As U.S. medicine becomes patient-based and patient populations grow more diverse — culturally, ethnically, racially, and in terms of faith, lifestyle and demographics — the national institution that is medicine in America must unravel and address serious disparities in health care treatment and delivery that the Institute of Medicine has decried as one of the nation's "most serious health care problems."

by David Wilkins

While diabetes is virtually non-existent among Arab nomads in the Egyptian desert, 40 percent of Arab immigrants in southeast Michigan are diabetic or glucose intolerant. African-Americans seeking medical help for chronic pain report more severe symptoms — greater pain intensity and severity, depression, disability and post- traumatic stress disorder — than white patients do. Native American women are less likely than women from other racial and ethnic minority groups to contract breast cancer, but more likely to die within five years when it is detected.

Why? The question is compelling, the answers complex and often unclear.

The tangled web of contributing forces to disparities in disease prevalence, health care and medical outcomes includes discrimination, acculturation, lifestyle and behavior, as well as limited access to and utilization of health care, variations in diagnosis and treatment, exposure to environmental health risks, genetic predisposition and socio-economic factors.

David Gordon Photo: Martin Vloet

"Disparities in the health of — and in health care delivery to — different racial, ethnic, and socio-economic populations have existed for at least decades and probably as long as statistics have been gathered on these issues," says David Gordon, M.D., professor of pathology and associate dean for diversity and career development. "Some gaps have lessened over the years with improvements in health care, but most remain and some may even be widening with time.

"Interestingly, medical conditions with the greatest disparities — including cardiovascular disease, obesity and related diseases, and prostate and breast cancer — are the major causes of death for the entire U.S. population," he says. "So the opportunity is clear. Any insights on how to improve health status and health care delivery in these areas should lead directly to improvements in the overall health of our country."

The Institute of Medicine's groundbreaking 2002 report, Unequal Treatment: Confronting Racial and Ethnic Disparities in Healthcare, concluded that:

• The most significant barriers to equitable care are factors which affect access to treatment — differences in income, lack of health insurance or reliance on publicly funded insurance, high co-payments, inadequate transportation, and a scarcity of nearby health care services.
• When limited access to care is removed from the equation, however, significant disparities in morbidity and mortality persist — suggesting bias and stereotyping play a large role.
• Minorities receive lower quality and less intensive health care and diagnostic services than whites across a wide range of medical conditions and treatment regimens.

While the important work of documenting disparities continues, medical researchers, including dozens at U-M, are digging deeper. Their goal is to understand why disparities occur. Why, after suffering a stroke, are Mexican-Americans less likely than non-Hispanic whites to take blood-thinning drugs that can reduce the risk of another stroke? Why are African-Americans four to five times more likely than other groups to develop kidney disease? Why are there often treatment delays when a woman suffers a stroke?

Understanding why these differences exist is the first step toward eliminating them, U-M researchers say — by toppling barriers to care, rooting out bias, correcting communication lapses between doctors and patients, identifying variations in treatment, understanding biological differences, and targeting unhealthy patient behavior with carefully calibrated education.

Carmen R. Green Photo: Martin Vloet

"We need to understand the underlying reasons for these disparities in order to design interventions that are person-specific or group-specific," says Carmen R. Green, M.D. (Residency 1992), a U-M anesthesiology professor who studies physician decision-making and disparities in pain management. "In doing so, we can improve the quality of life for all people."

Treating patients equitably, then, does not mean treating them identically. It means, in fact, accounting for and addressing the differences that create disparities — the tendency to under-prescribe pain medication for minorities, women and the elderly, for example, or the way Westernized lifestyles and genetic predisposition combine to make Arab immigrants susceptible to diabetes.

As a fellow in the U-M Multidisciplinary Pain Center in the early 1990s, Green wondered why some patients coped better than others with unrelenting pain. She also noted variations in the care they received. "I saw differences in treatments that had previously been provided or were currently being provided. I started asking why."

A decade later, Green is a national leader in the study of disparities in the way people perceive, assess, seek help for, and are treated for pain. In a series of studies published in Pain Medicine (2001-03) and the Journal of Pain (2003), for example, she and colleagues found that physicians are more likely to prescribe optimal pain treatment for men who have metastatic prostate cancer or postoperative pain after a prostatectomy than they are for women who have metastatic breast cancer or postoperative pain after a myomectomy.

In other recent studies, Green concluded that African-Americans across the age continuum are in significantly greater distress than white Americans when they initially seek medical help for pain. Further research is needed, she says, to understand the reasons behind this disparity. Possible causes include ineffective doctor-patient communications, differences in pain tolerance, variable health insurance and financial status, and physicians being less aggressive in providing minorities with pain medications and referrals for specialized pain care. African-Americans in Green's studies generally were covered by health insurance — but they were more likely than whites to be covered by Medicaid or Medicare and also were more likely to say medical care for chronic pain was a significant financial burden.

Green recently led a multidisciplinary team of researchers from around the country, including the U-M School of Public Health, which conducted an unprecedented review of research on factors contributing to racial and ethnic disparities in pain treatment. In a paper published in the September 2003 edition of Pain Medicine , they concluded that racial and ethnic disparities exist in pain perception, assessment and treatment in all medical settings and for all types of pain. They cited one study, for example, which found that white patients arriving in the emergency room with a broken leg were twice as likely as Hispanics to receive pain medication — and the disparity was not explained by the severity of the injury or the patients' gender, primary language, insurance status, or suspected intoxication. Little is known also about the pain experiences of other indigenous people.

Green and her co-authors explored the interwoven collection of causes that may contribute to these types of disparities, some of which involve patient attitudes and doctor-patient communication. Patients with serious medical conditions often under-report the severity of their pain, a phenomenon which directly affects the level of care delivered. This tendency may be more pronounced among minorities for several reasons. Many Hispanic and African-American patients adopt a stoic outlook, research suggests, and subscribe to the belief that pain is an inevitable part of a serious disease and must be accepted. Minorities also may tend to rely on alternative and complementary treatments, prefer to take analgesics only when pain is severe, and have a heightened fear of potential adverse effects of opioid drugs, including addiction, developing tolerance and intolerable side effects.

Studies also have found that minority patients may be less involved than non-minorities in decision-making about their treatment — although they participate more actively when their doctor shares their ethnic background — and that minorities are referred less frequently for specialized care.

Caregiver decision-making can be another factor. Health care providers have not widely employed consistent protocols and guidelines for treating many painful conditions, including back pain, cancer pain, and sickle cell anemia. As a result, clinical decisions are idiosyncratic and widely variable, which may contribute to less than optimal pain care in general, and disparities in care for racial and ethnic minorities in particular.

Access to health services also plays a significant role in pain treatment. It has been shown — notably in a study conducted by researchers from the Mount Sinai School of Medicine and published in the New England Journal of Medicine in April 2000 — that pharmacies in predominantly minority communities in New York City were less likely than pharmacies in white neighborhoods to maintain an adequate stock of opioid analgesics for the treatment of severe pain. In a study presented at a meeting of the American Pain Society in May 2004, Green and her colleagues replicated and extended this study across the state of Michigan and found that pharmacies in minority neighborhoods were less likely to stock opioid analgesics as well. These studies provide examples of the ways in which certain areas are under-served.

Green and her colleagues recommend improved training for health care providers and education for patients. Patients can benefit, they say, by seeking treatment, from information on how to discuss their pain with their doctor, and also from realistic expectations of treatment. Patients who expect pain relief and are willing to take appropriate analgesics may elicit more responsive pain management from their health care providers who, in turn, need to educate themselves regarding pain assessment and treatment and to be willing to listen to as well as elicit patients' pain complaints. The team of researchers also called for increased participation of minority subjects in biomedical research, increased federal funding for the study of health care disparities in vulnerable populations, and increased funding for minority researchers.

Pain, Green says, has an enormous impact on sufferers' productivity, personal relationships and enjoyment of life — and these burdens now fall disproportionately upon minorities. "Pain is one of the top reasons patients visit their physicians and it's the No. 1 reason for disability," she says. "The suffering, the impact on quality of life — it's huge, and we are a better country than this."

The issues raised in Green's research mirror, to a large degree, the agenda of the U-M Medical School's Office of Diversity and Career Development, established in 2002 to create a "workforce and new methodology that will improve the health of minority and disadvantaged populations."

The breadth of the office's focus is unusual among academic medical centers. Pipeline programs encourage talented grade school, high school and college students from historically disadvantaged and under-represented populations to pursue careers in the sciences, health services and biomedical professions. "Our overarching mission is to coordinate medical school efforts to identify and nurture those individuals from groups which are 'under-represented in medicine' as well as individuals from the majority population who will work to eliminate these health disparities," Gordon says.

Recruiting, retaining, and supporting career development for outstanding faculty, staff and students from under-represented groups and diverse backgrounds is critical to that mission. According to Gordon, "Professionals who are under-represented in medicine are more likely to serve disadvantaged populations."

Gordon's team also coordinates staffing to programs that support health care disparities research. "We have programs to promote increased networking among those investigators doing health disparities work and to promote increased interactions with our local communities concerning these unmet medical needs," he says.

These responsibilities commonly are spread among several departments within a medical school, but U-M's concerted, integrated approach is intended to promote knowledge sharing, synergy and diversity, and to generate momentum for initiatives that target disparities in health care.

The Minority Health Research Program, a key collaboration between the General Clinical Research Center and the Program for Multicultural Health, fosters disparities research within the U-M Health System and works to increase the participation of minority and historically underserved populations in U-M clinical research studies. "This program," Gordon says, "has two main goals: to provide educational contacts between our clinical research investigators and surrounding minority communities, and to promote increased inclusion of minorities in our clinical research studies. This makes our investigators better able to include such research participants, while at the same time addressing some of the health information and care needs of our local communities."

Recruiting minorities for clinical trials, Gordon explains, often requires investigators to overcome not only cultural barriers but also patients' suspicion that they are being exploited by a health care system which is otherwise unconcerned with their day-to-day medical needs.

"Our investigators need help with explaining how their research is addressing the unmet health needs these communities face," Gordon says. "Many of our investigators don't even know how to begin such dialogues. So we've established a community advisory board composed of individuals interested in the health needs of our local minority communities. This provides investigators with a ready audience who will honestly critique their research plans, ask representative questions, and provide them with advice on who to contact and what to emphasize or change to better appeal to potential trial subjects."

Researchers from U-M and Wayne State traversed this challenging terrain expertly for a study of diabetes prevalence among Arab-American immigrants in southeast Michigan.

William Herman

William Herman, M.D. (Residency 1982), a professor of internal medicine in the U-M Medical School as well as a U-M School of Public Health professor of epidemiology, investigated diabetes rates among Arabs residing in Egypt in the early 1990s. He found diabetes was virtually non-existent among Bedouin tribes dwelling in the desert, while prevalence climbed to 5 percent in rural villages, 10 percent among low-income people living in cities, and 20 percent among urban-dwellers with higher incomes. The variations, Herman says, appeared to result from lifestyle factors influenced by changing socio-economic status — increased caloric intake, decreased physical activity, and increased obesity, for example.

A decade later and thousands of miles away, Herman and colleagues are studying diabetes among Arab immigrants in Dearborn, home to the world's largest Arabic population outside the Middle East. The team recently unearthed startling and troubling findings: 41 percent of the study subjects suffered from diabetes or other types of glucose intolerance and half the cases were previously undiagnosed.

In a paper in Diabetes Care in 2003, the investigators characterized their findings as a major clinical and public health problem. They surmised that contributing factors may include lack of access to and use of health care, and culturally related attitudes and beliefs including fear of uncovering medical problems. "Community-based intervention programs to prevent and treat diabetes are urgently needed," they wrote.

Subsequent research explored the impact of acculturation on Arab-Americans' propensity for diabetes. Genetic predisposition also may play a role, Herman says, a supposition supported by the fact that other ethnic groups typically do not experience such a pronounced spike in diabetes following emigration to America.

These important findings came to light largely because the researchers exercised great care and sensitivity in working with the community. "We spent a great deal of time preparing the community for the project," says Herman, who credits Wayne State University pharmacy professor and co-investigator Linda Jaber with leading this outreach effort. The team collaborated on the study with Dearborn's Arab Community Center for Economic and Social Services, formed a 16-member advisory board of local leaders, and established a committee of local physicians to ensure that participants with medical problems uncovered by the study received appropriate referrals and follow-up care.

Primary care physicians in the community were contacted and provided information about the project so they could reassure their patients who were asked to participate and be assured the visiting health care providers were not competitors.

The study questionnaire and consent form were translated into Arabic, reviewed for linguistic and cultural accuracy, and field-tested. Questions about religious beliefs and income, and a standard disclaimer describing plans to share study data with government agencies, raised concerns about confidentiality, racial profiling and investigators' motives. Immigrants recruited for the study "often were from a place where that kind of information could be used against a person," Herman points out. The questionnaire and consent form were revised.

Another obstacle was a perception in the community that research amounted to experimentation and had no direct benefit to the participants or the community. Fear of uncovering a medical problem, culture that did not emphasize preventive care, and misconceptions about the seriousness of diabetes and the risks of treatment also generated reluctance to participate.

Investigators countered with an extensive media campaign that relied on the local Arabic newspaper and television and radio stations. They explained the scientific relevance of the study, discussed its methods and procedures, and shared information about the risks of diabetes and the benefits of early diagnosis and proper treatment.

Study participants who viewed the project favorably were asked to help recruit additional subjects, while people who expressed reluctance were immediately contacted by the principal investigator so questions could be answered and concerns alleviated.

The study required subjects to undergo medical testing, creating inconvenience and requiring a significant time commitment. To minimize the burden, researchers accommodated flexible scheduling for clinic visits, including weekend appointments. They also provided transportation and coordinated appointments for relatives, friends and neighbors. The clinic was equipped with a room supplied with magazines, children's books, toys, a television and a VCR.

In the end, the project's participation rate was an astonishing 87 percent. The research team had prepared the ground so well, in fact, that the study endured the tensions and suspicions spawned by the September 11 attacks, which occurred while the research was underway. Work was halted for two weeks, then resumed.

The current task, Herman says, is to better understand the causes of the community's elevated diabetes propensity and design effective interventions. "We don't have a clue whether people will want to pursue lifestyle interventions or medical interventions or both or neither," he says. "The goal is to make a difference — to recognize this problem, understand why it's occurring ... and do something about it."

ALSO:

From Cause to Intervention

Adolescent Girls and Lifestyle

Trust and Understanding in Nueces County

Chronic Renal Disease and African-Americans