A Conceptual Framework for Advancing Melanoma Health Disparities Research

Health disparities may be defined as “differences in incidence, prevalence, morbidity, mortality and burden of diseases and other adverse health conditions that exist among specific population groups in the United States.”^{1 (p7)} In this issue of the Archives, Hu et al² retrospectively compared the incidence of melanoma for white non-Hispanics, white Hispanics, and blacks. Their study detected significantly more advanced melanomas at presentation for white Hispanics and blacks in the state of Florida than for white non-Hispanic patients. This study adds to a growing body of literature that identifies a pervasive and persistent disparity in the stage of melanoma diagnosis between Hispanic and black patients vs white patients.

Hu et al had previously reported that late-stage melanoma diagnosis was more common among Hispanic (26%) and non-Hispanic black patients (52%) compared with non-Hispanic white patients (16%) (P < .001) in the Florida Cancer Data System (Miami-Dade County) from 1997 to 2002.³ A separate group of investigators reviewed the New Mexico Melanoma Registry, which like Florida has a greater proportion of Hispanics than is represented in the Surveillance, Epidemiology and End Results (SEER) program, and found that Hispanics had poorer prognoses than whites (1970-1986).⁴ A third study conducted in California found that 15% of Hispanic male patients in California were diagnosed as having melanoma after it had metastasized compared with 6% of non-Hispanic white male patients.⁵ Thus, 3 state registries with large Hispanic populations document later presentation for care. Survival is improved with early detection of melanoma, with the majority of survivors receiving treatment at stage IA. The survival rate 10 years after diagnosis was estimated as higher than 95% in those diagnosed as having stage IA melanoma and declined to less than 60% when the diagnosis occurred at a later stage (stage IIB or C). People who are diagnosed later rapidly progress to the advanced stage of the disease with metastasis to internal organs (stage IV).⁶ If the current trend of late-stage presentation remains unchanged, ethnic minorities will continue to disproportionately bear the burden of melanoma-related morbidity and mortality.

Elimination of health and health care disparities among the different segments of the population is one of the top 10 national health objectives for the US Department of Health and Human Services. Why? Because disparities carry a high cost at both the personal and societal level. At the individual level, a poor health outcome may lead to significant changes in productivity (loss), emotional distress affecting both patient and loved ones, disability, and mortality. Distal costs follow for society, including increased hospitalization costs, expense of medications and/or treatments, avoidable complications, and loss of work days/productivity.⁷ The effects of health disparities eventually affect all Americans.

Substantial resources have been allocated to reduce disparities including the passage of the Minority Health and Health Disparities Research and Education Act 2000, Healthy People 2010 (a comprehensive, nationwide health promotion and disease prevention agenda), and creation of the Office of Minority Health. Yet despite these efforts, comprehensive data collected in the National Healthcare Disparities Report developed by the Agency for Healthcare Research and Quality (AHRQ) confirms that significant inequalities continue to exist irrespective of medical condition, site of care, and/or point of health care intervention for racial and ethnic groups as well as those of poor socioeconomic status.⁸

Complex associations and interactions of environmental, cultural, behavioral, social, and biological factors contribute to this disparity. Research is needed to elucidate the role of the various components and identify effective intervention points for implementation of health policy programs. A multilevel framework must be developed to improve detection, understand why a disparity exists, and reduce the incidence of melanoma. A framework could be built on the epidemiologic work of Last,^{9 (p42)} which is defined as “the study of the distribution and determinants of health-related states or events in defined populations, and the application of this study to the control of health problems.”¹⁰ The identification of the affected vulnerable population and medical condition as well as valid measurement tools are components of the detection phase. Understanding why disparities exist as well as identifying factors responsible for health and health care differences constitute the second phase of research. Phase 3 involves development, implementation, and evaluation of interventions and their outcomes (Figure 1).¹⁰ Dermatologists need not only to be on the forefront of providing care to all those at risk to develop melanoma but also to serve them by participating in all phases of this research effort.

The work of Hu et al published in this issue adds melanoma to the list of documented health disparities in ethnic minority populations that includes asthma, cancer, diabetes, and cardiovascular disease among others.^{11 - 14} This group's observation of worse melanoma outcomes as well as that of other researchers provides our specialty with the foundation of the first phase of disparity research and should be the impetus for continued efforts. The concept of a health disparity seems straightforward, yet there is a surprising lack of consensus in the medical literature on some of the defining characteristics of a health disparity. Most researchers share the sentiment that the definition encompasses differences in the health between the population groups in question, and these differences are “significant” in magnitude.¹⁵ Disparities in vulnerable populations are thought to occur in groups who have “historically experienced ongoing bias and discrimination.”^{10 (p2115)} Many models of health disparities focus on the effects of race and racism on health.¹⁶ Dermatology researchers also appear to have defined the vulnerable population as ethnic minorities, but race may be a proxy for poor socioeconomic status, lower educational experiences, and occupational hazards (ie, agricultural workers and high levels of UV light exposure), among other factors. Our specialty should remain flexible in defining vulnerable populations for melanoma, since it may evolve from its current form as we develop a better understanding of this disease and its effects on the underserved and its subpopulations. Selection bias or errors that may confound findings must be addressed to prevent inadequate or incorrect health disparity care outcomes. Proper identification of all Hispanics and non-Hispanics in study populations as well as stratification of the wide range of pigmentation that occurs in Hispanics into high-risk and low-risk skin types are areas for further investigation. Melanoma incidence data are difficult to isolate for Hispanic patients prior to the 1990s because many cancer registries (including SEER) did not separately track this patient population. Continued tracking of this relatively new disparity is imperative.

Once a disparity is identified, understanding potential determinants is essential to formulate interventions to reduce differences. It is difficult to unravel the influence of an individual patient's biological, cultural, and environmental factors, along with health care system and provider and last but not least, clinical encounter issues and how they may contribute to a health disparity. We will divide the discussion in this section into 2 groups of potential health disparity determinants: (1) individual factors and (2) health care system factors (Figure 2). Issues that affect the clinical encounter are encompassed by the discussion of physician and health care provider factors and cultural competence sections herein.

There are several recognized melanoma subtypes that display distinct biological levels of aggressive behavior, age at onset, and increasing evidence of unique genetic etiology. Anatomical locations of cutaneous melanomas have shown glaring discrepancies between ethnic groups. Whites develop more than 90% of melanomas on sun-exposed areas of the trunk (men) and legs (women). However, in blacks, Hispanics, and Asians, melanomas more often arise on non–sun-exposed skin, particularly in acral areas.¹⁸ Along with varied clinical forms, several key genetic alterations have also been identified. Melanomas harboring BRAF gene mutations appear to be highly prevalent (59%) on skin without chronic sun exposure.¹⁹ Further research into the role that inherent genetic risk factors may play in melanoma disparities is needed.

Factors that often translate into a higher position in the economic system such as educational level, occupation, income, and health insurance coverage and benefits are associated with better health.¹⁶ Limited financial resources strain family resources and often require a choice between competing needs such as food and rent or mortgage vs health care. Lower socioeconomic status may also alter factors such as the availability of transportation (ie, public vs automobile), and living in high-crime neighborhoods may adversely limit community resources. Lower educational levels directly affect literacy and reduce the options for health education, ie, office handouts or instruction sheets that are not reviewed verbally with patients and rely solely on reading comprehension are ineffective for those with low literacy skills.²⁰

A recent skin cancer awareness report targeting Hispanics found that none had ever been taught how to perform skin self-examination and most Hispanics perceived themselves as not being at risk for the development of melanoma.²¹ Delay in melanoma diagnosis may have its roots in cultural beliefs and the patient's social network (ie, only whites get skin cancer, patient does not know anyone with skin cancer or have family members affected by skin cancer.) Skin self-examination is readily adopted by those with a history of melanoma, but is not likely to be readily performed in a population who believes their risk of skin cancer is virtually nonexistent.

Primary care physician (PCP) beliefs must also be addressed because they can provide important health care surveillance. Prior research indicates that skin examinations are infrequently performed by PCPs in general population.^{22 - 23} This lack of surveillance at both the individual and primary care level misses an important opportunity for surveillance for the early detection of skin cancer.

There are 2 main factors that may affect vulnerable populations in regard to patient-physician interaction: communication and conscious or unconscious stereotyping or racism. Ethnic minorities are less likely to be referred by their physicians for diagnostic procedures than whites despite reporting the same symptoms and controlling for socioeconomic factors and access to health care.⁸ Physicians have also been found to use a more dominant verbal style of address with African American patients than with whites in taped conversations.²⁴ Whether these differences are due to poor communication skills or stereotyping remains unclear.

Health care providers may not take into consideration cultural differences between patients and tailor health messages appropriately. Ethnic minorities often do not feel included in the decision-making process and may not attend for further treatment owing to mistrust.²⁵ All physicians need to be mindful regarding stereotyping and personal bias and continue to work toward social justice in our health care system.

The growing diversity of the nation calls for an improved understanding of culture and an effort to decrease language barriers to improve communication between physician and patient.¹⁶ The Liaison Committee on Medical Education mandated that all medical colleges include cultural competency in the curriculum in an effort to improve medical professionals’ ability to care for a culturally heterogeneous population. Improved cultural understanding should result in improved physician-patient communication. Reduction of language barriers and increased cultural competence should improve communication and assist in the reduction of patient trust issues and suboptimal care delivery.

There is a significant underrepresentation of ethnic minorities in many of the health care professions including physicians, nurses, and technicians. Increasing representation would assist in diminishing the problems of language barriers and culturally sensitive health care.²⁶ Continuing to provide opportunities for mentoring and increasing opportunities for exposure to possible careers in the health services may be invaluable.

Ethnic minority populations are overrepresented in government-provided health plans such as Medicaid. Efforts at cost containment for these federal- or state-funded programs may result in restricted access to specialty care including dermatological treatment.²⁷ Limited ability to access dermatologists may result in later stage presentations.²⁸ High-risk sites in ethnic minorities, ie, acral skin, may not be examined and an early opportunity for diagnosis and care may be missed.

Another adverse effect of budgetary cutbacks is neglect or closure of the public health infrastructure. As the need for publicly funded health services such as community health clinics and hospitals grows, many of these facilities located in underserved communities are eliminated to reduce expenses.²⁷ For patients with limited resources (ie, transportation), this limits access and reduces the potential for a patient to be seen for treatment early in the course of their disease.

Widespread mistrust of the health system is nowhere more evident than in health research, where participation remains low. Minority underrepresentation in research is not surprising given the history of past research abuses, including the most famous example, the Tuskegee Syphilis Study. Distrust, knowledge regarding health research, and the process of informed consent are all barriers to increasing recruitment of minorities into research studies and clinical trials.²⁹ The sequencing of the human genome has revealed much genetic diversity heralding the coming age of molecular profiling for individualized disease management. This is especially promising in clinical oncology, where targeted therapy should help maximize drug efficacy and minimize toxic effects. Disparate interethnic clinical presentations and melanoma tumor behavior are likely rooted in gene expression differences and germline susceptibility associated with ethnicity. Greater participation in research is of vital importance if minorities are to benefit equally in the promise of targeted treatment protocols for melanoma.

Interventions that do not specifically focus on the needs and characteristics of a minority community may have little impact on disparities. Often interventions are designed using technologies or approaches that may have limited access by socioeconomically disadvantaged individuals or special populations, such as elderly patients, or that may have limited appeal to special populations. Interventions that are tailored to specific subpopulations and that involve community members in the design and dissemination may have the potential for enhanced reach and acceptability. Community outreach and participatory research programs have been used with success in human immunodeficiency virus education and in promoting healthy cardiac health behaviors.^{30 - 31} Community participatory research approaches in which the target population has an investment and input into the design, recruitment, and evaluation of research may help to overcome fear and mistrust of professional organizations and language barriers and create links between the community and health services. Using residents to assist others in the same community often motivates participants to help create positive changes for their peers and neighborhood.

Fear and limited understanding of how to access health services often cause delays in ethnic minorities obtaining appropriate and timely medical care. Organized educational health communications via mass media or other trusted community leaders and organizations could be helpful in disseminating information regarding healthy behaviors. Dermatological societies must carefully tailor their messages to resonate with their target audience. Substantial changes are needed in current skin examination and cutaneous cancer campaigns because most Hispanics and African Americans do not believe they are at risk for skin cancer.²¹ Influencing individual health beliefs and behaviors are an important factor to consider in melanoma. Modifying attitudes about high-risk behaviors (ie, sun exposure) and susceptibility to skin cancer could result in earlier use of health services. Linguistic and conceptual preferences must be incorporated into any campaign attempting to address these populations.

All interventions and outreach programs that attempt to reach ethnic minorities and improve health knowledge should have preintervention and postintervention outcome measures to provide evidence of the effectiveness of their programs. Systematic data collection and analysis of changes in attitudes and health behaviors in the target population must be assessed to demonstrate the impact of the program. This would allow investigators to critically review and modify or eliminate any poorly received or underperforming research protocols.²⁷

In the absence of a significant advance in the treatment of late-stage or metastatic melanoma, the best opportunity to decrease the death rate is enhanced surveillance. The field of dermatology has begun to compile data regarding melanoma disparities between whites, blacks, and Hispanics. It is likely that no single risk factor will be responsible for this growing problem. As our specialty moves forward, it is of vital importance that we not relegate our efforts exclusively to surveillance and database reviews but move quickly in an attempt to intervene. An effective education and outreach model that transcends cultural and language barriers must be formulated. This article offers suggestions for a number of potential points of investigation and intervention (Figure 3). It is important for physicians, researchers, and the general public to realize that disparities are not inevitable. All populations groups deserve equal access, equal care, and an equal opportunity to enjoy good health.

Correspondence: Dr Hernandez, Department of Dermatology, University of Illinois at Chicago, Mail Code 624, 808 S Wood St, Room 376 CME, Chicago, IL 60612 (claudiah@uic.edu).

Financial Disclosure: Dr Hernandez is or has been a speaker, investigator, and consultant for Genentech, Amgen, Centocor, Biogen Idec, and Abbott Laboratories.

Funding/Support: Dr Hernandez is a Dermatology Foundation Women's Health Career Development Award recipient.