6210 Week 8 Discussion
COMMENTARY
Minority Group Status and Healthful Aging: Social Structure Still Matters
During the last 4 decades, a rapid increase has oc- curred in the number of sur- vey-based and epidemio- logical studies of the health profiles of adults in general and of the causes of dispar- ities between majority and minority Americans in par- ticular. According to these studies, healthful aging con- sists of the absence of dis- ease, or at least of the most serious preventable diseases and their consequences, and findings consistently reveal serious African American and Hispanic disadvantages in terms of healthful aging.
We (1) briefly review con- ceptual and operational def- initions of race and Hispanic ethnicity, (2) summarize how ethnicity-based differentials in health are related to social structures, and (3) empha- size the importance of atten- tion to the economic, politi- cal, and institutional factors that perpetuate poverty and undermine healthful aging among certain groups. {Am J Public Health. 2006;96: 1152-1159. doi:10.2105/AJPH. 2006.085530)
Jacqueline L Angel, PhD, and Ronald J. Angel. PhD
ALTHOUGH THE SUPREME
Courl outlawed the principle of sepajate but equal in 1954 with its famous Brown versus Bom-d of Education decision, many mi- nority y^mericans luul that they are still separate and unequal. Despite a century of impressive innovations in medical science and improvements in public health, poverty continues to un- dermine the pliysical and emo- tional health of a large number of Americans, and serious ra- cial/ethnic health disparities persist'"^ Low-income families have inadequate healtli care coverage,"'^ and individuals who lack adequate insurance are more likely to die from cancer and other serious diseases be- cause of late diagnoses and defi- cient care.^"" Perhaps the most basic question is wliether health disadvantages among minority Americans are the direct and almost complete resuit of pov- erty and its correlates. Well- documented correlates include low educationai levels, labor force disadvantages, and resi- dential segregation iii ghettos and barrios, where individuals are exposed to environmental and social health risks such as drugs. \'io!ence. and fainily disruption.'"^" ̂ ''
Radal/ethnic disparities in mor- bidity and mortality are so glaring that the federal govemment has been forced to respond, and a large body of research has exam- ined tlie role socioeconomic status (SES) and ailture play in these disparities.'̂ The ultimate goal Ls to identiiy the sodal stuictural
causes of inequities in health so that genera] population health can be impn)ved. We will present ap- proaches to studying radal/etlinic health disparities hy (1) reviewing operational definitions of race and ethnicity and tlie research tools tliat estimate difierential disease burdens and health au'e use, (2) assessing jast how far the field has come in understanding healtli. and (3) |iro]X)sing a future re- search agenda that examines the sodal, economic, and [xilidcal foires tliat peipetuatc health vulnerabilities.
GROUP CLASSIFICATION
Duriiig the past 2 decades, we have witnessed an increasing appreciation for the conceptual complexity of gi-oup dassification and its potential for intiTxiudng bias into studies of comparative health levels.'̂ Individuals can be of mixed race/ethnidt>', tliey can intermarry and identify with an adopted group, and they can even ivject a group clas.sification. pai- ticularly if that identity is imposed by others. Individuals who strug- gle e^aiiist the sodal stigma asso- dated with group dassification often embrace that identity as a political statement and a sign of defiance. Standard classifications of race/ethiiidty do not overlap with spedfic genetic profiles or at- tributes.'^ To a large extent, sudi classifications are political cate- gories defined by history ajid the sodal vulnerabilities imposed on minority gi-oups by the dominant majority.'" A political basis of gi-oup classification does not
translate directly into useful sden- tific or intellectual classification.'''' We no longer differentiate among non-llispanic White nationality groups, because distinguishing whidi nation an individual's an- cestors came from is no longer relevant. According lo Richard Alba, Americans of all European ancestries have come to be viewed and to view themselves as ethni- cally American."" Therefore, tlie radal/ethnic disdndions that re- main reflect enduring sodoeco- nomic viilnerabiiitiGs.
Because of the compiex social basis of radal/ethnic classifica- tions and identities, David Wil- liams proposed that Hispanic be included with African American and the various .̂ sian nati<inali- ties as a radal rather tlian an ethnic categorization.'" His justifi- cation is that the nmjority of His- panics self-identify as "Hispanic." and although tlie actual percent- age varies among studies, a large proportion do not further self-identify as either Black or White,"''^^ Subjectively, there- fore, what many consider to be an ethnicity is as basic as race in teiTOS of identity. What is sub- stantively important in such radal/ ethnic classification is the identi- fication of sociai and structural vulnerabilities assodated with group classification. Immigration adds another diinension of com- plexity to ethnic categories and identities. Immigrants who ar- rived from tlie state of Guererro in Mexico yesterday are very dif- ferent culturally and sodaily from ininiigi-ants of tlie same Mexican-origin census category
1152 I Commentary | Peer Reviewed I Angel and Angel American Journal of Public Health I July 2006, Vol 96. No. 7
COMMENTARY
whose ancestors arrived with the Conquest.* '̂
Census categories, even as they become more detailed aiid provide more choices, gloss over a great deal of heterogeneity ihat is of immediate importance Lo health and heaiUi service use?'' The reali^ is most health survey and census data use re- spondents' seli-ieported race, but only provide a limited number of choices. Biracial individuals or individuals who consider tliem- selves to be something other ihan White. Afiican American, Hispanic, or any of the other available categories answer questions about radal/ethnic j{i-oup classification in ways that are not yet understood.
Some data systems, such as the National Vital Statistics Sys- tem, do not even collect informa- tion on the race/ethnicity of the decedent, and data on mortality r-isks come from different and po- lentially contradictory sources. Data on the number of deaths. lor example, come from death certificates completed by fiineral directors or medical personnel on the basis of information from ;m infonnant, usually a family member.'̂ "' In other systems, such as those in which data are de- rived from hospital/patient care records, it is often unclear who made the racial/ethnic determi- nation. The different sources of radal/ethnic classification create a potential confounding factor when recording deaths." '̂' Infor- mation about the population at risk comes from survey data.^' !-;ach of these data sources intro- duces different possibilities for undercoujiLs or racial/ethnic mis- classification.
Additional reporting problems, such as the census undercount of minority group membei-s, af- fect population estimates. As a
consequence of the combined ef- fect of numerator and denomina- tor biases, it has been estimated that death rates are overstated by about 1 "/o for the White popula- tion and by about 5% for tlie Af- rican American popuiation. Such biases lead to underestimates of mortality for other groups, per- haps up to 21 % for the Ameri- can Indian or Alaska Native pop- ulations, up to 11% for Asian/ Pacific Islanders, up to 2% for Hispanics as a group,^'' and up to 6% for Mexican Americans.^''
In addition to gaining a better understanding of problems with administrative classification, re- seai'chei:s have become more aware of tlie potentially serious measurement biases that are in- hereiit when self-reported healUi data are used. Understanding the effect of these SES, cultural, and linguistic factors on the interpre- tation and response to questions about health is imperative if in- vestigators want to reduce poten- tial bias in the collection of data from survey and clinical respon- dents.'*" The group differences in cognitive schemas and world views that ethnographic studies of local and culturally based be- lief systems—including those that address disease and its causes- take as their objects of investiga- tion are methodological nui- sances for siurey reseai-chers and epidemiologists who want to deveiop valid and universal [jrobes that can be translated from one language to another for comparative use," Unfortunately. the figurative and impredse na- ture of language makes such an objective elusive.'^
Although researchers are aware of the potential confound- ijig of outcomes and predictors in comparative studies of the health of different groups, this potential confounding presents serious
pi'oblems to researchers who are only working with 1 cultural group. Individuals who have chronic conditions (e.g., diabetes) that have never been diagnosed by a doctor wiU answer nega- tively to a question about whetlier a doctor has ever told them they had the disease." Such confounding means that prevalence estimates for groups that have very different health care experiences, such as African Americans and non-Hispanic Whites, may vary gi^atly in their validity. In the absence of some objective criterion or other inde- pendent data about a respon- dent's actual condition, survey- based prevalence estimates must be inteipreted cautiously.
The ways in which individuals structure their responses to gen- eral health questions or to ques- tions about symptoms are poorly undei-stood." To make progress in measurement, researchers must have a much more sophisti- cated understanding of the im- pact of culture, language, SES, and other group-related factors on the complex response task. It is clear that reference group fac- tors affect how individuals evalu- ate their own healtli. Otlier cul- tui'ally based appraisals and valuations also may affect re- sponses. For example, it is possi- ble that in some cultures the fear of appearing arrogant leads indi- viduals to report their health as fair ratlier tlian as very good or excellent."''' One useful character- istic of comparative research is that it does not allow researehers to ignore the problems of compa- rability that probably affe(-t all data collecLJon efforts, even within the same cultural group.
We believe traditional epi- demiological approaches a:id re- search instruments, particularly those that elidt self-reports of
subjective states, should be com- plemented whenever possible by other techniques and should in- clude qualitative assessments of how respondents inteipret ques- tions and structuix' responses.^^ A multimethod approach may lead to a more sophisticated im- derstanding of subjective re- sponses spedficaily and the in- terview response task more generally.
Understanding social struc- tures and theii- impact on health requires an emphasis on both the cognitive aspects of culture and the social and material resources tliat individuals have at their dis- posal. '̂ The combination of tra- ditional epidemiological methods and ethnographic tecliniqiies is more effective for assessing the terminology that individuals use to talk about disease and the meaning it has for them. Com- bining qualitative techniques with surveys and even more ob- jective physiological data and [performance assessments will gi-eatly improve our knowledge of real comparative health levels among different populations and subgroups.
A CULTURE OF POVERTY?
The existence of minority group disadvantages in health indicators have led many to speculate about how poverly might cieate and |ierpetuate health disparities. Some theorists have suggested variations of the culture of poverty explanation (i.e., that chronic poverty leads individuals to develop a set of orientations and behaviors that are incompatible with sodal mo- bility and economic success or effective Involvement with social organizations) forwarded by Oscar Lewis several decades ago.*'' Susan Mayer, for example.
July 2006, Vol 96. No. 7 | American Journal of Public Health Angel and Angel \ Peer Reviewed I Commentary I 1153
COMMENTARY
argiied that poverty is a product of the loanied present orienta- tion ol" tliose who grow up in poverty."" Individuals who never witness a payofT to effective long-term pianning do not leam the niiddle-dass ability to delay gratific:ation and thus do not leam to plan lor their own fu- tures. From this perspective, tlie social environmenls in which such individuals grow up do not foster a strong work ethic, nor do Ihey encourage the resistance of immediate gî atification. Indi- viduals who have been social- ized in tliis way are unlikely to respond to educational opportu- nities or interventions for chang- ing their hehavior or reducing their health risks.
Blocked Opportunities More structural explanations
focus on the limited opportuni- ties available to individuals be- cause of their racial/ethnic chai'- acteinstics. From this perspective. Ihe deleterious heaith conse- quences of poverty are the result of exploitation and structural vul- nerabilities. Piven and Qoward, fttr example, explained higli rates of poverty among African Ameri- cans as Ihe result of institutional racism, which refers to the sys- tematic differential allocation of rewards (jn the basis of race. '** Institutional racism and discrimi- nation perpetuate poverty and its resultant individual-level healtli damage through unsafe and unhealtlifijl envii-onmenls, low educational levels, inadequate medical care, and feelings of helplessness and hopelessness." ~
Our reseairh and that of otiiers show that the fundamental nature of the laboi' market that places African /Xmericans and Ilispanics at a disadvantage in terms of health insurance also under- mines heath and well-being.' '*'
Historically, African Americans and Hispanics have been dispro- portionately confined to the low- wage service sector or to casual and informal jobs, where pay- ment is made in casli and where their ability to accumulate wealth is impaired. Discriminatory prac- tices iji the real estate market have confined many members of these groups to unsafe neighbor- hoods that liave few local em- ployment opportunities or com- munity rcsf)urces and inferior schools." Such confinement, and the inescapable poverty associ- ated with it. create chronically high levels of physical and social stress that increase the risk for poor health and vitality.""̂ Indi- viduals who live in tliese situa- tions lack adequate social capital and thus have few resources that might improve their lots.
Poverty and deprivation can undermine a people's sense of control and roh them of the opti- mism needed for a healthy life. Individuals who experience pov- erty, relative deprivation, and stress early in life become vul- nei-able to a variety of stressors throughout adulthood, which increases their risk for demoral- ization and depression late in life."" '̂ Older poor women, for example, are exposed to more social disruption in their lives compared with more af^uent in- dividuals, and these women's lives are often punctuated by a series of negative life events that are difficult to manage. At the same time, they are exposed to elevated levels of stress and have fewer resources for coping with life's hardships.''^
Disparities in Health Care Access
Among tbe reasons for the large differentials in health be- tween majority and minority
Americans ai'e the large differ- ences in adequacy of health care coverage, amount and quality of care, and access to long-tenn care.'"'~"^ Institutional racism that is rooted in ailturally insen- sitive and discriminatoiy prac- tices may explain the tendency for older minorities to receive fewer and lower-quality acute and chronic health care services.^ Those who sjiend their lives in low-wage service sector jobs are unable to save for retirement and the employers for whom they work rarely offer healtli or retirement benefits.̂ '̂ '
Even after contixil for SES dif- ferences, older African Ameri- cans perceive more discrimina- tion, personal rejection, and unfair treatment compared with non-Hispanic Wliites. and self- reported discrimination has been shown to inci"ease reports of de- pressive symptoms.^' In other cases, older minorities are sys- tematically excluded from pub- licly limded programs. Medicaid, for example, potentially penalizes poor elderly Mexican Americans and others who have lai"ge and complex families and want to care for frail parents. Under Med- icaid waiver programs, some states restrict eligibility to indi- viduals who have serious disabili- ties and are unable to function and who do not have access to other community-based services or family support. Although thii exclusion limits participation to those who have no other alterna- tives, it clearly discriminates against those who aie most de- pendent on their families. Rather than aiding family caregivers of elderly parents, this program may discourage their involvement^^
Immigration and Health Levels In addition to SES. nativity has
an important impact on health
outcomes. Studies on racial/ ethnic change in the United States have shown the increas- ingly important role nativity plays in determining the position of immigrants within the social structure.^' A generation of so- cial stratification has drawn atten- tion to ttie serious disadvantages immigrants may face in Ameri- can society. ̂ ''~ '̂' Although a se- lection effect may me-dn that im- migrants are healthier than those who remain behind or even indi- viduals who were bom in the United States,"" immigrants often suffer economic hardships and experience other strains as pail of the migration experience itself, which can undemiine their men- tal health and impede their social
As a result of inadequate health care in their aiuntry of origin, many immigrants may not be in optimal health when they arrive in the United States. In addition to the system-level barriers that may place the health of immi- grants at risk, disadvantages im- migrants face in the labor market also may place their health at risk. Many HLspanic elderly im- migrants have spent the majority of their lives outside the United States toiling in often harsh and dangerous conditions for very low pay. Many have been ex- posed to dangerous materials and have had inadequate preven- tive health care. Dangerous or difficult work and tlie lack of regular health care can result in serious health problems later in life, and a lifetime of low pay means that the financial re- sources necessary for maintain- ing a liealtliy independence can- not be accumulated.^" When these individuals become ill or incapacitated, they often have no recourse but to rely on family members for support.*''
1 1 5 4 j Commentary J Peer Reviewed | Anget and Angel American Joumai of Public Health i July 2006, Vol 96, No. 7
One reason why individuals kick health insuraiice is the em- ployment-based system of group health care coverage in the Uniled States. Few service sector jobs offer health insurance, and when they do, the premium that the employee is required to pay- particularly for family coverage- is prohibitive. Needless to say, jote that do not ofTer group cov- erage are unlikely to provide wages that allow employees to purchase private insurance. In the absence of a universal health care system in the United States, minority groups and re- i:ent immigrants are often con- fined to working in the low-wage service sector, which makes it difficult to obtain the care neces- .sary for maintaining optimal healtli with dignity.
Linguistic and Cultural arriers to Care
Racial/ethnic classifications say little about an individual s biological or genetic makeup. In Llie same vein, although such classifications indicate an individ- ual's origin, they say little about the individual's level of accultur- ation or cultural orientation. Broad census categories, such as Asian or Hispanic, combine various groups that have differ- ent cultures, belief systems, and histories. Specific nation-of origin groups also have very different immigration histories: they came to the United States at dif- ferent times in history, and they came for different reasons {e.g., economic opportunities vs polit- ical asylum).
Immigrants also have dilTer- ent levels of English proficiency and social competency, because of the age at which they immi- grated and other individual, family, and community factors.^'' Although immigrant children
quickly leam the language and customs nf the host society, older individuals and those who migrate to the United States late in life face particular problems in becoming fluent or proficient with the English language,''^ and many never do. Individuals who migi-ate during midlife or later often find the experience to be traumatic, because they are uprooted from familiar sur- roundings and are thrust into a new culture where they must leam a new language, new cus- toms, and a new set of social institutions. This can lead to mental health problems, such as depression.*"'
Cultural and Neighborhood Protective Factors
Although poverty and a lack of assets increase health risks among older minorities, other factors associated with culture potentially neutralize tliese health lisk factors and act in a protective manner. Cultural iden- tity and social Incorporation into a group that provides positive social involvement can improve health in and of itself, and group involvement can foster or en- courage positive health behav- iors.'""* Therefore, cultural factors that reduce the risk for social isolation are potentially health protective or enhancing.
Strong social institutions, such as family and church, can pro- vide similar support that pro- motes health and well-being.'"^'^ Hvidence suggests tliat religious involvement protects health gen- erally and plays an important role in minimizing the negative consequences of chronic condi- tions.''' Older Mexican American Catholics benefit fi-om frequent church attendance and report that it provides them with com- fort during times of trouble.''^
Church members can assist older inlinn members with daily tasks, which allows the older membei-s to remain in the community.*''*
Recent findings showed that residents who lived in high- density Mexican American and Cuban American neighborhoods were in better health than those who lived in lower-density neigh- borhoods.^" Although the data show a strong con-elation be- tween ghetto or barrio residence and poverty, other aspects of racial/ethnic enclaves may well protect health, possibly because of an enhanced sense of belong- ing, positive social interactions where the native language is spo- ken, and the availability of instru- mental social support.
DOES THE EPIDEMIOLOGICAL APPROACH MINIMIZE STRUCTURED INEQUALITIES?
Much progress in understand- ing health risks for individuals of all ages has been made in recent decades. Yet, it is dear that much remains to be understood if dis- parities in health are to be elimi- nated or even reduced and if everyone in the population is to enjoy optimal healtb at every age. To that end. we s u r e s t future re- search should improve our un- derstanding of how social policy and organizational structures and practices affect the opportunities available to minority Americans in ways that directly and iiidi- rectly affect group health levels.
The structured and institu- tional inequalities that have im- peded minority Americans' eco- nomic and social progress in tlie past and that continue to operate today—often in subtle ways- have their basis m a history of racism and systematic exclusion
from opportunities for economic and social advancement. Among African Americans and Hispan- ics. almost every aspect of social service delivery, educational op- portimities. and employment op- portunities have been infiucnced by race/ethnicity."''^ Data show that the health levels of entire groups are directly influenced by the fact that political and eco- nomic power are determined by both hi.story and the specific so- cial policies that perpetuate the social exclusion of specific groups of people.
Afiican Americans and His- panics lag far behind non4-lis- panic Whites in personal and col- lective wealth and political power. Lack of resources limits their ability to help their children and grandchildren buy houses and continue their education, and it translates into diminished eco- nomic and political power for the community' as a wbole. Although income and wealth do not guar- antee a good and virtuous life, poverty certainly does not guar- antee it either. Tbe intentional or unintentional exclusion of groups from sources of economic and political power is a major public health problem. We must develop a better understanding of the pathways to disadvantage and how health vulnerabilities are perpetuated Irom one generation to tlie next as the result of fonnal policies and institutional barriers to social mobili^.
Individuals choose to take ad- vantage of opportunities for economic and social advance- ment, and they make pei'sonal choices that affect their health. If opportunities for pei-sonal ad- vancement do not exist, or if they are blocked on the basis of group classification, members of that group find it difficult or impossible to avail themselves
July 2006, Vol 96, No. 7 American Journaf of Public Health Angel and Angel Peer Reviewed Commentary ' 1155
of potential avenues for social and economic advancement. Tliese blocked opportunities may result in frustrated hopes, demoralization, and deleterious health behaviors.
The complex association be- tween race/ethnicity and health has heen well doaunented at the individual level. African Ameri- cans suffer from more and more serious illnesses and die at higher rates compared witb non- Hispanic Whites. Although sur- vey-based studies tbat examine individuals and their vulnerabili- ties continue to provide useful information about health risks, Lheir failure to directly focus on the problems of institutionalized racism and exclusion is a serious sbortcoming. Studies tJiat ob- serve and analyze the individual have, for the most pail, not been accompanied by significant at- tempts to understand the role lai-ger social stmctures play in perpetuating racial/ethnic strati- fication and contributing to less favorable individual family, and community health profiles.
There are many reasons for Ihis i-elative neglect of structural and political factors. After World War II. the rapid development of survey researdi and the intro- duction of sophisticated analytic techniques pushed researchers in the direction of survey-based epi- demiological and health studies. Funding agencies, including the federal government tended to shy away from politically sensi- tive topics and instead focused on individual risk profiles. Tliis locus promised to inform public policy with educational and individual- level public health interventions. The power of individual-level biological approaches has mani- fested itself in the recent impe- tus to fund researdi projects that examine genetics and biology,'^
Yet the heath profiles of com- munities and groups are influ- enced by factors well above the level of the cell or the individual. Tliey are alTected by the ade- quacy of public healtti initiatives. federai and state health cai-e poli- cies, and other sodal policies. Be- yond that, bealtli levels are af- fected directly and indirectly by education, poverty, housing, physical and social environmen- tal stressors, and social exclusion and discrimination. These are emergent phenomena that can- not be undei-stood solely on the basis of individual-level studies.
New Directions in Research
Future investigations should build upon and add to the bio- medical mode! of disease and illness and should include a broader definition of health.̂ *' A more comprehensive and useful conceptual model of healthful aging might well hegin with a definition that includes not only the absence of disease and physi- cal infirmity at ils coi-e hut also tlie institutional and structural components and factors—such as educational opportunities, good housing, and sale neighborhoods— that have been shewn to affect health. The health of poor and minority Americans is under- mined by what has been termed tiie new morbidity. i.e.. threats to health from domestic violence, drug abuse, crime, and the perva- sive sense of inferiority that is the result of discrimination.
New studies on healthfiil aging should examine the underlying determinajits of illness within the community and develop better conceptual motiels and methods for assessing the stiiictured and in- stitutionalized stresses that minor- ity- Americans experience.̂ ^ We need to understand how these stresses affect individual-level
behaviors, patterns of sodal interaction, risk for \'idimization, aime. poverty, and other factors tliat inlluence health and func- tioning at all ages. Again, this approach should avoid purely individual- or family-level attribu- tions and should seaidi for the lai^er contextual factors that re- sult in structui'ed inequalities and disadvantage.
Ironically, the "diseases of af- fluence" in the United States- obesity, heart disease, cancer, and diabetes—take tiieir greatest toll on the least allluent, Tlie prevalence of these chronic dis- eases is affected by diet and other lifestyle factors and thus is influenced by SES. Almost one half (49.6%) of all African Amer- ican women anti more tliaii one third (38.9%) of Mexican Ameri- can women are obese.'*' To im- prove the health status of minor- ity women, newer and more aggressive efforts that educate medical care providers, extend community outreach, and im- prove compliajice with treatment regimens are necessary. Because ofthe pervasiveness of the struc- tui-al disadvantages minority .Vnericans face in the labor force, the entrenched poverty characteristic of urban ghettos and barrios and continuing dis- crimination efforts focused solely on individual health-related be- haviors aî e unlikely to be suc- cessful in improving population health levels.
In current practice, institu- tional and structural factors enter individual-level statistical models indirectly through controls for health insurance (private, Medic- aid. Medicare, or other coverage) and controls for income and edu- cation. Certain hierarchical tedi- niques indude ecological and larger geographical chai-acteris- tics. but these do not address
how institutionalized discrimina- tion, specific organizational structures, or fonnal aspects of public polides influence the health of spedfic groups. Because level of education, income, and wealth are determined by boUi opportunity structures and per- sonal choice, understanding how those structtires are maintained and how they operate to influ- ence health risks is necessary for understanding radal/etlinic health disparities.
The ina-easing awareness of tlie need to target research specif- ically at the unique health ml- nerabilides of poor and minority Americans is a welcome develop- ment Poverty, low educational levels, and other social disadvan- tages are the underlying causes of poor health generally, but these economic ajid .sodal disati- vantages are not randomly dis- tributed throughout the popula- tion and are greatest among Alrican Americans and 1 lispan- ics. Both groups will comprise a large proportion of the working- age population of the future, and they will comprise a growing pro- portion of the retired population. The capadty of tJie young to be productive and the general health levels and quality of life of Oie el- derly are both affected by factoi's closely associated with race, His- panic ethnicity, and inequality.
The Data Archive Attempting to better under-
stand tliese social vulnerabilities is an important research agenda. This effort will require the imagi- native use of existing data sets and an enhancement of san^)Ies to include larger oversamples of minority ,'\mericans. New data collection initiatives will be diffi- ailt during what is likely to be a period of retrenchment for majoi" funding agendes. Nonetheless,
1156 Commentafy F^er Reviewed Angel and Angel American Joumai of Public Health i Juiy 2006. Vol 96. No, 7
COMMENTARY
new and specialized data that examine spedfic vulnerabilities among groups that live and work in specific ecological and social niches will be necessary if we are to make progi'ess. During the past decade, the National Insti- tutes of Health have recognized the need for specially focused surveys that show the health sta- tus and functioning of minority elderly groups. ITie National Center for Health Stadstic's sup- plement to the National Health Interview Survey—the Longitu* dinal Study of Aging (LSOA)- presented new opportunities for documenting trends and cohort changes in the health and fiinc- tioning of a representative sam- ple of aging African Americans.'^ Data from several sources en- riched the LSOA data set and made the analysis of age-graded sodal processes possible.
Other existing and ongoing data sets include the National Health Interview Survey, the 1984 Health Insurance Supple- ment, the 1984 baseline Survey on Aging, the follow-up LSOA in- terviews. Medicare records, tlie National Death Index, and multi- ple cause-of-death files. Research- ers are usijig these data sets to examine patterns of health ser- vice access and use, including the impact of medical insurance, fam- ily dructure. housing, fomial and informal sources of care, employ- ment history, transpoi-tation, and sodal networks. There ai'e many unexplored possibilities for the in- novative and informative use of these data. The availability of lon- gitudinal data makes it possible to examine (1) the sequence and the consequences of morbidity and health care access on func- tional independence and dependence, as well tis death, within the community, and (2) the risk for institutionalization.
The possibility of new mod- ules in ongoing efforts provides new opportunities for under- standing the needs of spedfic groups. For example, although most survey questions were iden- tical in the first 2 waves in ttie LSOA series, new infomiation was gathered on individual risk behaviors, induding health opin- ions, during the third wave. In addition to interviews with sur- vivors, additional information was collected about decedents' hospitalization and nursing fadl- ity admission from their named next-of-kin contact. As part of the series, the Family Resources Sup- plement replaced the Health In- surance Supplement and pro- vided in-depth infonnation about caregiving, care receiver needs, unmet care needs, and reasons that needs were not met.
The Third Health and Nutri- tion Examination Survey (1988-1994) is a particularly useful source of information about the incidence and the prevalence of type 2 diabetes among the elderly. The sample had no preset upper-age limit and included individuals older than 85 years. This study in- cluded a medical examination of respondents and is one of the few data sets that provides both objective clinical observations and infonnation about the sub- jective experience of having dia- betes.'^ The life spans of older African Americans and Hispan- ics who have chronic conditions will prohably increase in the fu- ture as disease management im- proves. Yet, without substantial improvement to the economic and social situations of these groups, they will continue to fare worse than non-Hispanic Whites. Understanding all aspects of the assodation be- tween sodal factors, genetics.
and chronic illness is a higb- priority research objective.
The Hispanic Established Populations for the Epidemiologi- cal Stiidies of the Elderly is an important example of a spedal- ized study that is focused on a single group. This 10-year longi- tudinal study is ongoing and is sponsored by the National Insti- tute on Aging. It examines Mexi- can-origin individuals who tive in the Southwest and who were aged 65 years and older at the beginning of the study, its results are providing much needed infor- mation about tlie dynamics of aging throughout the life course. Studies of this sort are expensive and may gamer little political support if they focus on power- less groups. However, without such focused efforts, our tmder- standing of the physical and mental health and the health care needs of the minority elderly will remain superficial. National Insti- tute on Aging initiatives that are aimed at understanding and re- dudng health disparities among older persons and populations will foster these efforts.
In addition to important re- search on the health of older mi- nority Americans, special data makes it possible to investigate the impact of individuals' pre- retirement economic situations on welfare and healtli during their postretirement years. The Health and Retirement Study and the Study of Assets and Health Dynamics among the Oldest Old. for example, provide a better understanding ofthe complex interactions of race/ ethnicity, health, economics, and other social factors on aging processes for different groups. These data show serious income and aSxSet deficits among African Americans and Hispanics as they approach retirement, when
many of these individuals will lack resources for needed pre- ventive, acute, or long-term care and resources for living the most fulfilling life possible, including the possibility of help- ing their children.
Conclusions As we progress into the 21st
century, new and important med- ical innovations will increase life spans and will improve the qual- ity of those additional years. Much of that progress will no doubt result from a better under- standing of the genetic contribu- tion to disease. However, social structural factors that place cer- tain groups at a liigh risk for ill- ness and that impede their access to the highest quality health care continue to plague our sodety. As documented by the Institute of Medicine, institutionalized disad- vantages that manifest themselves most obviously as occupational, ijicome, and asset disadvantages across the life coui-se translate di- rectly into impaired health care access and poorer healtli among minority Americans.̂ ^ This fact makes it imperative that we con- tinue to examine sodal factors in health service, epideniiological, and health poficy research. The necessities of a healthful living and healthful aging are dear, but they ai'e out of tlie reach ol" many minority Americans. •
About the Authors ]acquelme L Angel is with ttie Schoot of Puhtic Affairs ami Depurtment ofSodot- Hgy. and Ronald J. Anget ts icith the De- partmettt of Sodotogy. Vmversity of Texas, Austin. Both authors are with the Popula- tion Hesearch Center. University of Texas, .'histin.
Requests for reprints should he sent to Jacqueline L. Angel. LBJ School of Public .•Iffairs. University of Texas at Austin. PO Box Y. Austin. TX 78713-8925 (e-mait: jangel@mail. utexas.edu).
This articte was accepted February 19. 2006.
July 2006, Vol 96. No. 7 | American Journal of Public Health Angel and Angel ' Peer Reviewed | Commentary | 1157
Acknowledgments The aulhors Uiank David WilliBms for hiK valuable comments.
Refereirces I. Institute of Mediant'. Coi'eruge Matters: Insurance ami Heaiih Ctire. Washington, DC: National Atmiemy
i. 2001.
2. link BK. Phelan JC. McKeown and tlie Idea that social cxjnditions are !iin- danicnEal causts of disease. AmJ Pubhc Htallh. 2002;92:730-732.
3. Phillips S. T7ic tmpaci ofPbeerti/ on Ikalih: A Scan of liesearcii Liletatiirc. Ollawa, Ontaiio. Caniida: Canadian In- stiluie b r Health Information; 2003.
4. Iiislilutf ol' Medicine. .'1 Shared Destiny: Community Effects afUmitsur- unce Washington, UC: NalionaJ Acad- emy Press; 2 0 0 3 .
5. Williams DR, Collins C. US sodo- ectmomic aiid racial differences in liealth pHtiems and exfilanalions. Annu
1995:29.349-380.
6. Ayanian JZ. Kohler BA, Abe T. I'pstein AM. The relation bclwe<jn hciillh insurance crjvcrage ami dinical oiitnme; anwmg women with bi-east can- cer. NEnt^JMed l993;329:326-33].
7. Pep|)pr Commission. A Call for Action, t'inul licport to the US. Bipartixan Commission im ("umpwAwwiiw Health Cun Washington, DC: US Government Printing OfTice; 1990.
H. Roelzlieim RG. PB\ N. Gonzalez HC, Ferrantt'JM. Van Diirme f)J. Kriiiclicr JP. Kffccts of heaUh insurance and race on cfjlorectal cancer ircntmenLs and out- a)m<«, AmJPiihljcHeaitk 2000:90: 1746-1754,
9. D<)wd JJ, Ucngtson VI.. Aging in minority [lopulations: an examination or the lioiibli- jeofiardy hypothesis. JGtroniol. 1978;33:427-436.
10. Fanner MM. Ferroro KR Are radal disparities in healtli tondiuonal on sri- doeconcimic status? Soc Sci Med 2 0 0 5 ; 60:191-204.
11. Ferraro Ki'. Fanner MM. Double jeopardy Lo health hyjx)thesis for Afri- can .\m('rican.s: analysts and critique. J Health Soclhhav. 1996:37:27-43.
12 Ferraro KF, Kelley-Moore J. Self- rated health and mortalily anioiig black «iid white adults: pxamininK the dy- naniic evaluation thesis. J Cenntiol St>c 5a. 2OOl,56:S195-S2O5.
13. Williams DR, Wilson CM, Race, cthnidly anti aging In: Binslork RH. Gtin'ge LK, MIS. Handbonk of Aging and Social .Sciences, 5tii ed. San Diego, Calif; Academic Prtss, 2001:16O-178,
14. HaywBitl MD. Gorman HK. llic
long arm of childhood: the influence of eariy-life social conditions on men's mortality. Deinograpliif. 20O4;41: 87-107.
15. Bulatao RA, Anrlersnn NB. eds. UndiTsumding Ratio! and Ethnic Differ- ettces in lieallh in iMie Life: A tiesetirch A^tda. Washington, DO. National Academies Press; 2004.
16. WillianiB DR. Race/ethnidly and socioeconomic status; ineasiircniPnt and rnetliodological issues. Inll J of Heaiih S&v. I996;26:483-5O5.
17. Helms JE, Jcmigan M, Mas<'Jier J. The meaning of race in psychology and how to change it: a meUuKtologica! per- spective. Am PsijchoL 2005;60:27-36.
18. Smedley A, Smedley BD. Raee as biology is fiction, radsm as a sodal problem is real: anthropological and hi.siurieal perspectives on the srtcial con- struction of raee. .-Im Psycho!, 2005:60: 16-26.
19- Williams DH. Raco and health: !)a.sic questions, emerging dircdiuns. Ann Eptdenml 1997:7:322-333.
2t)- Alba RD. llthnic Identity: the Transfunnatiim of W^ite Amenca. New Mavm, Conn: Yale University Press; 1990.
21. Tucker C, McKay R, Kojetin B, et al. Testing Methods of Collecting Hudul and Ethnic Injbniuiiiun: Results of the Current Population Survey Supplement on Hace and Ethttidti/. Bureau of Labor Strt- OstKS Siatislical .'Vores. No. 40. Washing- ton, DC; Bureau of Labor Statistics: 1996.
22. Daza P. Mazas C, Nguyen L. Wetter DW Categorizing race among Mis|>anic .stnokera. Cancer Omtroi 2005:12{suppl2):9l-92.
23. Burchard K.G.. BoiTell I.N, Choudhry S. et al. Latino populations: a unique t^portunity for the study of race, genetics, and sodal environment in epidemiological research. AmJ Public Health. 2005:95:2161-2168.
24. l.aVeLsl TA, Beyond dummy vari- ables aoil sample selection: what healtli services researchers ought to know about race as a variable. Healt/i Sem ties. 1994:29:1-16.
25. RosenlieigHM, MaurerJD, Sorlie PD, et al. Quality of deatli rates by race and Hisiianic origin; a summary of cur- rent researdi. Vital Health Stat 2. 1999: 128:1-13,
26. National Center for IicaJth Statis- tics, Vital Statistics of the United States, 1992. Vol li Mortality. Part A Techni- cal aiiprntiix. HyatLwille, .Md; US Dept of Health aiid Human Sta-vjces: 1996.
Health Survojs. San Fnincis«i. Calif: Josscy-Bass; 1989,
28. Sodie PD, Rogot p:, Johnson NJ, Validity of demographic characteristics on the death ccrtificatt;, Epidemiolog/. 1992;3;iai-184,
29. F t̂el KV. Rschbach K. Ray LA, Markides KS, Letters to the ed, rKgarding Evaluation of mortality data for older Mexican Amencans: implications for tlie Hispanic paradox, AmJ Epidemiol. 2 0 0 4 : 160:1030-1031.
30. Skinner JH, Teresi JA, Holmes D, Slahl SM. Stewart .'\L, eds. Multicultural Measurement in Older Riputatiotis. New York. NT: Springer: 2002.
31. Hogler, L,H, Methodological sources of atltiiral insensitivity in men- tal health research. Am Psychol. 1999: 54:424-43:).
32. D'Anclradc R, The t)evelopment of Cognitive Anthrojmlogy. Cambridge, UK: Cambridge University Press: 1997,
33. Angei RJ, Narrative and Ihe fimda- mental limitations of quantification in fToss-cultural research. Med Care. In pit'ss.
34. Angel R, Thoits P. l l i e impact of culture on the cognitive structiiR- of '\\\- ncss-Ctdt Med Psiichuary. 1987:11: 2,3-52.
35. IjiVeist TA. Disentangling race and sodoeconomic status: a key to solving health dl-qjarities. / Urban Health. 2 0 0 5 ; 82;26-34,
36. Lewis 0 five Families: Mexican Case Studies in the Culture of Poverty. New York. NY: Pfrseiis Books; 1975.
37 Mayer SF. What Mona/ Cant Buy: Family Income and Children's Life Chances. Cambridge, Mass: Harvard University Press: 1997,
38. Piven FF, Qoward RA The New Class War. New York. N'Y: Pantheon Books; 1985.
39. Williams DR. Radsm and health; a reseai-di agenda. Ethnidty Dis. 199fi: 6; 1-6.
40. Angel RJ. Lein L, Henrici J. Pour Families in .'Uiierica's Health. Care Crisis: How the Other Half Pays Cambridge. UK: Cambridge Universi^ Press; 2006.
41. Wilson WJ, When Work Disap- pears: The World of the Urban Poor. New York. NY: Knopf; 1996-
42. Kawac-hi I, Berkman LF, Soda! ties and mental health, / Urban Health. 2001:78:458-67
4 3 . Mechanic DC. tnescapdjle Deci- sions. New Brunswick, NJ: Transaction PublLshers: 1994.
27 Aday LA. Designing and Conductitig 44, Dohrenwend BP, Levav 1, Shrout
PE. et al. Socioeconomic status and psychiatric disorders: the causation- selection issue. Science. 1992;255; 9 4 6 - 9 5 2 .
45. McLeodJD, Kessler RC. Sodoeci)- nomic status differences in vulnerability to undesirable life events. J Health Soc Belmv. 1990:31;162-172.
46. Angel RJ, .^ngel JL. The extent of private und public health insurance coverage among adult Hispanics. Gerontologist- 1996;3H;332-34O,
47. Mor V. Zinn. Angelelli J, Teno JM, Miller SC, Driven to tiers: socioeco- nomic aiid radal disparities in the qual- ity of nursing home care. Milbank Q. 2 0 0 4 : 8 2 : 2 2 7 - 2 5 6 .
48. Dela Torre A. Friis R, Hunter HR. Garda L, The heaitii insurance status of US Latino women: a profile from the 1982-1984 HHANFS. AmJPublw Health. 1996:86;533-537
4 9 Margraves JL, HadleyJ. Impact of hcaldi care context: the cxtntribution of insurance coverage and (X)mmunit>' re- sources to reducing radal/ethnic dwpar- ities in access to care. HeaWt Serv ties 2 0 0 3 ; 3 8 ; 8 0 9 - 8 2 9 .
50. Ctystal S, Shea DG, Cumulative advantage, public policy, and inequali^ in later life, Ann Rev Cenmtol Geriatr 2 0 0 2 ; 2 2 ; l - 1 3 .
51. Bame,s LL. Mendes De Leon CF, Wilson RS, et al Radal differences in perceived dLscriminatioii in a commu- nity population of older blacks and whiles-J Aging HeaUh. 2004;16: 315-337.
52. MutdderJE, Angel JLPbUcy de- velopment and the older Latino popula- tion in the 21 st centurvv IntlJ Sodol Soc R)/in/, 2000:11:177-188.
53 .\iigel JL Devolution and the sodal welfare of elderly immigrants: Who will bear the buitlcn? Public Adm Rev. 2003:63:79-89,
54. E^>enshade T], Fu H. An analysis of FJiglish-language pmfidency among US. immigrants. Am Sodol Rev 1997: 6 2 : 2 8 8 - 3 0 5 .
55. Orlield G. Commentary on the education of Latino youtli. In: Suarez- Orozco MM. Pae;* MM, eds. Latinos^ Hemakir\g Amenca. Berkeley, Calif: Uni- versity of California Press: 2 0 0 2 : 3 8 9 - 3 9 7
56. Massey D. Denton N. American Apartheid. Cambridge. Mass: Harvani University Press; 1993.
57. Mai'kides KS. Eschbach K. Aging, migraiion, and mortality: current status of research on the Hispanic paradox, J Cerontol Soc Sd. 2005;(>0;S68-S75.
58. Angel JU Angel RJ. Age at m^(ration.
1158 I Commentary Peer Reviewed Angel and Angel American Journal of Public Health July 2006, Vol 96. No, 7
COMMENTARY
sodal conneclions, and well-being among elderiy \iispamcs. J /\giiig Health. 1992;4: 4 8 0 - 4 9 9 ,
59. Hao L, Johnson RW. Economic, ciiltural, and social origins of emotional well-being. Res Aging. 2000:22: 5 9 9 - 6 2 9 .
('(>, Angei JL, Buckley CJ, Sakamoto A, Duration or disadvantage? exploring na- ijvity, ethnicity, and health in midlife. / Gerontot B Psychol Sa Soc Sd. 2 0 0 1 : 56:S275-S284.
fii. Angel RJ. Angel JL, Lee GY, Markides KS. Age at migi-aiion and fam- ily dependency among older Mexican imniigrants: recent evidence from die Mexican Aniencan KPKSIi Geroniolo- gist. 1999:39:59-65.
62. Burr JA, Mutchler J[i. lingli.sh lan- guage skills, ethnic concentration, anti household composition: older Mexican immigrants. / Gerontol B l^choi Sd Soc Sd. 2003:58B:S83-S92,
63. Angel RJ,, Angel JL. Markides KS. 1^ salud fisica y mental de los Mexi- canos tnigr'antes mayores en Los feta- dos Unidos. In: Salgado de Snyder VN. Wong K, eds, Envejedendo en La Po- breza: Una G^eiv, Salud y Calidad De Vida, eds. Ciiemavaca, .Mexico: Instituto Nadonal de Saliid Piihlica: 2 0 0 3 : 153-172.
fi4. Lara M, Camboa C, Kahramanian MI, Moi-ales 1^, Hayes Bautista DE, Ac- culturation and Latino health in tlie United Status: a review of the literature and ils sociopoiidca! context. Ann Rev PuhlK Health. 2 0 0 5 : 2 6 : 3 6 7 - 3 9 7
65, MarkidesKS, BoldtJS, R a y L Sources of helping and intergenerational solidarity; a three generations study of Mexican-Americans, y Cefonro/. 198fi: 41:506-511.
fi6. Taylor RJ. Chatters LM, ftittems of intimnal support to elderly Black adults: family, Iriend.s, and church members. SiK' Work. 1986: November-(December: 4 3 2 - 4 3 7
67 Idler E, Benyamini Y. Self-rated health and mortality: a review of iwenty-seven eommunity Studies. / Health Soc Behav. 1997:38:21-37.
1)8. HiUT, Anget JL, Ellison C. Angel RJ. Religious attendance and mortality: an eight-year follow-up of older Mexican Americans. / Gemntoi B Psychol Sd .Soc 5ct. 2005:60B:S102-109.'
69. Berkman LF. Looking beyond age and race: the stnictiire of networks, functions of support, and chronic stress. Epidemiol. 1997;8:469-470,
70. Eschbach K, Ostir GV, Pate! KV, Markides KS, Goodwin JS. Neighbor- hood context and mortality among older Mexican Americans: Ls thei-e a
barrio advantage? AmJ Public Health. 2004:94:1807^1812.
71. GilensM. W!a/Americans Hate Welfare: Rare. Media, and the Politics of Antipoverty Policy. Chicago, III: Univer- sity of Chicago: 1999.
72. Quatlagno J. Why the United States has no national health Insurance: stakekholder mobilization against the welfare state 1945-1996./Wea/rt Sof Behav. 2 0 0 4 : 4 5 : 2 5 - 4 4 .
73. Whitfield KE, McOeam G. Genes, environment, and race: quantitative ge- netic qjproaches. Am Psychol 2005:60: 104-114.
74. Brestow L. Health measurement in the tliinl era ol health. AmJ Public HraM. 2006:96:17-19.
75. Alwin DF, Wray LA. A life-,span developmental persjjective on sodal sta- tus and health. / Genmtnl B Psychol Sd Soc SCI. 2005;60:7-14,
76. National Center for Health Statis- tics. Health, United States. 2005. With Gharthook on Trends in the Health of Americans. Table 73. MyatLsville. Md: National Center for Health Statistics: 2005.
77 Kovar MG. Fitti JE. Chyba MM. The Longitudinal Study of Aging: 19S4-1990, Vital Health Stat I. 1992; 2 8 : 1 - 2 4 8 .
78, National Center for Health Statis- tics. Plan and opcradon of the Thin! National Health and Nutridon Examina- tion Survey, 1988-94. Vital Heaith Stat 1. 1994:32:1-407.
79, Institute of Medicine. Insuring America's Health: tYindples und Recom- mendations. Washington, DC; National Academv Press: 2004,
Fighting Global Blindness Improving World Vision Through Cataract Elimination
By Sanduk Ruit, MD,
Charles C Wykoff,MD, D.Phil., MD,
Geoffrey C Tobir), MD
Unoperated cataract is the cause of millionsof cases of visual impairment and blindness in poor populations throughout both the developing and the developed worid. This wonderfully written volume shares the experiences of a team of surgeons who have demonstrated how the surgical procedures can be simplified and made more efficient, accessible, and far less expensive.
ISBN 0-87553-067-2 • spiral bound • 2006 S31.50 APHA Members • $45,00 Nonmembers
American Public Health Association 800 I Street, NW. Washington, DC 20001 www.apha.org
To ORDER: web www.aphabookstore.org
email apha@pbd.com fax 888.361 .APHA phone 888.320.APHA M-F8am-5pm EST
July 2006, Vol 96, No. 7 | American Journaf of Public Heaith Angel and Angel i Peer Revievred | Commentary ' 1159