© Springer International Publishing Switzerland 2016
Perry M. Nicassio (ed.)Psychosocial Factors in Arthritis10.1007/978-3-319-22858-7_1111. Revisiting Unequal Treatment: Disparities in Access to and Quality of Care for Arthritis
(1)
Gender and Diversity Division, Social Sector, Inter-American Development Bank, 1300 New York Avenue, NW, Mail Stop W0608, Washington, DC, 20577, USA
(2)
Department of Sociomedical Sciences, Mailman School of Public Health, Columbia University, 722 West 168th Street, New York, NY 11530, USA
Keywords
DisparitiesDiscriminationMinoritiesPolicyThe opinions expressed in this publication are those of the authors and do not necessarily reflect the views of the Inter-American Development Bank, its Board of Directors, or the countries they represent.
Introduction
Racism and social discrimination harm health (Krieger, 1999; Williams & Jackson, 2005). That is the conclusion to which health disparities research has come over the last two decades based on the finding that in the United States, blacks, American Indians, and Latinos of all races suffer disproportionately from both chronic and acute illness (Williams & Jackson, 2005). According to the Centers for Disease Control and Prevention (CDC, 2015), American Indians who comprise 2 % of the US population, have the highest prevalence of arthritis (25.2 %) of any racial or ethnic group in the United States (Dall’Era, Snipes, Cisternas, Gordon, & Helmick, 2014). The arthritis prevalence rate is 11.1 % among Latinos, compared with 19.4 % for non-Hispanic blacks and 23.8 % for non-Latino whites. Despite lower prevalence of arthritis, arthritis-attributable limitations are greater among non-Hispanic blacks (44.6 %) and Hispanics (43.2 %) compared to non-Hispanic whites (36.2 %). Blacks, American Indians, and Latinos also report a higher prevalence of activity limitations in work (almost 42 % for non-Latino blacks) and more severe pain than do non-Latino whites with arthritis (CDC, 2015).
Black, American Indian, and Latina women also experience higher morbidity and mortality from autoimmune diseases such as Systemic Lupus Erythematosus (SLE) (Alarcón et al., 1999; CDC, 2013; Society for Women’s Health Research) that, in 90 % of cases, includes some acute or severe arthritic inflammation and resultant activity limitation.1 Among black women in the United States, SLE incidence is almost eight times that of the general population (8.1–11.4/100,000 population per year)2 and Latina, American Indian, Native Hawaiian, and Asian3 women have almost four times the risk of diagnosis when compared with Caucasian women (Domino, 2011; Nass, 2001; Richards, 2003).
The reasons for this increased risk and severity among Black, American Indian, and Latino arthritis patients have been examined in a range of studies, from genetic (Bomback & Gharavi, 2013; Molokhia et al., 2003; Raman & Mohan, 2003; Tsao, 2003) and biomedical (Eudy, Vines, Dooley, Cooper, & Parks, 2014; Kim-Howard et al., 2014; Schulman et al., 1999) to environmental (Williams, Watkins, Anderson, & Tumiel-Berhalter, 2011) and sociocultural (Allen et al., 2010; González, Toloza, McGwin, & Alarcón, 2013). The search for the genetic underpinnings of health disparities for blacks and other minority groups has a long history in the United States. In the late nineteenth century, diseases from which blacks seemed to suffer disproportionately were deemed “black” diseases and were considered evidence of “inherent black frailty,” inferiority, and immutable elements of blacks’ evolutionary lot (Washington, 2006). However, the advent of health disparities research in the 1990s led to an exploration of social and cultural explanations for the disparate health outcomes seen between blacks and whites. Public health research on health disparities reached a fever pitch in the late 1990s and early 2000s following closely the economic, political science, and sociological work on income inequality within and among countries and across social domains including employment, education, housing, health, and social stratification (Beckfield, 2004; Kawachi, 2000; Ryscavage, 1999; Sen, 1992).
In one of the earliest statements defining the health disparities movement, “Social Conditions as Fundamental Causes of Disease,” Link and Phelan (1995) assert that socioeconomic status (SES) , treated largely until that time as a confounding variable in risk factor-based epidemiologic research, must be evaluated contextually rather than tangentially if we are to understand the distribution of disease. In addition to contextualizing, they suggest that researchers take into account the “fundamental causes” of disease, that is, factors that affect access to resources that promote health. These fundamental causes do not change despite the elimination of the mechanisms that link these causes to disease (Link & Phelan, 1995). For Link and Phelan (1995), class (in conjunction with gender and race/ethnicity) is the most important “fundamental cause” of inequality in the United States, deserving of independent attention by public health problem-solvers. They conclude:
Specifically, if the social factor is a fundamental cause, one cannot claim to have accounted for its effects by having “explained” its association with the inclusion of intervening variables in a path or regression model. Second, to understand associations between fundamental causes and disease, medical sociologists need to examine the broader determinants of the resources that fundamental causes entail. This distinctly sociological enterprise will link medical sociologists to the broader discipline in a productive way as we seek to understand how general resources like knowledge, money, power, prestige, and social connections are transformed into the health-related resources that generate patterns of morbidity and mortality. (Link & Phelan, 1995, p. 88)
As a result of these theories and related research, national public health priorities turned toward testing and identifying the central aspects of class, gender, and race/ethnicity that are most relevant to addressing the health of disadvantaged groups in the United States (Link & Phelan, 2000; Marmot, 2004). At the federal level, in 2003 at the request of the U.S. Congress, the Institute of Medicine (IOM) of the National Academies of Medicine issued a landmark report on health inequalities called Unequal Treatment: Confronting Racial and Ethnic Disparities in Health Care (Smedley et al., 2003). The report was striking in that it shed light on disparities between racial and non-English speaking ethnic groups across a wide spectrum of health care including (but not limited to) treatment of cancer, cardiovascular disease, management of chronic diseases such as diabetes, referrals for a variety of tests, and mental health services. The report concluded that there are important differences between racial/ethnic groups in their rates of receipt of medical procedures, even when insurance status, income, age, and severity of conditions were comparable among patients. In addition to identifying multiple disparities in health care, the IOM report highlighted the importance of considering access to health care as a multicomponent concept. Specifically, it described factors that contribute to disparities, such as features of health care systems and settings, of physicians and the clinical encounter, and of patients.
In sum, the IOM report showed clearly that in the United States, racial and ethnic minorities received less optimal treatment and lower quality health services than did whites (Smedley et al., 2003). Later, in 2010, the U.S. Department of Health and Human Services’ (DHHS) Healthy People 2010, “a long-term national agenda aimed at improving health in the United States” (National Research Council, 2004), named elimination of health disparities across gender, race, ethnicity, education, income, geography, and sexual orientation as one of its two main goals.4 Interestingly, at the international level, the Pan American Health Organization (PAHO) also issued goals for the study and eradication of health inequalities based on poverty, education, ethnicity, and gender (Alleyne et al., 2002). Finally, it is also widely understood that the fulfillment of the Millennium Development Goals (MDGs) will require a concerted effort to address disparities in health (Busso, Cicowiez, & Gasparini, 2005).
Chapter Overview
Drawing from some of the levels of influence outlined in the IOM report, this chapter provides a discussion of disparities in health care among people with arthritis by focusing on the clinical encounter and patient factors. We first describe groundbreaking research on physician bias (much of which focuses on cardiovascular disease). We highlight experimental studies and other novel methodologies that include conversation and interactional analyses. We then turn to classic sociological concepts, including power, privilege, and stigma, as analytic tools for understanding these disparities. We propose that the interplay of power, prestige, and stigma create a thorny relationship between physicians and patients. Next, we highlight research and issues relevant to arthritis, with a particular focus on rheumatoid arthritis, osteoarthritis, and systemic lupus erythematosus, common forms of arthritis for which a growing body of research exists on health disparities. Finally, we conclude with research and policy recommendations. Although we acknowledge the population differences inherent in the terms Black and African American, we use these terms interchangeably and in accordance with the terminology employed in the studies we cite. The same convention is adopted for the use of the terms Hispanic and Latino.
Research on Physician Bias
Experimental Studies
In this section, we highlight briefly three notable biomedical and medical sociological health disparities experiments that explored whether physicians’ biases may account for differences by race in use of life-saving interventions. A fourth study (Armbrister, 2014), the only of its kind to focus on arthritis-related disease, is described later in the chapter. Given the strengths of experimental methods in drawing causal inferences, these studies merit mention.
To date, the most groundbreaking clinical work on the effect of racism, sexism, and classism on health has been conducted through studies of cardiovascular disease (CVD) and coronary heart disease (CHD) (Arber et al., 2006; Ayanian & Epstein, 1991; Giles et al., 1995; Schulman et al., 1999). These studies find that women and blacks often are overlooked as candidates for potentially life-saving interventions and receive lower quality care than whites.
In a novel study, Schulman et al. (1999) measured physicians’ propensity to refer patients for cardiac catheterization by patient race and gender. Patient confederates were presented to physicians in video vignettes in which scripts and performances were strictly controlled and all patients had the same income, insurance coverage, and personality affects. Physicians were randomly assigned to watch one of eight videos of “patients,” who varied only in age (middle-aged versus older), gender (male versus female), and race (Black versus White). The study found that physicians were least likely to refer black women for the procedure:
our finding that the race and sex of the patient influence the recommendations of physicians independently of other factors may suggest bias on the part of the physicians. However, our study could not assess the form of bias…. [it] may represent overt prejudice on the part of physicians or…could be the result of subconscious perceptions rather than deliberate actions or thoughts. Subconscious bias occurs when a patient’s membership in a target group automatically activates a cultural stereotype in the physician’s memory regardless of the level of prejudice the physician has. (Schulman et al., 1999, p. 625.
This study and others that followed underline the potential importance of the physician’s own personal characteristics (e.g., length of experience, medical specialty, race, gender, and personality) in his or her decision-making process (McKinlay, 1996; McKinlay, Potter, & Feldman, 1996; Sabin et al., 2009).
Using case vignettes, Sabin, Rivara, and Greenwald (2008) explored whether (1) quality of care would vary by patient race; and (2) strength of implicit race bias would be related to quality care for pediatric patients. The web-based study, conducted with 95 pediatric faculty, fellows, and residents from one department at a research university, presented the Race Attitude Implicit Association Test (IAT); the Race and Compliant Patient IAT; and the Race and Quality of Medical Care IAT as measures of implicit discriminatory beliefs. The IAT is a measure of unconscious bias. Participants were also asked to respond to two explicit bias items: “‘My feelings towards African Americans are…’ and ‘My feelings towards European Americans are…’ (Answer options ranged from 0 = cold to 10 = warm)” (Sabin et al., 2008, p. 680). Finally, participants were presented with two of four possible pediatric case vignettes in which patient confederates were either African American or white. They then assessed whether treatment recommendations for pain control, management of urinary tract infection (UTI) , Attention Deficit Hyperactivity Disorder, and asthma control varied by patient race. Specifically, following the vignettes, participants were asked to indicate the appropriateness of a series of treatment options using a 5-item Likert scale that ranged from: I strongly disagree, this is clearly the wrong treatment option to I strongly agree. This is clearly a good treatment option. One of the treatment options included the ideal care for the case depicted, but none of the options represented negligent or inadequate care.
Their analyses revealed little evidence of overall implicit bias as measured by responses to the IAT tests administered. However, results indicated mild implicit stereotypes about race and compliance skewed toward African Americans, who were conceived as more compliant than European American patients (76 % versus 19 %) and more likely to receive “preferred” medical care (88 % versus 0 %). In addition, the authors also found that for all cases except asthma care, participants identified the ideal treatment recommendation more frequently. In contrast to the earlier work of Schulman et al. (1999), treatment recommendations did not vary by patient race except for management of UTI, for which ideal care occurred more often for African Americans than whites (Sabin et al., 2008, p. 681). The study concluded that the small sample size, use of a convenience sample, a physician sample skewed toward women, and the newness of the Compliant Patient IAT and the Quality of Medical Care IAT may have influenced their findings. In a similar 2005 study, Sabin and Greenwald found that increasing pro-white bias scores of physicians on a measure of implicit attitudes about race were associated with less optimal recommendations for narcotic pain medication for black pediatric patients, but found no other significant treatment associations (Sabin & Greenwald, 2012).
A third study by Green and Carney (2007) asked 202 internal and emergency medical residents in four academic medical hospitals in Massachusetts and Georgia to consider a written patient vignette and photograph of a 50-year-old man suffering chest pain. Participants were asked to read the vignette and indicate whether they would recommend thrombolysis (clot busting techniques) and to specify the strength of their recommendation. Physician participants were then asked to comment on their patient’s personal characteristics (e.g., patient cooperativeness) and were given an explicit discrimination measure and three Implicit Association Tests (IATs) . The study found that physicians exhibited slight pro-white bias and that they indicated the need for thrombolysis equally between black and white patients. The authors emphasized that, given the large body of evidence of black men’s increased likelihood of coronary artery disease, equal treatment constitutes a disparity. They also found that, compared to those with lower IAT scores, respondents with higher IATs (more pro-white bias) were less likely to recommend thrombolysis to black patients. In sum, two of the three studies described above found racial bias in physicians’ decision-making and stereotyping of patients.
Other Methodological Approaches
In complement to the results of the biomedical studies of physician decision-making, studies of patient perceptions of racial bias have also led to the conclusion that physicians’ implicit and explicit biases may play an important role in health disparities. For example, Berrios-Rivera et al. (2006) found that for the RA and SLE study participants both their ethnicity and physician’s communication style (the amount of information provided, sensitivity and patient-centeredness) were associated with satisfaction with the medical encounter. In explanation, Berrios et al. suggest that “perhaps physicians (and/or patients) have subconscious biases, stereotypes, perceptions, or misconceptions…that interfere with open communication, empathy, and the development of trust” (p. 391). They base this hypothesis on the same studies of cardiovascular disease (CVD) and coronary heart disease (CHD) summarized above.
Cruz-Flores et al. (2011) reviewed the stroke care literature from 1972 to 2010 and found differences in the burden of stroke risk and stroke care between racial and ethnic minorities in the United States and whites. The disparities include among the myriad explanations for these disparities patient mistrust and physician bias. They explain that “differences in socioeconomic status and insurance coverage, mistrust of the healthcare system, the relatively limited number of providers who are members of minority groups, and system limitations, may contribute to disparities in access to or quality of care” (Cruz-Flores et al., 2011, p. 2103).
Another type of study, led by epidemiologists using some microsociological tools (e.g., conversation and interaction analysis), has sought to examine physician–patient communication in depth. These studies analyzed differences in physicians’ interaction styles and behaviors with black and white patients, finding that many times, physicians’ conversations with black patients differed in content and quality from their conversations with white patients. Specifically, physicians were more likely to lecture black patients and “adopt a narrowly biomedical communication pattern with African Americans, a pattern associated with low satisfaction ratings for patients as well as doctors” (Ashton et al., 2003).
Power, Prestige, and Stigma
The literature on social interactions in the medical encounter offers some explanation for physicians’ behavior, implicating power imbalance and stigma as the main driver of continued health inequalities in medical treatment. Talcott Parsons (1951), for example, presents illness not only as a biological phenomenon, but also as a social role (“sick role”) whose boundaries are dictated by social expectations. Specifically, the “sick role” is an expression of social deviance with the express purpose of exempting the sick person from her prescribed social duties for the duration of the illness, absolving her of responsibility for the illness. To perform the role successfully, the patient must additionally desire to get well, seek professional help in recovering from the illness and cooperate with the prescribed treatment. In Parson’s conception, the power differential between physician and patient is needed to absolve the patient of her responsibility for accomplishing her duties.
Another branch of the power literature focuses on relations between patient and physician as a reflection of traditional power structures (Conrad, 1992, 2001; Pappas, 1990; Todd, 1989), the “embodiment” of culture and the established moral order (Alexander, 1982; Farmer, 1992; Kleinman, Eisenberg, & Good, 1978; Krieger, 1999). Farmer (2005) views this “embodiment” as a reflection of structural violence. Social constructionists legitimize explanations for disease based on the lived experiences of the sufferer that are often at odds with biomedical findings and methods that purport objectivity (Taussig, 1980).
Swidler (1986) takes the analysis of social interaction and its manifestation in culture in a different direction—away from the power of individuals or groups and toward the power of culture itself. She analyzes the relationship between cultural influence and action offering a view of culture as “a ‘tool kit’ of symbols, stories, rituals, and world-views which people may use to solve problems (Swidler, 1986). The tool kit imagery serves to emphasize that the organizing functions of culture are not uniform…but actors select relevant tools to construct their behaviors.
Prestige and status are often used interchangeably to indicate “the esteem, respect, or approval that is granted by an individual or a collectivity for performances or qualities they consider above average…[p]restige (negative or positive) may also be granted for qualities with which the individual is endowed at birth, such as nobility, membership in an ethnic group, or perhaps musical genius” (Goode, 1978). Prestige can also be earned by granting the “right kind of approval” to others (i.e., learning from the response of those around us that we have reacted “correctly”).
At the same time as societies grant praise and esteem, they also remove it through dispraise and disapproval. Dispraise can be communicated through gesture, facial expression or verbally, and has perceptible effects. In fact, social psychological studies show that the more an individual’s identity falls below the norm set by those in interaction with him or her, the greater the frequency of any performance below the norm (Goode, 1978). This cycle spirals downward as it repeats, increasing the likelihood of receiving more direct dispraise and the probability that others will join in the heaping on of disapproval. The repetition of low performance also increases the chance that “other sanctions or controls will be applied against the offender…[and] that the criticism will contain moral overtones or be expressed in moralistic rhetoric; and that the critic will display anger” (Goode, 1978). Steele and Aronson (1995) have termed this phenomenon stereotype threat, the feeling that one is at risk of confirming stereotypes of one’s group and the consequent in-group underperformance when reminded of these stereotypes.
In sum, prestige (esteem) and its converse, disapproval, function to control and correct behavior, both of the criticized and of the group to whom the criticized belongs. In this social order, this type of negative control is applied inequitably: the more powerful (socially superior group) being less likely to receive strong overt sanctions than the less powerful (socially inferior group) (Goode, 1978). Furthermore, many times sanctions are based solely on stereotypes about a socially inferior group. For example, as Rubio and Williams (2004) assert, negative stereotypes about Blacks and other minorities “are likely to have profound implications in situations ranging from personal day-to-day interactions to public policy. Research on stereotypes indicates that the endorsement of negative racial stereotypes leads to discrimination against minority groups” (Rubio & Williams, 2004).
Closely related to the concepts of prestige or status is stigma, a phenomenon that occurs when the labeling of human differences results in negative stereotyping and the categorization of people into “us” and “them” (Link & Phelan, 2001). Not only do societies recognize and shy away from the stigmatized, they actively reduce the life chances of the stigmatized person (Goffman, 1963).
Kwate et al. (2006)) and others argue that racism is a form of stigma for two reasons: (1) “blackness” and black people are linked to a set of undesirable characteristics; and (2) the experience of residential and social segregation constitutes a de facto demarcation of “us” and “them.” Stigma is revealed in several pervasive, though neither necessarily logical nor consistent, stereotypes. These include among others: the ascription of inferior intellects to black people that “in school settings…underlies peer perceptions of undeservedness, inequalities in educational placement, and attacks on affirmative action”; the ongoing message that black people, especially black men, are dangerous and prone to criminality; the perception of blacks as lazy and uninterested in work; that black people, especially black women, are servants; and are able to “withstand extremes of heat, physician labor, and pain.” Recent studies suggest the continued reliance on this latter stereotype in medicine, as African Americans are still less likely than whites to be prescribed analgesic medication (Heins et al., 2006). Moreover, the application of these stereotypes in and of themselves can harm the health of African Americans as “stereotypes of African Americans fuel a moral economy that not only takes an intrapsychic toll on Black minds, but also subtly interweaves policies and practices that subordinate Black people” (Kwate & Meyer, 2011).
We propose that power, prestige, and stigma converge to create a thorny relationship between physicians and patients. Physicians, in a place of authority and thus greater prestige, have and are exhorted to gain the compliance of patients, perhaps members of a stigmatized racial or ethnic group to which all manner of negative stereotypes have been ascribed (Zola, 1972).
Research on Arthritis Populations
In the following sections, we provide an overview of two overarching classifications of direct and indirect causes of disparities in the sociocultural arm of arthritis, research in which issues of power, prestige, and stigma contribute to the creation and perpetration of health disparities—studies focused on patient-level factors (such as medical mistrust, adherence to medications, and health literacy) and research that highlights elements of the clinical encounter (focusing on physician biases in medication prescriptions and other forms of treatment). Given recent reviews of research on disparities in arthritis (see McBurney & Vina, 2012; McIlvane, 2009), we selected issues which we see as key for further research.
Related posts:
Sleep Disturbance in Rheumatic Disease
Mechanisms of Arthritis Pain
Socioeconomic Disparities in Arthritis
The Importance of the Biopsychosocial Model for Understanding the Adjustment to Arthritis
Physical Activity and Psychosocial Aspects of Arthritis
The Nature, Efficacy, and Future of Behavioral Treatments for Arthritis
Stay updated, free articles. Join our Telegram channel
Full access? Get Clinical Tree
