Sociological contributions to race and health: Diversifying the ontological and methodological agenda

Abstract

Sociologists have made fundamental contributions to the study of race and health in the United States. They have disrupted biological assumptions of race, uncovered individual and structural factors that drive racial health disparities and explored the effects of racism on health. In recent years, however, with broader shifts towards big data, the work to understand the dynamics between race and health has been increasingly pursued from a quantitative perspective. Often, such analyses isolate intermediary mechanisms to further explain race as a cause of disease. While important, these approaches potentially limit our investigations of underlying assumptions about race and the complexity of this critical social construct. We argue that the resulting dearth of qualitative research on race and health substantially limits the knowledge being produced. After providing an overview of the overwhelming shift towards quantitative methods in the study of race and health, we present three areas of study that would benefit from greater qualitative inquiry as follows: (1) Healthy Immigrant Effect, (2) Maternal Health and (3) End-of-Life Care. We conclude with a call to the discipline to embrace the critical role of qualitative research in exploring the dynamics of race and health in the United States.

INTRODUCTION

From theoretical conceptualisations of race (Bonilla-Silva, 1999, 2019; Omi & Winant, 1994; Ray, 2019) to empirical studies exploring mechanisms that drive racial inequality across all facets of social life, sociologists have extensively explored race and racialised experiences in the United States. Quantitatively, scholars have examined the relationship between race and various individual-level outcomes including socioeconomic status, educational attainment (Telles & Ortiz, 2009), employment and housing (for review, see Pager & Shepherd, 2008). Sociologists have also analysed neighbourhood effects (for review, see Sampson et al., 2003) that lead to concentrated disadvantage (Du Bois, ([1899]1967)), residential segregation, and the creation of the underclass (Massey & Denton, 1993; Wilson, 1987), which disproportionately impact people of colour. Qualitatively, scholars have focused on lived experiences, particularly those of people of colour residing in urban settings (Desmond, 2016; Duneier, 2000). A robust sociological tradition of urban ethnography, that dates back to the 1940s, has extensively explored neighbourhood contexts and how the increased risk of marginalisation, discrimination and criminalisation directly shapes the social worlds of people of colour living in these spaces (Goffman, 2014). Furthermore, more recently scholars have looked beyond the ‘urban underclass’ and instead have examined the experiences of middle class and elite racial minorities living in urban (Pattillo, 1999), suburban (Lacy, 2007) and rural settings (Eason, 2017). While themes of marginalisation, discrimination and stigma still emerge in this work, this scholarship has also shed much needed light on the multi-dimensional lived experiences, cultural meanings, decision-making and interactional orders surrounding people of colour living in the United States.

In the subfield of medical sociology, US scholars have documented extensive racial disparities in health, often explored from a quantitative demographic tradition. For example, age-adjusted black-white death rate ratios have worsened for blacks for heart disease and cancer over the past several decades, blacks have higher rates of premature mortality than their white counterparts (Geronimus et al., 2001), greater incidence and severity of hypertension (Kershaw et al., 2011), obesity (Kershaw et al., 2013) and Type II diabetes as well (Bancks et al., 2017), and the homicide rate was over seven times higher for Black men than for White men in 2016 (National Center for Health Statistics, 2018). Resonant with several aspects of broader sociological race research, explanations for these racial disparities in health often invoke differences in socioeconomic status, neighbourhood residential conditions and medical care, frequently the consequence of enduring structural inequalities and racism (Riley, 2017; Sewell, 2016; Williams & Collins, 2001; Yearby, 2018).

Despite these strengths, however, the subfield of medical sociology exhibits some key departures from the sociological study of race more broadly. First, as noted by Williams and Sternthal in their contribution to the 2010 special issue of Journal of Health and Social Behavior, ‘for much of the twentieth century, as reflected by publications in the two leading journals in American sociology, the health of the black populations has not been a central focus of the discipline’ (Williams & Sternthal, 2010, p. S16). Indeed, their review of American Journal of Sociology and American Sociological Review revealed a total of only 18 publications on racial-ethnic disparities in health by 1989, with fourteen more published from 1990–2008 (six in AJS and 8 in ASR). In that article, Williams and Sternthal identify a series of important sociological contributions to scholarship on race and health, including, for example: (1) showing that race is largely a social rather than biological category; (2) capitalising on the multidimensionality of socioeconomic status to more fully articulate its relationship to race; (3) developing multilevel constructs of racism; and (4) identifying a multitude of empirical pathways through which segregation can affect health. The authors conclude by calling for quantitative health data to be ‘routinely collected, analysed, and presented simultaneously by race, SES, and gender’ (2010, p. S23). As a whole, this 2010 review characterises flagship medical sociology research on race and health as making critical contributions but also as limited in volume and largely oriented to using quantitative operationalisation and analytic tools in the measurement of racial disparities.

Using the same search terms as Williams and Sternthal (2010, p. S16), we observe that flagship sociological research on race and health has modestly increased in volume in the intervening decade, but retains the largely quantitative ontological and methodological orientation of earlier work. For the years 2009–2020, out of approximately 385 articles published by American Journal of Sociology, four were captured by our search for the terms ‘race and health’, ‘health inequality and race’, ‘health inequality and ethnicity’ or ‘health disparity’ in abstracts; of those, none were qualitative. For American Sociological Review, eighteen articles addressed these topics, but only three of those used qualitative methods. For Journal of Health and Social Behavior, there was a larger volume of work addressing race, health, and inequality but a similarly small proportion was qualitative (for example, 68 unique articles came up under a search for ‘race and health’, one of which used qualitative methods). In total, only four qualitative JHSB papers were captured with the full set of search terms – a meagre 1.5% of all the articles published over eleven years by American Sociological Association’s lead medical sociology journal. Thus, while flagship publication of research in these areas has certainly increased from the pre-1989 period, and perhaps shown modest gains since 2009, developments have been overwhelmingly in the analysis of large secondary data sets and negligible qualitative approaches. Furthermore, this pattern is not fully explained by assuming that qualitative studies of race and health are simply published in other outlets. A broader search of Dissertation Abstracts International and Sociological Abstracts, using the same search terms and time periods, indicates that the proportion of research on race and health that is qualitative is generally higher than in the flagship journals, yet still quite modest. For example, our search on ‘race and health’ yielded 3741 (English language) doctoral dissertations, of which 419 were qualitative (11.2%). Considering the full set of peer-reviewed scholarly articles and books listed in Sociological Abstracts, the same search yielded 1841 publications, of which 63 (3.4%) were qualitative, a figure more in line with the flagship journals. A search of current NIH grants shows a current portfolio of 35,322 grants on ‘race and health’, but only 6% of those also use qualitative methods. Based on these data, we conclude that methodological foci of US flagship journals may be partially explained by selection biases in the placement of papers but that a broad survey of sociological scholarship suggests that these trends towards quantitative investigations of race and health also hold more generally.

Critically, we are loathe to conflate ‘flagship’ with ‘quality’ in this discussion. We recognise that there is a plethora of excellent qualitative research on race and health published in outlets other than AJS, ASR and JHSB. This type of research is regularly published in respected outlets such as Social Science and Medicine, The Milbank Quarterly and Journal of Health Care for the Poor and Underserved. Indeed, Sociology of Health and Illness has been a leader in this area for many years exploring racialised experiences across different regions and health-care contexts (e.g., see Cain & McCleskey, 2019; Gunaratnam 2001; Persson & Newman, 2008; Smart & Weiner, 2018; Younis & Jadhav, 2019). However, we believe this ontological and methodological dominance of quantitative approaches to studies of race and health has major implications for sociology more broadly, in the United States and internationally. The assumptions and conceptualisations embedded in the intellectual traditions we see calcifying in sociology around race and health inherently narrow the types of research questions posed, concepts developed, evidence marshalled and conclusions drawn about race and health. This dominance implicitly positions qualitative approaches to these topics as ‘alternative’ or secondary to the main discoveries.

Far from being secondary, we view this type of work as fundamental for moving beyond the treatment of race as an essentialist concept and having a deeper engagement with the ways that race is configured in relational, processual and contextual terms (Loveman, 1999; Morning, 2014; Saperstein & Penner, 2012). Such understanding entails unpackaging of systems, organisations, variables, racialisation and proxies for race, all of which are embedded in cultural, social, and economic systems and resources which have the potential for multiple types of health influences (Nazroo et al., 2020). Furthermore, stratification systems, including the ways that people explain, rationalise and justify their biases, are reproduced through dynamic interactions (Bonilla-Silva, 2006), which are insufficiently captured in quantitative measures. Qualitative approaches are ideally equipped for the elucidation of such processes and refinement of our assumptions and theories about how race operates. These approaches can help reveal our own biases and points of stagnation as theorists and researchers.

Furthermore, the implications of these patterns are cascading and reifying, mattering for intellectual and professional reasons. To the extent that hiring, funding and career trajectories are tethered to dissemination outlets, and reputations of departments and programmes are built on those indicators of individual researchers’ successes, the type of research that is published in flagship outlets matters. Rather than be relegated to the ‘alternative’ spaces within and around sociology, we believe that qualitative research on race and health should also be central to the modern sociological agenda in order to help the field outgrow outdated essentialist conceptualisations of race and bring health research into alignment with cutting edge race theory and sociological studies of race more broadly. Without research examining assumptions about race categories, racialisation processes, and the ways that racial bias and discrimination become built into our organisations, cultures and emotional economies, our understandings of race and health will be fundamentally limited.

In this article, we examine the implications of this limited qualitative research agenda around race and health in sociology, with careful attention to the types of empirical questions being asked, the methodologies being used, and the potential gains to be had with increased ontological and methodological diversity. Here, we selected three key areas of research that would greatly advance scholarship on race and health through more qualitative inquiry: Healthy Immigrant Effect, Maternal Health and End-of-Life Care. That said, this sort of evaluation is needed in most aspects of research on race and health and could be explored in any area where race is treated as a quantitative variable related to health. We selected these areas as they: (1) represent the full spectrum of the life course; (2) are not entirely centred on clinical interactions, thus providing an opportunity to examine how race is operative upstream of medicine and how those processes are not well understood; and (3) as elaborated below, each of these subfields is faced with prolonged puzzles around the role of race and are thus ripe for qualitative inquiry.

First, we provide a general overview of the current literature in each research area, focusing on empirical studies published in the flagship US sociological journals American Sociological Review, American Journal of Sociology and Journal of Health and Social Behavior. We highlight successes and opportunities lost in each area and propose qualitative research agendas to fill critical gaps in the current scholarship. We must note that the broad, multidisciplinary scope of these topics makes it impossible to fully address all of the studies on each of these areas within a single article, but we do our best to provide comprehensive overviews of the research areas within the United States. We conclude with a discussion of how the dominance of quantitative methodology and ontology in sociology not only hinders intellectual advancements in scholarship on race and health, but also has cascading effects that directly shape institutions and the profession more broadly. We also consider possible reasons for these observed patterns.

HEALTHY IMMIGRANT EFFECT

The Healthy Immigrant Effect, or the Latinx Health Paradox, refers to the fact that foreign-born individuals, particularly Latinx immigrants, tend to perform better on health indicators such as all-cause mortality, infant mortality and self-rated health compared with Whites and their US-born counterparts, despite experiencing lower socioeconomic status and educational attainment (Markides & Coreil, 1986; Scribner, 1996). The health assimilation model has been widely accepted to explain this paradox: as Latinx immigrants become more acculturated through prolonged US residence, they lose protective health effects that give them this health advantage. In recent years, however, scholars have increasingly questioned the paradox's efficacy arguing that while Latinx individuals may live longer, they tend to have faster physical decline and poorer quality of life compared with Whites (Boen & Hummer, 2019).

Sociologists have confronted assumptions in the paradox primarily through quantitative methodologies that have been extensively published in the JHSB. Individual-level factors explored include socioeconomic status (Brown et al., 2016), educational attainment (Crosnoe, 2006), gender (Gorman et al., 2010), English language proficiency (Lebrun, 2012), country of origin (Montazer & Wheaton, 2011), arrival year (Hamilton et al., 2015) and length of residence in the United States (Tegegne, 2018). Others have looked to cultural and structural factors, including intergenerational family dynamics (Schmeer & Tarrence, 2018) and the impact of institutions, neighbourhoods and policies on immigrant health (Dondero et al., 2018; Rosenbaum, 2008; Viruell-Fuentes et al., 2012). One persisting critique of the paradox is that studies have mostly ignored the health status of undocumented individuals. The available scholarship however has found that undocumented individuals are much less likely to access necessary healthcare services (Torres & Waldinger, 2015) and experience an exacerbation of chronic conditions (Hacker et al., 2011).

New empirical research continues to problematise the Healthy Immigrant Effect's depiction of US immigrant health. For example, the relationship between variables like citizenship and English language acquisition, common measures of acculturation and assimilation, and immigrant health do not consistently align with the paradox. Lebrun (2012) found that immigrants who remained in the United States and Canada for longer and acquired greater English proficiency, tended to have better healthcare access, utilisation and experiences with healthcare providers. Tegegne (2018) argued that immigrants who were embedded in diverse networks and acquired greater English language proficiency, reported better health over time and were less likely to be diagnosed with a chronic condition.

While quantitative scholarship on the Healthy Immigrant Effect is robust, qualitative research in the area has been ostensibly absent in the discipline's flagship journals. Sociologists who have utilised qualitative methods have frequently done so to explore the efficacy of the Latina Birth Paradox, which asserts that Latinx immigrant women have better birth outcomes than their White and US-born counterparts (Fleuriet & Sunil, 2015). In their interview study of pregnant Mexican women in Texas, Fleuriet and Sunil (2015) found that while infant birth weight decreased as length of time in the United States increased and Mexican American women reported higher levels of depression and stress, Mexican immigrant women reported higher levels of anxiety directly related to the pregnancy itself. These findings resonate with recent quantitative research that also reveals that foreign-born Latina women do not necessarily have better birth outcomes than White women (Hoggatt et al., 2012) or only in the context of specific birth outcomes and migration histories (Urquia et al., 2012). Collectively, this research points to the ongoing need to reconsider our assumptions of immigrant health. For instance, these inconsistencies may point to the need to embrace a life-course perspective when thinking about immigrant health, recognising that health and immigrant status are fluid and deeply interconnected experiences that cannot be measured simply as single data points in time.

This relative dearth of qualitative explorations of the Healthy Immigrant Effect, which could be attributed to the model's historical roots in epidemiology and assumptions that qualitative measurement of this paradox may be too difficult, leads to persisting gaps in the literature. Variables across studies share the same name yet result in vastly different findings due to their complexity and the great variation across immigrant groups. These inconsistencies beg for greater qualitative inquiry, which is particularly well-equipped for the refinement of our assumptions related to the Healthy Immigrant Effect. One way to target these persisting puzzles would be to incorporate more qualitative inquiry to explore how the paradox relates to undocumented status, racialised experiences, discrimination and other obstacles that emerge in daily life (Viruell-Fuentes et al., 2012). It is important to note that there has been valuable qualitative research on immigrant experiences in health-care encounters, revealing how culture, contexts and social structures directly shape immigrants’ clinical experiences (e.g. Lo, 2010 and Lo & Stacey, 2008). We argue that such conceptualisations and approaches should be applied to examine the Healthy Immigrant Effect as well.

Another area of fruitful inquiry is qualitatively exploring the impact of neighbourhood effects on immigrant health. Cagney et al. (2007) found that while foreign-born Latinx have much lower rates of asthma and other respiratory conditions compared to their US-born counterparts, this advantage was contingent on the physical environment. When Latinx immigrants lived in communities with high rates of other immigrants, the respiratory health advantage remained, yet when Latinx immigrants lived in largely non-immigrant communities, their respiratory health deteriorated (Cagney et al., 2007). What are the resources, opportunities and experiences in majority-immigrant communities vs. non-majority-immigrant communities that may lead to such differential health outcomes? How do neighbourhoods reproduce racial inequalities via norms, culture, and various other mechanisms? And how do individuals in a community differentially encounter agency, segregation and discrimination in different structural contexts (Ray, 2019; Wingfield & Chavez, 2020)? Participant observation, interviews and other qualitative approaches would be able to offer critical insights into these questions.

There is also value in applying qualitative methods to understand how health intersects with race, policing and the implementation of both local and federal immigration policies. Recent qualitative scholarship on policing and incarceration has elucidated the complex, and at times surprising, ways in which race operates in these contexts (Eason, 2017; Western, 2018). Negative perceptions, fear of, and past traumatic experiences with providers and law enforcement officials may shape individual decision-making and patient–provider interactions, exacerbating inadequate access to care, leading to undetected, poor health outcomes (Read & Reynolds, 2012) and expectations of poor health and survival in the long-term (Warner & Swisher, 2015). Collectively, these examples of possible qualitative inquiry would help unpack both individual and structural factors that directly shape immigrant health in the United States.

MATERNAL HEALTH

Clinicians and public health scholars have primarily used quantitative methods to explain sustaining racial disparities in pregnancy-related morbidity and mortality in the United States (Creanga et al., 2012). In contrast, within sociology, the relationship between race and maternal health has been less central to the research agenda, which has focused largely on reproductive health, childbearing experiences and maternal and infant outcomes. Much of the scholarship published in top US sociology journals has relied on quantitative methods and have operational-ised maternal health in terms of infant birth weight or pre-term births (Dooley & Prause, 2005) and mothers’ physical and mental health (Hartnett & Brantley, 2020; Williams & Finch, 2019). Individual maternal characteristics, like age (Mirowsky, 2005), education (Kravdal & Rindfuss, 2008), socioeconomic status (Strully et al., 2010), employment (Frech & Damaske, 2012) and relationship status (Meadows et al., 2008), and structural factors, such as social networks (Balbo & Barban, 2014; Sandberg, 2006), neighbourhood context (Morenoff, 2003) and residential segregation (Walton, 2009), have been targeted as critical factors that shape disparate maternal and infant health outcomes.

In the past decade, quantitative sociologists have accelerated a research agenda exploring how race impacts maternal health (Colen et al., 2019; Reyes et al., 2018; Williams et al., 2015). Reyes et al. (2018) found that Black women experienced greater emotional distress during pregnancy than their White counterparts. Williams et al. (2015) found that African American women, who had a child when they were adolescents or young adults, reported worse health outcomes later in life compared with White women who had children at similar ages. Furthermore, Colen et al. (2019) found that African American women, whose children experienced acute and chronic discrimination, had faster and poorer declines in health compared to their White counterparts.

In contrast, qualitative scholarship has largely investigated the organisational, sociocultural and political structures that shape women's experiences navigating their reproductive health (Almeling, 2007; Waggoner, 2017), pregnancy (Armstrong, 2003; Casper, 1988; MacKendrick & Cairns, 2018) and childbirth within the United States (Dillaway & Brubaker, 2006; Morris, 2013) and abroad (Storeng et al., 2010; Story et al., 2012). While this research has provided key insights revealing experiences, mechanisms and consequences of maternal healthcare, race has largely remained on the periphery – particularly in the research that has been featured most prominently in the discipline. For instance, Almeling’s (2007) ASR article exploring the impact of sociocultural notions of gender and family on the markets of egg and sperm donation provides invaluable information on how such market dynamics result in divergent compensation, marketing tactics and interactions with egg versus sperm donors. Yet, there is little discussion of how racial backgrounds of donors and clients inevitably shape the market.

Similarly, in her study on foetal alcohol syndrome, Armstrong (2003) found that while low-income women of colour were generally less likely to drink than White women were, a small group of these women reported high levels of drinking during pregnancy. There, however, is a missed opportunity to unpack the factors that may lead to this pattern of distribution of FAS: although Armstrong (2003) notes structural inequalities that put low-income women of colour at greatest risk, much of her analytical argument focuses on how clinical and policy approaches to FAS lead to the social regulation of all women. Anthropologists, on the other hand, have delved more deeply into the relationship between race and maternal health (Bridges, 2011; Davis, 2019; Galvez, 2019). Both Bridges (2011) and Gálvez (2019) found that women of colour frequently encountered racialised health-care experiences, often labelled as ‘at’ or ‘high risk’, and were subsequently subject to increased surveillance, interventions and sanctions. And Davis (2019) found that the medical racism of providers and institutions increased negative birth outcomes for African American women across different SES levels.

While maternal health research is extensive, there remain key limitations. In quantitative social scientific scholarship, conceptualisations of maternal health largely rely on infant health outcomes and mothers’ self-reported health status. Research on racial disparities is limited and primarily reveals disparities related to the ‘usual suspects’ (e.g. access to care and SES) without fully exploring broader social contexts and structural inequities. Qualitative sociological scholarship has been more minimal, with little focus on racial dynamics, which is a lost opportunity to unpack current racialised trends in maternal morbidity and mortality. For instance, recent qualitative research published in the JHSB reminds us that patient–provider interactions are directly shaped by provider perceptions (Fenton, 2019; Stevens, 2016). Such empirical inquiry is especially necessary in maternal health, where recent publicised cases of maternal (near) deaths point to ineffective interactions with providers.

Despite repeated visits to the hospital after giving birth to her daughter, Dr. Shalon Irving, a 36-year-old African American epidemiologist at the CDC, died three weeks postpartum. Upon waiting over 7 hours for any clinical intervention after first reporting problematic symptoms, Kira Johnson, a 39-year-old African American woman in excellent health, died from the elective c-section of her second child due to haemorrhaging. Serena Williams, a sports icon, nearly lost her life after the birth of her daughter: health-care professionals initially refused to provide her with a CT scan that she requested and failed to adequately acknowledge her pain, even though she had a medical history that placed her at greater risk for maternal morbidity and mortality. These anecdotal cases disrupt the assumptions that racial health disparities largely mirror the racial distribution of poverty within our society, as these women were not short of resources yet struggled to receive necessary care.

Furthermore, pain management research reveals that Black patients are the least likely to have their pain sufficiently treated by health-care providers (Hausmann et al., 2012) and are frequently approached as criminals, especially in the context of opioid medication prescriptions: African American patients are monitored more closely than their White counterparts (Becker et al., 2001; Hausmann et al., 2012) and labelled as drug seekers (Zickmund et al., 2015). Patients of colour are also burdened with needing to ‘prove’ their pain and are also subject to racialised assumptions that shape how their concerns are addressed (Rubin et al., 2018). Considering the high morbidity and mortality penalty for Black women, it is crucial that studies similarly examine the relationship between women of colour and providers during their pregnancy and childbirth to discern how discriminatory experiences and inequities contribute to current racial maternal health trends. Qualitative approaches would offer valuable opportunities to explore decision-making processes of both providers and patients and the nature of interactions between providers and patients of colour, which qualitative scholarship has shown frequently reproduces and reinforces broader social inequities (Shim, 2010).

END-OF-LIFE CARE

Vast multidisciplinary literatures spanning sociology, demography, health services and medicine have quantitatively documented socioeconomic and ethnoracial disparities around end-of-life. In a large international review of research on high-income countries, Davies et al. (2019) document robust socioeconomic differences in place of death and type of care received in late life, with blacks at increased risk of adverse events such as having four or more transitions in the last six months of life (Wang et al., 2019), being hospitalised, or treated in the emergency department (Rizzuto & Aldridge, 2018). Similar work has demonstrated that racial differences exist with regard to use of palliative care (Johnson, 2013) and engagement in end-of-life planning (Orlovic et al., 2019). As a backdrop to all of this, persistent disparities in pain assessments and receipt of intensive treatments have been well-documented (Hoffman et al., 2016).

Historically, qualitative sociology has been prominent in the study of end-of-life. In the 1960s, 70s and 80s, substantive studies of death and dying were abundant, supporting, for example, an Annual Review of Sociology piece (1983) and a 1975 special issue of qualitative studies on the topic in Urban Life (Lofland, 1975). Classic qualitative sociological monographs on the topic were generated in this period (Glaser & Strauss, 1968; Sudnow, 1967) as well as in more recent decades (Anspach, 1997; Timmermans, 1999; Zussman, 1992). Today, qualitative research on death, dying and end-of-life remains quite active (Bandini, 2020; Broom & Kirby, 2013; Chapple et al., 2015; Cortez & Halpin, 2020; Exley, 2004; MacArtney et al., 2016). These works have embraced thick description of the social processes and contexts that surround death and dying, pushing fields beyond quantitative conceptualisations. Unfortunately, however, they have rarely focused on racial, ethnic, or even socioeconomic disparities (Cain & McCleskey, 2019; Livne, 2019 as exceptions).

Furthermore, consistent with broader trends outlined above, there has been a dearth of qualitative research on end-of-life published by flagship American sociology journals in the past twenty years (i.e. American Sociological Review and American Journal of Sociology). In Journal of Health and Social Behavior, publications in this substantive area have largely used quantitative analyses to examine demographically oriented questions of morbidity, mortality and wellbeing (Aneshensel et al., 2004; Carr, 2003; Cornwell & Waite, 2009; Turner et al., 2017; Umberson, 2017). Qualitative exceptions in JHSB are present but rare (Bloor, 1991; Thomeer et al., 2017). Several recent monographs written by sociologists have used qualitative methods to study various aspects of ageing, end-of-life, health care and race in this time period (Abramson, 2015; Raudenbush, 2020; Rodriquez, 2014).

The implications of these methodological biases are considerable, often reifying problematic assumptions underlying quantitative analyses. For example, in a sample of Medicare beneficiaries, Barnato et al. (2009) report that minorities have a greater preference for intensive treatment near end-of-life, including Blacks being more likely to want mechanical ventilation, preferring to die in a hospital, but also being concerned about having too much medical care. By conceptualising the mechanisms for these disparities as a matter of individual preferences, knowledge or attitudes, we miss an opportunity to examine the ways racialised health care, patient–provider interaction, ecologies of care or historical trauma may contribute to black distrust of health care or cultural ideas about end-of-life planning.

Another example is how epidemiologic differences over the life course provide upstream context for end-of-life. We would expect one’s long-term health experience to shape expectations for activities of daily living, thoughts about caregiving and the role of social support. Consider: middle-aged non-Hispanic Blacks start with more chronic diseases and develop multimorbidity at an earlier age than their white counterparts (Quinones et al., 2019), and also fare worse in terms of disability adjusted life years (McGrath et al., 2019). Two late life patients with the same end-stage diagnosis, one white, one black, may have the same diagnostic ICD-10 code in a dataset but have vastly different lived experiences and decision-making processes that are not captured by quantitative measures.

These illness experiences are, in turn, nested in other social structures and environments. Having a caregiver is often touted as a key characteristic of patients who are able to use hospice or to die at home (National Hospice & Palliative Care Organization, 2019). However, authors of quantitative studies have drawn some conflicting conclusions on this matter, for example with Blacks believed to be at increased risk of not having a caregiver because they are less likely to be married (Carr & Luth, 2019) but also possibly more likely to have a caregiver as a result of living in larger households and having stronger social networks (Orlovic et al., 2019). These big data studies are ill-equipped to address the quality and racialised context of these family and caregiving structures, and whether they translate to end-of-life caregiving. Qualitative studies, whether ethnographic, interview, historical or content analyses, are well-positioned to examine the meaning and processes that lead up to these data snapshots.

American systems around death and dying – including hospice – are not race-neutral, but grow out of a specific racialised history and therefore must be studied with intersectional and multilevel lenses of analysis. Questions of what patients have been told, how they have been pressured or supported, whether they have received goal-concordant care and how those narratives are passed through social support systems over time are all part of understanding assessments of quality of care or medical decisions. The role of medical providers must be similarly and seriously assessed. Assumptions and incentives related to the value of hospice itself are deeply embedded in history of medicine and health policy, and those racialised contexts contribute to conceptual frameworks and resulting research agendas. Qualitative approaches are needed to not only clarify assumptions underlying treatment decisions and demographic analyses, but also examine pathways, processes and contexts connecting these domains as end-of-life trajectories unfold.

CONCLUSION

We began this article with the observation that scholarship on race and health published in sociology's flagship US journals lacks methodological diversity, and this trend is generally consistent with sociological literature broadly. Despite the historical richness of qualitative studies of race in sociology generally, and of qualitative research within the subfield of medical sociology, this pattern has been robust over time and in recent decades it has become more acute. In the past decades, we have instead seen a focus on quantitative approaches and epidemiologic and demographically oriented research questions. In this article, we have explored three substantive cases related to race and health – the healthy immigrant effect, maternal health and end-of-life care – to examine opportunities for qualitative inquiry as well as potential limitations and distortions in our knowledge that may emerge from a penchant for methodological and ontological homogeneity. These alternative research approaches have the potential to not only shed light on context, processes and lived experiences that are essential for understanding race and health, but also help resolve mixed results and paradoxes that are evident in quantitative literatures in these areas.

There are many other cases in the arena of race and health that would similarly benefit from increased qualitative attention, and our selected cases are intended to offer some general observations that may speak to other areas of inquiry. First, our cases speak to the importance of not treating race as a proxy for other characteristics. Rather than seeing race as an individual demographic characteristic, the lived experience of race reveals some of the ways various social institutions and policies become racialised over time. To only seek explanations at the level of the individual or family is a major limitation for our knowledge, inviting continuations of outdated deficit models when we could instead be asking how racism is built and sustained by various social structures. More centrally, our cases illustrate the urgent need for generating valid understandings of context, meaning and processes of race, not only for individuals, but also cultures, institutions, policies and the ways in which these domains interact with one another.

We speculate that race and class confounds, which can be challenging to evaluate qualitatively, may be part of the reason that we see little qualitative work on race and health. This begs the question: Is it better to over- or under-engage race given fear of confounds? Since we know from quantitative data that unexplained race effects exist, it seems not only intellectually myopic but also unethical to avoid the pursuit of explanations of how those racialised effects operate. Erring on the side of attributing disparities to socioeconomic status by default, or avoiding race altogether, has the effect of perpetually relegating race to the far reaches of our inquiry. The risks of this approach, as evidenced with the exemplars we present above, include compromised validity, deracialised assumptions and distorted agendas. Furthermore, per our opening discussion of race and sociology, we observe that urban ethnographers do not shy away from drawing conclusions about how race operates in ethnographic settings – in other words, it is feasible and necessary.

While we have focused on the outcomes of these patterns in research dissemination, we also recognise that there is likely a diversity of upstream causes as well as potential challenges in the implementation of our recommendations. Without more data, much remains unknown about where and how patterns emerge in the publication pipeline. We are not able to assess the types of research submissions received by flagship journal editors in the years we evaluated, who can of course only publish from the pool of submissions they receive. It may be that the sociological workforce lacks the diversity (in training, in experience) to feel compelled to engage these lines of research. (However, we believe this seems implausible given the representation of qualitative dissertations being published in sociology). It may be that there are interested researchers, but the logistical hurdles to conducting time- or resource-intensive qualitative research are obstacles for career, funding, or other related reasons. It may be that authors doing qualitative research on race and health disproportionately submit their articles to non-flagship outlets, or publish them as monographs. Indeed, the data above show that the proportion of qualitative publications on race and health is greater across Sociological Abstracts than in the flagship journals, although that increase is only from zero to three percent. It may be that flagship journals disproportionately lack reviewers or editors with the expertise to review these kinds of submissions, thus affecting their likelihood of being published and reifying methodological biases in journals. Perhaps decision makers are hesitant to publish such work out of other concerns such as reproducing stereotypes, exploiting communities of colour, or facing challenges in ethically recruiting marginalised persons to participate in research. In the end, we lack transparency about the upstream factors preceding the dissemination patterns we observe, but we suspect it is a multifactorial problem.

We have focused on research published in specific flagship sociology journals because we assume these outlets signal key priorities, strengths and results from sociology generally (ASR and AJS) and medical sociology specifically (JHSB) within the United States. While some of our intellectual neighbours engaged in scholarship on race and health have deep qualitative traditions (e.g. anthropology and science and technology studies (STS)), others either do not use qualitative methods (e.g. demography) or have a more limited tradition in that area (e.g. medicine, health services research). To the extent that quantitative research agendas thrive in sociology to the exclusion of qualitative work, we are concerned that qualitative scholarship on race and health may wither. Even more likely, qualitative scholars of race and health will disperse to areas beyond sociology, publishing in journals that are more ontologically diverse, accepting positions in departments that are more open to these approaches (e.g. African American studies), and cultivating research networks outside of sociology. Indeed, the cases we describe above suggest such changes are already afoot.

In the context of our three cases and the various sorts of qualitative race and health investigations occurring in neighbouring disciplines, we believe sociology has the potential to take a leading role in cultivating a more sophisticated research agenda around health and race. Taking such a role is contingent on the extent to which we are willing to continue to revisit our historical roots, tools, concepts and methods. The point here is not only that quantitative studies can be improved with more accurate measures or models of race, but that we need to fundamentally e-brace these varied forms of knowledge production – an endeavour that requires not only broad intellectual engagement but also collective investment in the research, professional and educational infrastructures that support the diversity inherent in our sociological enterprise.

DATA AVAILABILITY STATEMENT

Data sharing not applicable to this article as no datasets were generated or analysed during the current study. This is a review article that reviews current scholarship published in the field of Sociology.