Researchers find that the salaries of top hospital leaders can be linked to the use of advanced technology and favorable hospital experience survey scores. Facilities with low 30-day mortality rates pay their CEOs less than hospitals with the highest mortality rates.
|
How much a CEO gets paid to run a non-profit hospital in the U.S. is not linked to how well that organization performs on quality and safety measures, readmission or mortality rates, its profit margin, or the extent of its charity care.
Rather, the top leader's salary is associated with the organization's level of advanced technology, and to the extent its patients respond favorably to federally required hospital experience surveys.
Those are some of the conclusions from the first study of executive compensation since non-profit CEO pay became publicly available last year in federal 990 tax returns for 2009. Those packages were then paired with outcome and process quality measures including those reported on Medicare's Hospital Compare, and mortality rates for 19 medical and surgical conditions.
The survey was undertaken by Ashish Jha, MD, and collaborators at the Harvard Department of Health Policy and Management and was published Monday in the JAMA Internal Medicine.
Jha postulates that since physician pay is partially determined by certain measures of quality, why shouldn't the compensation packages of hospitals' top executives be as well?
"If I, as a practicing physician, have part of my salary tied to quality of care that I deliver, which by the way, I do as a practicing physician at the VA (Veteran's Affairs Boston Healthcare System), I wondered if this was happening in the management ranks as well.
Governance a Factor
"And what we found is that it really didn't seem to be happening. So we said, boy, we think there's an opportunity here for improvement, for hospital boards of directors to focus on this and make it part of a CEO's compensation package. It would, at least, give us one more lever to drive some change."
In fact, the study found that hospitals with low 30-day mortality rates paid their CEOs $4,667 less than did hospitals with the highest mortality rates.
The study included salaries for 1,877 top executives who are responsible for 2,681 non-profit hospitals, and found they earned $595,781 in unadjusted mean compensation. CEOs in the lowest 10% received a median compensation of $117,933, and were largely in charge of rural, small hospitals, most frequently in the Midwest.
In the highest 10%, CEOs made $1,662,548. They oversaw large urban hospitals that often included medical schools.
Technology Boosts Pay
Other factors linked to higher CEO paychecks included bed size. Researchers found about $550 in additional CEO pay per additional bed and number of hospitals in a system, with $32,609 more for each additional facility.
Perhaps not as surprising, Jha says, is the link between the level of a hospital's technology, as measured by the industry's "technology index," and CEO pay. Hospitals with high levels of technology paid their CEOs about $663,800, or $136,000 more than hospitals with the lowest levels of technology.
Asked if the fact that there was no link between CEO pay and quality of care seemed odd, Jha says that it did surprise the research team.
"We didn't think there would be a big effect, but we were hoping that, given the national conversation we've been having about the importance of quality and reducing unnecessary healthcare spending, that boards would have taken this on a little more and made it a little more a priority, and [tried] to hold their senior managers a little bit more accountable than they are doing."
Asked which quality measures he thinks should be in the equation to determine CEO pay, Jha says that outcomes should be included and "should trump process measures."
"What do patients care about when they have to go to the hospital because they're sick? They want to survive and do well. So I say the things that really matter are those that measure patient outcomes, avoiding infections, medication errors, and other medical errors."
And while patient experience, as measured by the Hospital Consumer Assessment of Providers and Systems (HCAHPS) surveys, is important, too and should be included, hospital boards should direct their CEOs to work on whatever quality areas need improvement. "It depends a little bit on where you are on the performance curve already."
Research "Missteps" Cited
How Jha and colleagues interpreted their findings was criticized by Warren Browner, MD, CEO of California Pacific Medical Center in San Francisco, who wrote in an accompanying invited commentary that the Harvard researchers made "several missteps" in confusing correlation with causation.
"…Their most disturbing finding was that CEO pay correlated with patient satisfaction, but not with quality. They see this as a missed opportunity and recommend that hospital boards provide incentives for CEOs to meet quality goals," Browner wrote.
"That advice seems strange, since every hospital CEO I know who receives incentive compensation already has quality-related goals, as did approximately three-quarters of CEOs who were surveyed recently."
A more likely scenario, Browner wrote, was that "hospitals are concerned about, measure, and reward different quality metrics" than the ones the Harvard researchers focused on, for example, severity-adjusted sepsis survival rates.
"Indeed, the profusion of quality metrics challenges the ability of hospital quality departments to keep up with measuring and reporting them."
"Indeed," he added, "some hospital CEOs have likely figured out how to game the system by working to align their incentive compensation with the metrics that their hospital usually achieves—and then trumpeting those results. Thus, the opportunity that the authors have identified is to standardize how quality is measured, rather than allowing the indiscriminate proliferation of metrics to continue."