Economic Grand Rounds: Social Inequality and Health: Will Psychiatry Assume Center Stage?

The linkage between social inequality and physical health has been recognized since the 19th century. The association between social inequality and mental illness was first elaborated in the mid-20th century (1), inspired by the work of Faris and Dunham, Hollingshead and Redlich, and Srole and colleagues. By the 1970s, interest had waned because of limited analytic tools, underdeveloped interventions, and a political climate that was not conducive to addressing issues of social inequality (2,3).

However, social inequality has emerged as a focal point of health concerns in the 21st century. The British government's new health policy recognizes that "the root causes of ill health are mostly social, economic, and environmental and require policies that target help at those who are worse off" (3). In the United States, the Surgeon General has identified "health disparities" as a major public health concern (4).

Nevertheless, in psychiatry, issues pertaining to social inequality have received comparatively little attention. The requirements of the Accreditation Council for Graduate Medical Education for psychiatry do not mention the terms "poverty," "social inequality," "disparity," or "social class," although "socioeconomic status" is among a long list of items to be included in the curriculum. Between 1993 and 2002, of around 20,000 articles about mental disorders that were listed by MEDLINE, fewer than 400 mentioned poverty, social inequality, or social class. In articles that mentioned the latter, social class indicators were typically controlled for rather than studied for their effects on health (2).

Social inequality refers to economic and social disparities in a population in terms of various indicators such as income, wealth, education, occupation, social class, and deprivation. These indicators are often encompassed under the theoretical concepts of "social stratification" and "social class."

There are compelling reasons for psychiatry to begin a serious engagement of social inequality. First, such engagement can promote linkages between psychiatry and other areas of medicine by highlighting common social factors in the etiologies of mental and physical disease and by demonstrating psychiatry's critical role in the prevention and treatment of many physical disorders. Second, such an approach can reinvigorate one of community psychiatry's key missions of extending its therapeutic role to the community. Finally, it can stimulate research and innovative treatment methods for addressing the psychosocial effects of social inequality at the individual level.

Social inequality, medicine, and psychiatry

In advanced industrial countries, despite improved material conditions for all population strata, there remains a strong inverse pattern between health and social class. While this association is more pronounced for impoverished groups, some association can be seen at all socioeconomic levels. Moreover, it appears that income distribution is more strongly related to mental and physical health and longevity than is per capita income (5).

For example, Wilkinson (5) calculated that greater life expectancy is found in those European countries in which the least well-off 70 percent of the population receive a greater proportion of the earned income in their country. This finding may be particularly critical in the United States, where wealth has become increasingly maldistributed and social mobility has been stagnant. Recent data from the United States have shown a strong relationship between income inequality and mortality that has not been observed as consistently in other developed countries (6).

Two key concepts have emerged from these epidemiological findings that suggest several potentially fruitful linkages between psychiatry and medicine. The first, the notion of "fundamental causes," was elucidated by Link and Phelan (7). They proposed that regardless of what the profile of diseases and known risks happens to be at any given time, persons with greater access to critical economic and social resources will be less affected. As Cassel (8) noted, "A remarkably similar set of social circumstances characterize people who develop tuberculosis and schizophrenia, become alcoholic, are victims of multiple accidents, or commit suicide. Common to all these people is a marginal status in society." Marmot (9) believes that this phenomenon reflects the impact of both material resources, such as education, nutrition, housing quality, and occupational health, and psychosocial resources, such as social ordering and empowerment, social supports, and community cohesiveness.

The second concept, "epidemiological transition," is based on Wilkinson's (5) observation that advanced industrialized countries have seen a shift from direct material causes of disease to indirect psychosocial pathways. It is now postulated that "emotions provide the missing link between personal troubles and broader public issues of social structure" (10). That is, social and economic problems affect health indirectly through various forms of worry, stress, insecurity, and vulnerability, thereby making the mind the crucial gateway through which social influences affect physiology to cause disease. Thus, paradoxically, psychiatry's cross-fertilization with general medicine may rest more firmly on its social affinities than on its biological ones.

Despite the well-established principle that social inequality may affect psychiatric illness, interest in this topic has been largely dormant. Unlike other health disciplines, psychiatric theory remains bogged down between two competing individual-focused models in which socioeconomic differentials associated with mental disorders are viewed as an effect (social selection, or "drift") or as a cause (a social stressor). Dohrenwend's review (11) showed that both processes play a role in most mental disorders, although for some disorders, such as depression and alcoholism, social causation has proportionately more impact, whereas for others, such as schizophrenia, social selection may be more important. Regardless of causality, studies have consistently shown that socioeconomic factors affect the course and outcome of mental disorders.

Social inequality and pathogenesis

Recent models explicating the relationship between social inequality and the pathogenesis of mental and physical disorders have focused at the individual level—for example, network supports and participation, level of personal control, and neurobiology—or at the social level, such as organizations, communities, and mutual cooperation (9). Indeed, an evolving conceptual assumption among theorists is that inequality must be examined at both the individual level and at the community level, although an integration of these two levels has not been realized (9). A potential unifying point for these models is emotion, because it is a hybrid category that incorporates biological, psychological, and social categories and thereby occupies a central place in linking the social world to the body and mind (10). Five models are described below.

First, social inequities may affect personal health through endemic stress. Fried (12) defined endemic stress as persisting scarcity, enduring conditions of loss or deprivation, and continuing experiences of inadequate resources or role opportunities. He suggested that endemic stress produces subtle and subclinical manifestations of apathy, alienation, withdrawal, affective denial, decreased productivity, and resignation. Turner and Lloyd's (13) studies of "cumulative adversity" have confirmed Fried's hypothesis. These studies showed that accumulated lifetime traumatic experiences correlate with increased psychological distress and mental disorders.

Second, at the physiological level, an increased vulnerability to stress as a result of early or accumulated life experiences may be mediated by a persistent alteration in neurobiological systems that are known to be responsive to stress, such as the hypothalamic-pituitary-adrenal axis, corticotropin-releasing factor systems, and immune function. Recent evidence indicates that stress-induced steroid production can produce hippocampal atrophy, loss of hippocampal feedback, and decreased neurogenesis (14). Thus chronic stress may produce structural and physiological changes that predispose individuals to depression, anxiety, and stress-related physical disorders, such as gastrointestinal disorders and heart disease (15).

Third, Wilkinson (5) believes that nonegalitarian socioeconomic structures are reflected in problems with social cohesion (or "social capital")—deficiencies in social trust, reciprocity, and concern for the well-being of one's community. Social cohesion is defined as participation in public affairs, civic responsibility, and involvement in public life and can be assessed by such indicators as whether a person votes, whether he or she reads a newspaper, and the number of volunteer associations to which a person belongs. Diminished social cohesion may create feelings of powerlessness, stress, or hopelessness.

Empirical support for this framework has come from the British Household Panel Study (16), in which respondents who categorized their neighborhoods as lowest in social capital measures had a greater risk of psychiatric morbidity, whereas respondents who categorized their neighborhoods as highest in social disorganization measures had the greatest risk of physical illness. Similarly, Dalgard and Tambs (17) found that as improvements occurred in the social environment in a low-functioning neighborhood in Oslo, Norway, so too were improvements in mental health noted that could not be explained by selective migration. In a U.S. study, Diez Roux and associates (18) found that coronary heart disease was significantly associated with living in a disadvantaged neighborhood.

Fourth, social networks have been identified as being able to buffer against life stressors (19). Working-class populations are often highly connected to social networks, whereas those who are very poor typically have small networks (19). Middle-class networks tend to be larger and diffuse, a configuration that facilitates vocational, financial, and recreational opportunities. Chronic illness often puts a strain on the members of a network and may have an especially pronounced effect on the networks of the economically disadvantaged (15).

Finally, deficits in social ordering and empowering may help to explain why income distribution may be more critical to health in advanced industrial countries than is per capita income per se (9,19). Indeed, Ilfield's (20) study found that adults in the lowest income category were three times as likely as those in the highest income category to express low levels of self-esteem and self-efficacy. British researchers found that having little control over workplace activities was an important risk factor for coronary heart disease and depression (9). Primate studies, in which material differences between animals were not appreciable, have likewise supported the association between subordinate status and both increased risk of coronary heart disease and higher serum cortisol concentrations (9).

Implications

The epidemiological transition in which socioeconomic factors have come to affect health primarily through psychosocial pathways rather than direct material pathways provides an opportunity for psychiatry to assume center stage in addressing the effects of social inequality. Although psychiatry has neglected the problem of social inequality, partly because of the absence of a solid research and clinical base, the development of new causal models, assessment tools, and clinical methods means that now is a propitious time for psychiatry to seriously take on this problem.

First, at the community or group level, community psychiatrists can play a role in augmenting social capital by using novel organizing techniques to work with neighborhoods to strengthen community bonds and resources (21) and by using social network intervention strategies that have been developed for supporting and strengthening social bonds (19).

Second, at the individual level, clinical assessments might include an "ecoanalysis" that examines clients' economic and ecological contexts (22). This Meyerian approach entails examining the economic conditions of a client's childhood, the wealth of the client's neighborhood, the parents' social class, and the client's work history, current occupation, and educational level as well as recognizing that income fluctuates over time.

An ecodynamic analysis also entails differentiating between proximal and distal causes of disorders (7). Thus psychiatrists must extend their analyses beyond immediate factors that suggest biological causality to examine the social contexts of these factors. For example, what are the antecedent social factors that make a hallucinating person with schizophrenia more likely to become homeless? To probe these issues, clinicians can supplement their assessments with instruments that have been developed to examine financial strain, daily hassles, social class, and wealth (2).

Considerable work must be done at the individual therapeutic level to address the effects of social inequality. It is not clear whether stress management techniques, techniques that are used with persons who have posttraumatic stress disorder, or an amalgam of the two techniques would be beneficial, nor whether novel therapeutic strategies should be developed. In Britain, the Critical Psychiatry Network has sought to complement biomedical explanatory models with interpretations of symptoms on the basis of a person's social perspective, experience, and context (23).

Krieger and colleagues (2) observed that the longer lives and superior mental and physical health of people who are "better-off" demonstrate what is attainable for all persons. Psychiatry can play a central role in the realization of that goal.

Acknowledgments

The author thanks Kenneth Thompson, M.D., Carles Muntaner, M.D., and Barbara Singh.

Dr. Cohen is professor of psychiatry at the State University of New York (SUNY) Health Science Center, Box 1203, 450 Clarkson Avenue, Brooklyn, New York 11203 (e-mail, [email protected]). A version of this paper was presented at the annual meeting of the American Psychiatric Association held May 5-10, 2001, in New Orleans. A more extensive bibliography is available from the author. Steven S. Sharfstein, M.D., is editor of this column.