At a Convention of the Medical Committee for Human Rights held in Chicago in March 1966, Martin Luther King Jr declared: “Of all the forms of inequality, injustice in health care is the most shocking and inhumane”,¹ and in the Institute of Medicine report “Crossing the Quality Chasm: A New Health System for the 21st Century”² published in 2001 it is stated that a healthcare system should provide “care that does not vary in quality because of personal characteristics such as gender, ethnicity, geographic location, and socioeconomic status”.

In the 21st century there are still countries where health systems put a price on care that the patient must pay at the point of service. For example, in the United States, where over 45 million citizens are without healthcare insurance, many must purchase care with personal resources. As a consequence, health care competes with the other necessities of life like food and clothing. Such health systems can be characterized by variation in healthcare outcomes that is related to patients’ ability to pay.^3,4 Bangladesh is geographically and culturally far from the US, but the predicament of pregnant women in Bangladesh described in the paper by Pitchforth and colleagues published in this issue of QSHC⁵ would probably be familiar to many uninsured and underinsured Americans.

While the general adherence to evidence-based care processes varies little across sociodemographic groups in the US,⁴ the patient’s ability to pay has been linked to substantial variation in healthcare outcomes. The 2004 National Healthcare Disparities Report³ prepared by the US Federal Agency for Healthcare Research and Quality (AHRQ) identified racial, ethnic, and socioeconomic disparities in accessing quality acute, chronic, and preventive care for Americans. By employing 31 measures of healthcare access, AHRQ found that people who had incomes of less than 100% of the US federal poverty level for a family of four—$18 850 (£10 860) in 2004—had worse access to care for about 80% of those measures.

Such gaps in healthcare quality outcomes in the US exist at both the national and the local level. A woman who lives in the South Bronx region of New York City, where the population is 95% Black or Hispanic, and very low income, has a 20 times greater chance of dying of the complications of diabetes than a woman living in the higher income Upper East Side of Manhattan.⁶ A survey of four diabetes related specialty services at three South Bronx hospitals showed that many specialty services were not available to patients whose care was supported by fee-for-service Medicaid, a government payment system that provides modest payment to clinicians. Just as a bureaucratic “system” may impede a Bangladeshi woman’s chance of receiving relief from the Social Welfare Organisation or the “Poor Fund”, a South Bronx patient is usually unaware of the availability of charity funds for uncompensated care that could be applied to his or her medical bill.

The six aims promulgated by the IOM report “Crossing the Quality Chasm”² for a better healthcare system in the US—healthcare that is safe, effective, efficient, timely, patient centered and equitable—provide a framework that is increasingly being adopted by improvement scholars. For example, Chin and Chien⁷ have proposed that scholarly reports of healthcare improvement should strive to accommodate disparities when documenting improvement research. They postulate that there are untapped opportunities for quality improvement scholarship to address inequitable care. On the other hand, Asch and colleagues⁴ found that the difference in the quality of evidence-based health care in the US was small when they evaluated groups with incomes over $50 000 (56.5% of recommended care) and those with incomes less than $15 000 (53.1%). These investigators emphasized the need to focus on inequity at the point of access, but they found that the differences between sociodemographic subgroups were relatively small compared with the gaps in quality across all groups.

It is noteworthy that Pitchforth and her colleagues have characterized variation that is related to ability to pay for care as an opportunity for systems improvement.⁵ Based on their study, they proposed a local system change that would make financial assistance more readily obtainable for a woman with a complicated obstetric delivery. Similarly, the Bronx Health REACH program has provided a framework for addressing inequity at the community level that brings together the many groups that play a part in this problem.⁶ Their coalition of social service agencies, healthcare providers, after school programs, housing development programs, and faith based groups has sought to identify the roots of the problem. Moreover, REACH has formulated a regulatory and legislative agenda that includes the far reaching goal of universal health insurance along with more immediate goals such as increased public health education and greater diversity in all healthcare professions.

Stories at the personal level and statistics at the community and national level converge to make clear the dimensions of inequality in health care when the patient pays. An obstetric suite in Bangladesh and a primary care clinic in the South Bronx both provide distressing examples of settings in which such variations in quality occur. An emphasis on systems improvement offers an opportunity to improve inequitable care. Unfortunately, until national resources and/or political will can lead to the larger system changes that are required to change the economic context of care, improvement strategies at the local level will have to suffice.