Disparity clarity

One of the central themes of the Patient Protection and Affordable Care Act, mentioned numerous times throughout the law, is the pressing need to eliminate healthcare disparities.

But some health policy experts worry that provisions of the law designed to improve healthcare quality could exacerbate gaps in access and outcomes by penalizing the hospitals that minority and poor patients depend on the most.

Racial and ethnic minorities tend to seek care at hospitals that often don't perform as well as others on quality metrics, says Dr. Anne Beal, chief operating officer of the Washington-based Patient-centered Outcomes Research Institute, an independent, not-for-profit organization established by the healthcare reform law to promote comparative-effectiveness research.

That's not because clinicians at those hospitals are not interested in providing top-notch care, Beal says. Rather, it's a problem of resources, especially at institutions that depend heavily on Medicaid reimbursement.

“The differences in care that we are seeing are driven not by who you are but where you go,” she says. “Providers in those settings have complicated, high-risk patient populations that make it much more difficult to achieve high levels of performance. Therefore, they're much less likely to benefit from programs that reward it.”

In a commentary piece in the October 2011 issue of Health Affairs—which was dedicated to the topic of health disparities—Beal, who was then president of the Aetna Foundation, praised the healthcare reform law's requirement that quality data must be collected and stratified by race and ethnicity.

But despite the increasing availability of such data, serious challenges will remain for those institutions where most low-income and minority patients seek care, Beal says.

Low-performing hospitals, which are concentrated mainly in the South, care for much higher proportions of minority patients than do top-performing hospitals in the Northeast, according to another study that appeared in the same issue of Health Affairs.

The study's authors, led by Dr. Ashish Jha, used a number of sources to gauge hospital quality and costs, including the CMS' Hospital Compare, the American Hospital Association's annual hospital survey from 2007, and the 2008 Hospital Consumer Assessment of Healthcare Providers and Systems survey.

A total of 122 hospitals received a “best” rating, while 178 were rated “worst,” according to the study. Nearly 15% of discharges from the lowest-performing hospitals were older black patients. But that same group accounted for only 6.8% of patients at the best hospitals.

The lesson, says Jha, associate professor of health policy at the Harvard School of Public Health, Boston, is that the site of care plays a critical role in disparities.

“Minorities and the poor are much more likely to end up in facilities that are poor performing for everyone,” he says.

In the context of the CMS' value-based purchasing program, first proposed in January 2011 and finalized in April, those differences in baseline performance could mean financial penalties and even fewer resources for already struggling hospitals, Jha says.

The first year of the value-based purchasing program ties hospitals' incentive payments for discharges after Oct. 1 to performance on a set of measures of clinical processes of care and patient experience. Funds for the incentive payments come from a 1% across-the-board cut in base operating DRG payments, rising to 2% by 2017.

Hospitals can earn points for achievement—based on where an organization's performance on a particular measure fell within a benchmarked achievement range—or for improvement. For instance, in an example provided by the CMS, a hypothetical hospital scored 0.91 on a pneumococcal vaccination measure during the performance period. Because the benchmark for that measure was 0.87, that hospital received the maximum 10 points for the measure, according to the CMS.

Another hypothetical hospital scored 0.7 on the same measure, earning six points for achievement. But because that hospital's performance improved substantially, from 0.21 to 0.7 during the performance period, the hospital also received seven improvement points. The CMS uses the higher of the two scores to determine the score for each measure, so the hospital's final score was seven. The individual scores are used to calculate each hospital's total performance score, which is then used to determine the incentive payment.

The CMS says the improvement score keeps the program from unfairly penalizing low performers.

It was that safeguard that three CMS officials cited in a January letter in Health Affairs, responding to the study authored by Jha and his colleagues. The officials argued that the highest-performing hospitals and those that showed improvement could be rewarded equally.

“In fact, the program will not penalize low-performing hospitals for failing to meet the benchmark set by top performers,” they wrote in the letter. “Hospitals can earn the full incentive payment by improving their performance relative to their own baseline or by attaining an established benchmark.”

Also, no hospital would be left “to sink or swim,' the officials wrote, citing assistance available through Medicare quality-improvement organizations and other sources.

In a response letter in the same issue of the journal, Jha and his colleagues acknowledged that the improvement score would create the opportunity for low performers to capture the same payments as high-performers. But that's only if those hospitals at the bottom actually improve. And with already tight budgets and scant resources, improvement is not a certainty, Jha says.

Dr. Mona Fouad, professor and director of the division of preventive medicine at the University of Alabama at Birmingham, says the environment is challenging for physicians who struggle to provide the extra support and attention that high-risk patients need to manage their health.

“These patients often seek care in the emergency room,” says Fouad, also the founder and director of the UAB Minority Health & Health Disparities Research Center. “They get treated, it's often a one-time thing, and then they disappear until there is another acute issue.”

And as pay-for-performance initiatives roll out, hospitals that serve these populations will be pressured to come up with creative ideas to avoid being dinged by penalties, she says.

The stakes will be even higher in the second year of the value-based purchasing program, when the CMS begins to look at clinical outcomes through the use of several inpatient mortality measures, says Dr. Karen Joynt, an instructor in the department of health policy and management, also at the Harvard School of Public Health.

High-risk patients often come to the hospital sicker, she says, and they have less access to resources outside the hospital.

“These are generally not hospitals that are operating on big margins, and the program does run the risk of exacerbating disparities,” Joynt says, adding that the CMS' inclusion of the improvement score could provide some protection.

There is no such safeguard for the healthcare reform law's preventable readmissions penalty, Joynt says.

Beginning Oct. 1, hospitals with high readmission rates for acute myocardial infarction, pneumonia and heart failure over the previous three years will see reductions in their Medicare payments up to 1%. That penalty rises to a maximum of 2% of payments in 2013 and 3% in 2015.

In a February study in the Journal of the American Medical Association, Joynt and her colleagues found that among elderly Medicare beneficiaries, black patients had higher 30-day readmission rates for all three conditions, a gap that was attributable both to race and to site of care, they said.

Using preventable readmission rates as a quality metric is problematic, Joynt says, because it's not clear that lower readmission rates translate to higher quality.

“Lower readmission rates could mean a higher mortality rate,” she says. “Also, if patients have access to good-quality outpatient care, the pool of patients that are admitted to the hospital will be sicker and more likely to be readmitted. Improving overall quality can actually increase hospital readmission rates.”

Ensuring adequate support and access to care after discharge becomes much more difficult with patients who are transiently housed or who don't speak English or who can't afford medications, Joynt adds.

“Hospitals with a large number of those patients have to do a lot more to keep those patients from being readmitted,” she says. “It's hard to hold them accountable. It's probably a better measure for a system or an accountable care organization.”

Not all policy experts are convinced that minority patients receive care at lower-performing hospitals. Darrell Gaskin, deputy director of the Center for Health Disparities Solutions at Johns Hopkins University, Baltimore, calls that theory false.

The reality, says Gaskin, who is also an associate professor of economics at the university's department of health policy and management, is that hospital quality depends heavily on which measures and minority groups you're looking at.

“People should be very careful about saying minority patients are more likely to use lower-quality hospitals,” Gaskin says. “It's just not uniformly the case.”

He argues that the idea has taken hold because of studies that looked specifically at care for heart patients. That research uncovered significant differences in quality, he says, and the tendency has been to extrapolate the findings to hospitals and minority patients in general.

In a study published in December 2011 in the journal Inquiry, Gaskin and his colleagues used several years of inpatient discharge data as well as inpatient quality and patient-safety indicators from HHS' Agency for Healthcare Research and Quality. They looked at mortality rates for a long list of procedures, conditions and adverse events, including hip replacement, stroke, heart failure, hospital-acquired infections and postoperative hemorrhage.

“What we found is that these things are not correlated with one another,” Gaskin says. “There's no pattern. Just because a hospital did poorly on one measure did not mean it would do poorly on another.”

For instance, the researchers found that having a high percentage of Hispanic patients was actually associated with higher performance on patient-safety indicators. And as the percentage of black patients increased, rates of seven adverse events rose but rates of four others fell, they reported.

“The take-home message is that if you want to talk about the relationship between caring for minority patients and the quality of hospital care, you need to make sure you are talking about specific services and conditions,” Gaskin says.

Perhaps because of the results of that study and other past research, Gaskin says he is more optimistic than some about the prospects of value-based purchasing. Despite concern in the field that performance-based incentive payments could unfairly penalize hospitals with fewer resources, he contends that as long as the measures are reasonable, the program should work.

“The idea that we are going to have higher complication rates, higher mortality rates and lower rates of concordance with guidelines just because we have hospitals that treat poor and minority patients does not hold water with me,” he says.

Write us with your comments. To send us a letter electronically, go to modernhealthcare.com/letters. By fax, it's 312-280-3183; or through the mail it's Modern Healthcare, Letters to the Editor, 360 N. Michigan Ave., Chicago, Ill. 60601. To publish letters, we need your name, title, affiliation, location and phone number.