spending,” Niall Brennan, CMS' chief data officer, said in a statement.
In 2014, Medicare spent, on average, $11,393 per hospice beneficiary for 70 days of care. But that varied substantially by state. In Iowa, South Dakota and Wyoming, average hospice spending per beneficiary ranged from $7,000 to $8,600. But Medicare spent between $13,400 to $15,000 on the average hospice beneficiary in Georgia, Texas and four other states.
Some have called into question the reliability and accessibility of CMS's vast ocean of spending data.
In a separate report released Thursday, the U.S. Government Accountability Office scrutinized expenditure data from skilled-nursing facilities. Medicare spent $28.6 billion on skilled-nursing facility services for nearly 1.7 million beneficiaries in 2014. These facilities have to self-report spending in annual reports, information that the CMS publishes in a raw format on its website.
“CMS has not provided the data in a readily accessible format and has not posted the data in a place that is easy to find on its website, according to public stakeholders and GAO's observations,” GAO's report said. “In addition, CMS does little to ensure the accuracy and completeness of the data.”
In its own analysis of variation among these facilities, the GAO found that for-profit skilled-nursing facilities' direct and indirect care costs as a percentage of revenue were lower than those of government and not-for-profit SNFs. Nurse-staffing ratios also varied, and among for-profit groups, the GAO found a correlation between a decrease in time nurses spent with patients and an increase in profit margins.
The report recommended that the CMS make skilled-nursing facilities' spending data more accessible, accurate and complete.
As the CMS attempts to tie a growing proportion of its healthcare dollars to value rather than volume, it has rolled out a slew of spending programs and care models that hold hospitals, physicians and other providers financially accountable for the services they use. For these groups, knowing where their dollars return the greatest value is key.
In bundled-payment programs, for instance, a hospital receives a fixed amount to oversee the full spectrum of care, including after discharge, for an episode such as a heart attack. Having more data helps them ensure that they send patients to post-acute care facilities that offer the greatest bang for their buck.
Value-based payment models are only expanding, as HHS has set the goal of tying 90% of traditional Medicare payments to quality or value by 2018. The CMS' voluntary and mandatory bundled-payment initiatives alone encompass hundreds of hospitals across the U.S.
The CMS also updated its Market Saturation and Utilization Data tool, it said Thursday, to include hospice services for the first time. Now in its third iteration, the tool also contains information on utilization and services from home health facilities, ambulances, independent diagnostic testing facilities and skilled-nursing facilities.
