The Health Disparities Question:
Will Pay-for-Performance Measure Up?

A Presentation By

Sandra L. Gadson, M.D.
President, National Medical Association

At the

AMA / National Minority Health Month Foundation
2006 Leadership Summit on Health Disparities

April 12, 2006

The Fairmont Hotel
Washington, DC

Good afternoon, everyone!


Since 1998, the National Minority Health Month Foundation has helped us to move forward the national conversation on racial and ethnic disparities in health.

The Foundation has gone beyond conversation and brought tools to the table to help us better understand and ultimately eradicate these disparities.

The issue of ensuring the highest standards of physician performance is certainly a vital part of this national conversation.

Quality improvement is an important goal for our nation’s healthcare system.

But policy initiatives designed to achieve this goal must carefully scrutinize their potential effect on minority populations.

In this regard, Pay-for-Performance is a “carrot-and-stick” concept that brings cause for concern.

In its evolving form, it offers bonuses for high performers, while tightening the purse strings for all others.

This may be a viable approach for improving the quality of widgets produced in a factory.

But it will certainly play out much differently when dealing with patients—particularly those populations with the highest rates of morbidity and mortality in the nation.

In my allotted time, I want to share with you the reasons why Pay-for-Performance—as it stands—will likely compromise both access and care for minority populations, and therefore widen racial disparities in healthcare.

Even a cursory comparison of black and white populations shows that these groups bring substantially different health profiles to the clinical encounter.

Significant racial disparities exist in health status; health system quality and access; healthcare utilization; and patient compliance.

In the interest of time, I will not delve into each of these, but I want to at least highlight the most salient features of these disparities.


Black Americans lead the nation in 12 of the top 15 leading causes of death, including heart disease, cancer, diabetes, and kidney disease.

Racial disparities in health status persist across the entire human lifespan. At the start of life: Black infant mortality is two and a half times higher than that of white babies.

And at the end of life: White men outlive black men by 7 years; and white women outlive black women by a half-decade.

When quantify the human toll of health disparities, the results paint a clear and troubling picture.

Depending on which causes of mortality you include in the equation, there are 85,000 to 130,000 excess deaths among blacks each year.

The term “excess deaths” refers to the number of deaths that would not have occurred if blacks shared the same mortality rate as whites.

This slide shows the racial gap in health status for infants between 1950 and 2000.

Although there has been an appreciable decline in infant mortality for both races, the racial gap remains striking, and the gap is widening.

In fact, as you can see from the red line, the racial gap in infant mortality was wider in 2000 than it was a half-century earlier in 1950.

When we look at mortality for “all causes,” here again, we see the distinctive health status gap between whites and blacks.

As with infant mortality disparities, we see the overall widening effect.

And here again, the racial gap in overall mortality was higher in 2000 than it was in 1950.

Comparing the life expectancy of black and white Americans with that of other nations brings the health status picture into sharper focus.

The women of Sri Lanka, Ecuador, and Bosnia outlive black women in the U.S.

The average life expectancy for black women is closer to that of Colombian, Syrian, and Palestinian women than it is to that of white American women.

When we turn to the international health standing of black American men, the picture is worse.

The life expectancy at birth for black males in the U.S. (68.8) is lower than that for males living in Sri Lanka, Colombia, and the Occupied Palestinian Territories—populations identified by the United Nations as having “medium human development.”

In fact, the average life expectancy for black males is much closer to that of Viet Nam, El Salvador, and Iraq than it is to that of white males in the U.S.

There are also sharp racial disparities among the uninsured.

In 2004, approximately 1 out of every 3 Hispanics and 1 out of every 5 Blacks compared to 1 out of every 10 whites was uninsured.

The uninsured have worse health and higher morbidity compared to the insured.

The uninsured are also more likely to forego needed care and obtain inadequate care for even the most serious illnesses like diabetes, heart disease, hypertension, kidney disease, cancers, and AIDS.

The uninsured are also less likely to receive preventive services such as screenings for breast, cervical, and colorectal cancer. When they do receive these services, they receive them less frequently than recommended.

When minorities do have healthcare coverage, there are still deep disparities in healthcare delivery which results in worse health and higher morbidity for minority patients.

In March 2002, a 15-member committee from the Institute of Medicine (IOM) released its 600-page report titled, “Unequal Treatment.”

The committee’s 18-month investigation found that racial disparity in healthcare was “remarkably consistent” across a 10-year corpus of literature.

This pattern occurred in every investigated disease area, including cardiovascular disease, HIV, diabetes, and end stage renal disease.

And the pattern persisted even after researchers took into account such factors as insurance, disease severity, and compliance with doctor’s treatment plan.

The IOM committee showed that these inequities result in significantly higher death rates for minority patients.

Lastly, there are racial and ethnic disparities in healthcare utilization and medical compliance.

Racial and ethnic minorities are more likely to avoid or delay seeking care. And they are less likely to comply with medical care.


These patterns also result in poorer health status, including greater mortality.

There are many reasons for these disparities in utilization and compliance, including cultural health-seeking patterns, lack of insurance, and a deeply embedded distrust of the medical system.

Taken together or individually, these factors constitute important differences in the profile of minority and non-minority patient populations.

These radically different patient profiles tell us that performance measures—and actions based on these measures, must take these differences into full account.

Otherwise, we engage in the business of comparing apples with oranges.

But patient profile differences are only one side of the coin.

We must also take into account physician practice patterns and patient perceptions of physician care.

First, minority doctors are more likely to serve minority populations. This means that minority doctors are more likely to have the disparate population profile that I just described.

Second, patients prefer physicians who share their racial or ethnic background.

Patients tend to rate their physicians’ communication style higher in race-concordant relationships.

And minority patients report higher levels of healthcare satisfaction when receiving care from minority physicians.

These patterns paint a clear picture about the uniquely beneficial clinical encounter between minority patients and providers.

This is why the IOM committee that reported on health disparities in 2002, and the Sullivan Commission that examined healthcare diversity in 2004, both called for an increase in the proportion of minority healthcare providers.

This all means that policy initiatives that threaten the viability of minority providers are counterintuitive to improving minority health and must be approached with caution.

And when you take these two scenarios together—poorer health status among minority patients, and the uniquely beneficial role of minority providers—you have a bigger picture that evokes a new set of questions:

• How will Pay-for-Performance affect physician practice patterns as they relate to sicker minority groups?

• Given the sicker caseload of minority doctors, how will Pay-for-Performance affect the viability of these physicians? And how will it affect the movement toward greater—not less—diversity in the healthcare workforce?

• What are the implications of all of this for minority access and quality of care?

• And ultimately, will Pay-for-Performance help or hurt the national effort to eradicate health disparities?

As I stated at the outset, I believe that Pay-for-Performance is cause for concern.