IRS survey data unreliable, MHA says
The Internal Revenue Service recently released its final report on a survey of more than 500 nonprofit hospitals, focusing on community benefit and executive compensation.
But a spokesman for the Michigan Health & Hospital Association discounted data, arguing that the revised Form 990 and its companion Schedule H, required for the first time for tax year 2009, will provide a much better look at nonprofit hospital operations.
“The report that just came out is a little bit confusing in the sense that the data they collected from a small segment of not-for-profit hospitals really, we don’t believe, are representative of all the not-for-profit hospitals in the U.S., and specifically here in Michigan,” said Jim Lee, MHA’s vice president of data policy and development.
Community benefit was established in the 1960s as a standard for determining a hospital’s federal tax-exempt nonprofit status. The concept encompasses a wide variety of hospital activities that have a positive effect on their communities — from classes to events to free care for the needy — but there is no exact formula for deciding what should be included as community benefit.
The study found that the average community benefit reported by the hospital responding to the questionnaire was 9 percent of hospital revenues, and the median was 6 percent. Other than 15 research hospitals, the major component of community benefit was uncompensated care, accounting for 56 percent. Just 9 percent of the hospitals reported 60 percent of the sum of community benefit.
The American Hospital Association criticized the questionnaire and said the hospitals varied widely in the type of items they reported as community benefit. For example, the gap between the cost of providing care and Medicaid payments is a common measure, but not all hospitals report on the same gap for the Medicare program and can be inconsistent in the cost basis.
For example, Spectrum Health routinely includes the Medicare gap in reporting its community benefit to the public, but Saint Mary’s Health Care uses only the Medicaid gap.
“They gave some guidance in collecting this data to not-for-profit hospitals, but they didn’t give clear guidance in terms of how hospitals should report information,” said Lee, whose organization represents 144 Michigan hospitals.
“So in their report, they acknowledge some hospitals included Medicare losses and others did not. Some included bad debt and others did not. Some calculated things at cost while others calculated things at charges. Those things obviously have dramatic impact on the data itself,” Lee said.
According to the MHA, in 2007, the latest year for which data is available, Michigan hospitals in 2007 provided: $209 million in charity care; $606 million in bad debt; $607 million in Medicaid shortfalls; $586 million in Medicare shortfalls; $48 million in other government program shortfalls; $95 million in community health improvement services; $25 million in health education and screening; $35 million in free and reduced-cost clinics; and $23 million in cash and in-kind support.
In its audited financial statement for fiscal year 2008, Spectrum Health reported spending $111.1 million on community benefit, citing $84 million in losses on shortfalls in payments from government programs. CFO Michael Freed said at last year’s public budget meeting that community benefit was expected to rise to $154 million in the current fiscal year and attributed $129 million of that to Medicare and Medicaid losses.
At a press briefing about the report, Lois Lerner, director of the IRS Exempt Organizations Division, said the agency plans to move ahead with discussions with “stakeholders” about the 40-year-old community benefit rule and its modifications over the years.
Lee said reliance on the current report is a shaky foundation.
“I think the prudent thing for everyone to do is to wait for the 990 process to play out,” Lee said. “I think you’ll start to see a more standardized data collection process, a more standardized set of data to really do some analysis on.”
Regarding executive compensation, the IRS review showed that most hospitals use comparability data and independent personnel to levels for top executives. The IRS took a closer look at 20 hospitals whose executive compensation, as reported on the 990 form, was at the high end for their type and size. It did not identify which hospitals it reviewed.
Overall, compensation was lower at critical access and smaller hospitals. The overall average was $499,000 and the median was $377,000. But at the 20 chosen for special examination, average compensation was $1.4 million and the median was $1.3 million.
On its Form 990 filing for the 2007 tax year, which was submitted in 2008, Spectrum Health Systems listed total compensation for 2006, the most recent full calendar year, for President & CEO Richard Breon at $1.8 million, including $774,528 in base compensation and the rest in incentive bonuses, benefits and “personal fringe benefits.”
For the tax year ending June 30, 2007, Saint Mary’s reported through its Doran Foundation that President & CEO Philip McCorkle was compensated a total of $641,061 for 2006, including pay of $533,168.
For the same period, Metro Health Hospital reported compensation for President & CEO Michael Faas during 2006 of $689,603, including a base compensation of $636,000.
Lerner said the agency hopes to review the quality of comparability data and the practice of setting aside the objective measures in negotiating the first contract with the executive.
She also said that the report provides important information as the Obama administration is expected to forge forward with health care reform.
“While we believe the data we’ll be receiving from the new Form 990 hospital schedule will give the IRS and the public a better view of how nonprofit hospitals operate, we won’t be seeing any significant data until 2010 or 2011, when the new schedules start to be filed in substantial numbers,” Lerner said. “In the meantime, we will need to work with others to evaluate what guidance might be needed, and we believe there is a lot of information in this report that can inform those discussions.”