Vaccine Hesitancy and the Accumulation of Distrust

ABSTRACT

Scholarship on vaccine hesitancy portrays racially marginalized populations as undervaccinated, undereducated, or under the influence of social movements. However, these explanations cannot account for vaccine hesitancy among the Somali diaspora in Minneapolis. Drawing on interviews with Somali parents and health, education, and government professionals in Minneapolis, we argue that vaccine hesitancy among marginalized populations stems from accumulated distrust. Somalis’ distrust is relationally produced through their interactions with the healthcare system, where they experience both epistemic and corporeal harm. When health experts ignore Somalis’ history, knowledge, and embodied experiences, distrust accumulates. Our account reveals different expressions of vaccine hesitancy, thus highlighting the contingent, relational, and cumulative nature of distrust.

vaccine hesitancy, trust, distrust, race/ethnicity, immigration

Since the beginning of the COVID-19 pandemic, news of Black Americans’ distrust of vaccines, their resistance to vaccine clinical trials, and their broader distrust in American medicine has circulated widely. And yet, before the coronavirus pandemic, Black Americans’ distrust of biomedicine was rarely linked to vaccine hesitancy. Throughout this paper, we use the term “vaccine hesitancy” to refer to the broad array of context-specific doubts and concerns people may have about vaccines, despite their availability (MacDonald and Sage 2015; Yaqub et al. 2014).¹ Most media and scholarly attention to vaccine hesitancy focuses on the vocal movement commonly referred to as “anti-vaccine,” composed of white, middle- and upper-class Americans. When racially marginalized groups are discussed in relation to vaccine uptake, they are portrayed as undervaccinated, undereducated, or under the influence of the highly organized vaccine hesitancy movement. And yet, these explanations cannot account for vaccine hesitancy among Somali Americans in Minneapolis, who make purposeful and knowledgeable decisions about vaccine refusal and delay for their children. Vaccine hesitancy among Somali Americans raise critical sociological questions about the intersection of medical trust/distrust, immigrant status, and race.

We were first made aware of Somali parents’ concerns about vaccination in 2011, when news surfaced of a measles outbreak concentrated within the Somali diaspora of Minneapolis (Karnowski 2011). A county epidemiological report conducted in 2013 found that the rate of Measles Mumps and Rubella (MMR) coverage in the Somali community had dropped from 90 percent in 2007 to 55 percent in 2011 to 30 percent in 2013 (Bahta and Ashkir 2015; Fieldnotes 6/13/13). In 2017, after another measles outbreak among Somalis, the county found only 42 percent of Somalis had been immunized with the MMR (Minneapolis Department of Health 2017). Among members of the Somali diaspora in Minneapolis, vaccine hesitancy is linked directly to fears about autism.² As we will explain, the prevalence of autism is high in the Somali community in Minneapolis (Hewitt et al. 2016), and Somalis believe they are especially vulnerable to disabling forms of autism, including nonverbal communication, intellectual disabilities, self-injury, and aggressive behavior.

Based on interviews with Somali parents and health, education, and government professionals, this paper takes up the case of vaccine hesitancy in the Somali American community of Minneapolis to argue that many racially marginalized populations experience an accumulation of distrust in relation to health systems and experts. While the accumulation of distrust is rooted in historical and structural inequities that lead marginalized people to question the legitimacy of science and medicine, we focus in this paper on the experiential and interactional level of distrust. More specifically, we document Somalis’ experiences of what we term epistemic and corporeal harm in the U.S. healthcare system, revealing how distrust accumulates through harmful interactions with experts, producing uncertainty about vaccination for some, and outright rejection for others.

We distinguish between epistemic harm and corporeal harm to unpack the black box of medical distrust, highlighting diverse experiences of alienation in health interactions that lead to vaccine distrust. “Epistemic harm” identifies what happens when marginalized groups’ realities, lived experiences, and knowledge systems are denied or disbelieved in interactions with health authorities. “Corporeal harm” describes what happens when marginalized groups believe that their own or their families’ bodies have been physically harmed in discriminatory interactions with health systems. Both epistemic and corporeal harms accumulate over time in interactions with the healthcare systems and public health experts, layering to produce vaccine hesitancy. Further, we find that “vaccine hesitancy” encompasses a diverse set of beliefs and strategies that are routinely misunderstood. As we will show, experiences of distrust accumulate differently through interactions with health professionals, resulting in distinct types of vaccine hesitancy.

Our empirical investigation of people’s pathways to trust and distrust reveals that scholars and commentators have been wrong to homogenize racially marginalized groups’ vaccine decisions. We don’t just ask if people are vaccine hesitant, but how vaccine hesitant beliefs and practices are generated over time within marginalized communities – as sense-making and knowledge production projects. In this way, our accumulated distrust framework reveals distrust to be a practical response to: 1) medical and public health experts’ denial of marginalized groups’ situated knowledge (epistemic harm); and 2) incidents with Western medicine that are considered injurious to body/mind (corporeal harm). As such, we build on literature in medical sociology that theorizes actors’ medical knowledge and practices as both embodied and cumulative (see Shim 2010) – as well as patterned by systemic exclusions.

Vaccines are produced and monitored at the global and national levels but reach into the most intimate spaces of our familial lives (Leach and Fairhead 2007:2). The global implementation of mandatory vaccinations has demonstratively reduced the incidence of infectious diseases worldwide, but vaccine biopolitics are complicated. They are represented as a universal good, but many perceive vaccination as forced assimilation to a Western model of health. As Emily Martin explains, “accepting vaccination means accepting the state’s power to impose a particular view about the body and its immune system – the view developed by medical science” (1994:194). Marginalized populations understand that their lives are undervalued by biomedicine and by the state, undermining their sense of inclusion in the social contract.³ Racial disparities, discrimination, and medical abuse belie inclusionary protectionism. As medical sociologists have shown, when immigrants experience consistent, degrading interactions with medical professions and neglect within U.S. healthcare settings, their trust in abstract systems (including vaccine programs) is undermined (Wright and Perry 2010). It is our contention that trust is the theoretical link necessary to understand vaccine decision-making as it intersects with marginalized groups’ discriminatory interactions with healthcare systems and experts. This paper explores the racialization of vaccine policy and refusal.

BACKGROUND: VACCINE HESITANCY

COVID-19 has brought a renewed focus on the politics of vaccine refusal. Before COVID-19, hesitancy around childhood vaccinations was already on the rise.⁴ A 2010 nationally representative household survey found that more than 1 in 10 parents of young children use an alternative vaccination schedule (Dempsey et al. 2011). And according to the CDC, 2017 was the third consecutive year that vaccine exemptions increased across the United States (Mellerson et al. 2017).

According to the media and public health experts, vaccine hesitancy stems from: 1) a knowledge deficit; 2) influence by an organized vaccine hesitant movement; and/or 3) barriers to access. Sociologists, on the other hand, have argued that vaccine hesitancy can be explained by the neoliberalization of healthcare, which individualizes decision-making and discourages collective concern for “the herd.” While each of these explanations has merit, none is sufficient for explaining the Somali case.

First, public health experts often assume that people are unfamiliar with infectious diseases because these diseases have been eradicated in the United States, they are unaware of the required vaccine schedule, and they do not understand herd immunity.⁵ For example, to explain the 2015 measles outbreak at Disneyland, CDC Director Thomas Frieden suggested: “They’re folks who just may not recognize, one, that measles is still here, two, that it can be very serious, and three, that by not vaccinating they’re not just putting their own kid at risk, they’re also putting other... vulnerable people... at risk” (CNN 2015). This is the “knowledge deficit” explanation of hesitancy.

The second explanation proffered by public health experts is that vaccine hesitancy is on the rise due to recruitment efforts by those in vaccine hesitant movements, led by politically conservative, middle-class, educated white Americans. These activists are often described as taking advantage of “insular” communities, who are portrayed as vulnerable because they do not know better. This was the case in the media coverage of a 2019 measles outbreak in an Orthodox Jewish community in Brooklyn, who were portrayed as “insular and close-knit,” and, therefore, easily influenced by vaccine hesitant movements (Reddy 2019). Presuming that “insular” marginalized communities are less educated and easily manipulated invokes racial stereotypes of these communities. As Blume (2006:638) explains, joining “anti-vaccine” groups is more often the result – not the cause – of parental concerns.

A third explanation is that marginalized people encounter barriers to access, and there are indeed racial disparities in vaccine uptake. In 2016, although the overall rates of coverage were high, the CDC found that childhood vaccination coverage was lowest among Black children, children living below the poverty line, and children who lacked health insurance or who were covered under Medicaid (Hill et al. 2017). Using demographic mapping, Berezin and Eads (2016) reveal the uneven landscape of vaccine coverage – showing that the lowest vaccine rates are in both wealthy enclaves and “low-income and low education” areas. They attribute the cause to lower education levels and poverty rates. However, it is rare for social scientists to provide qualitative data on marginalized communities’ beliefs about vaccines. Marginalized communities are presumed to be unable to vaccinate but are rarely portrayed as choosing not to vaccinate.

Unlike public health experts, social scientists explain vaccine hesitancy by situating it as part of neoliberal ideologies and infrastructural shifts. Neoliberal transformations have made individuals responsible for their own health decision-making and transformed medicine into a consumer “choice.” Elite parents who are vaccine hesitant embody the neoliberal norm to become “expert parents,” responsible for addressing each individual risk their child faces (Kaufman 2010; Lakoff 2015; Reich 2014; Senier 2008). In a risky world, parents face contradictory information about safety and harm, and they assume the burden of responsibility to consume that knowledge and act upon it (Kaufman 2010:23). Many even feel as though blindly trusting vaccines is shirking one’s parental duty (Yaqub et al. 2014). Neoliberal parenting dovetails with the increasing medicalization of childhood behaviors, which expands the domain for parental decision-making and intervention and situates therapeutic professionals as arbiters of childhood “normalcy” (Rafalovich 2013; Sweet and Decoteau 2018). Medicalizing childhood behavior is racially stratified, with children of color more likely to be diagnosed with disorders that warrant harsher interventions, which can also contribute to greater expressions of medical distrust (Bosk 2013; Decoteau 2021).

Jennifer Reich argues that class and race privilege are essential for understanding vaccine hesitancy because not everyone has the same access to “choice” (2014, 2016). More resources, she argues, allow for more freedom from state mandates as only knowledgeable parents can easily navigate state bureaucracies without inviting surveillance. Families of color who opt out are more likely to face state sanctions. For example, welfare recipients have been punished if their children are not fully vaccinated (2016:14-15). Reich’s explanation is well-founded: higher median household income and whiteness are predictors of vaccine exemptions (Yang et al. 2016). Estep and Greenberg further argue that residential and school selection processes in the United States allow elite parents to self-segregate into “pockets of homogeneity,” in which they can avoid criticism of their vaccine hesitancy (2020).

And yet, while explanations that attribute vaccine hesitancy to neoliberal parenting are well-supported, they do not hold in the case of Somali communities, who do not articulate neoliberal ideologies of parenting or individualized approaches to health decision-making. Somalis in Minneapolis have little difficulty accessing vaccines, are very familiar with measles, and they forge alliances with broader vaccine hesitant movements in purposeful ways. Existing public health and sociological theories cannot explain Somali parents’ vaccine hesitancy – in part because there are few studies that focus on marginalized communities’ vaccine beliefs.⁶ While people of color may face sanctions when they are “non-compliant” (Reich 2016), marginalized populations nevertheless make choices and develop knowledge about vaccines. In fact, marginalized groups’ biographical experiences of health disparities may be at the root of their distrust of vaccines. As Jamison, Quinn, and Freimuth (2019:92) note, blindly trusting in vaccines is also a privilege. Therefore, we require a different framework to make sense of marginalized communities’ vaccine concerns. To better understand the relationship between social marginalization and vaccine hesitancy, we turn to the sociological literature on trust and race/ethnicity.

TRUST/DISTRUST, RACE, AND MEDICINE

Questions of trust are central to vaccine policy. Sociologists have analyzed how groups’ trusting relationships with each other and with experts matter for decision-making. Scholars have been particularly interested in how racial groups differ along these lines. Despite advances in sociological work on trust, this literature suffers from two problems that we seek to correct. First, literature on trust tends to represent race as an “interrupter” of general social trust. In this formulation, racialized groups are depicted as broadly distrusting in a way that homogenizes their experiences and reiterates a deficit model of race (see Benjamin 2014; Wilkes 2011). This issue replicates some of the problems we have already outlined in the way vaccine hesitant people are represented in the public health literature. We argue instead for thinking about distrust as accumulating over time and as productive of distinct kinds of knowledge. Second, literature on distrust in medicine, while more nuanced in its approach to race, tends to focus on histories of racialized medical violence and their role in producing group-based distrust. Intervening in this literature, we show that trust and distrust are produced through ongoing, situated experiences with health experts and local communities, which affect marginalized groups’ knowledge claims and embodiment. Our approach to trust and distrust is therefore rooted in epistemic and corporeal experiences. Finally, the literature on trust tends to bifurcate “Black Americans” from “immigrants” in a way that ignores Black immigrant experiences.

Much of the sociological literature on trust is motivated by a core assumption that trust is a normative good, a necessary social glue. Trust in others, scholars argue, is the stuff that holds liberal democracies together. Statements such as, “generalized trust is vital for a healthy society” (Glanville 2015:30) are common. Alongside these normative claims is an anxiety that trust is currently “under threat:” eviscerated welfare programs lead to distrust of the state (Mewes and Giordano 2017); widespread distrust in expertise abounds (Eyal 2019); people have lost faith in political parties and institutions.

By suggesting that racial diversity causes distrust to foment, anxieties about threats to the social order subtly racialize distrust. As Sandra Smith argues, race is the “life experience” that has the biggest impact on trust (2010). The racialization of trust is especially relevant to what researchers call “generalized trust” (Smith 2010), wherein trust is a characteristic of individuals or groups that can be measured through surveys or experiments – a psychological state that stems from accepting one’s own vulnerability in relation to the intentions of others (Glanville 2015). It is common for researchers to suggest that Black Americans experience generalized distrust, which is represented as a rupture to the normative order. As Rima Wilkes puts it, “Most of the literature on trust continues to imply that black Americans’ distrust rather than white Americans’ trust is the deviation that needs explaining” (2011:1598). If trust is needed for the good life, Black Americans’ distrust is understood as an obstacle to collective flourishing.

Studies that measure generalized trust usually rely on General Social Survey (GSS) questions about the fairness and trustworthiness of “others.” Both formal and informal ties are linked to positive generalized trust (Glanville, Andersson, and Paxton 2013). Generalized trust is thought to be conducive not only to a healthy society, but to healthy individuals (Mewes and Giordano 2017). Most studies consider generalized trust to be a “stable psychological propensity” of groups, turning “distrust” into what Ruha Benjamin calls “culturally essentialist beliefs” about trusting versus non-trusting groups (2014). In this way, “trust talk” reifies racial group boundaries (Benjamin 2016). Many scholars consider Black Americans’ distrust to be an outcome of discrimination and exploitation, which has remained relatively stable over time in the U.S. context (Smith 2010; Wilkes 2011, 2015). This history of discrimination is especially relevant for scholars interested in distrust of medicine among the racially marginalized (Benjamin 2014; Latkin et al. 2021; Sullivan 2020; see also Glanville et al. 2013; Wilkes 2015).

Scholars have also documented racial differences in strategic trust (Smith 2010), or the instrumental variety of trust in which people expect others to act in accordance with their expectations. Strategic trust is associated with social networks and social capital and is, therefore, relational (Smith 2010). In social capital theories, trust is not a psychological trait, but a social resource generated through cooperative ties (Barbalet 2009:377).⁷ In fact, the concept of trust is so heavily associated with social capital that medical researchers often use survey questions about trust as a measure for social capital itself (Carpiano and Fitterer 2014). According to James Coleman, the strength of social ties depends on an expectation that relationships are reciprocal (1988). In this model, people trust others because trusting relationships increase their social capital.

Despite their acknowledgement that trust is relational, social capital theories also employ a deficit conceptualization of race, wherein those marginalized by race are thought to have disadvantageous social networks (Desmond 2012). Diversity itself is often conceptualized as a threat to trust. For example, scholars have argued that diverse neighborhoods experience reduced social trust (Dinesen and Sønderskov 2015). But as Maria Abascal and Delia Baldassari observe, such findings are just measuring nonwhites’ lower levels of trust, since nonwhites are more likely to live in “diverse” neighborhoods (2015). Further, there is no reason to believe that all forms of trust are beneficial for “the collective,” especially because whites tend to create social networks in which they hoard resources (2015). And as Raudenbush shows, people in marginalized communities develop strategic trusting and exchange relationships with others even when they express an overall sense of distrust (2016). Assumptions about the negative – or interruptive – effects of race on trust pervade network models of trust. In our account of accumulated distrust, distrust is knowledge-generating, rather than a social capital deficit.

Building on a social capital approach to trust, scholars have also asked critical questions about trust and distrust of medicine.⁸ Medical distrust among Black Americans specifically has become a medical truism that tends to reify racial group differences (Benjamin 2014:756). This literature focuses on Black Americans’ hesitancy to engage in clinical research (Braunstein et al. 2008; Corbie-Smith, Thomas, and St. George 2002; Crawley 2000), their distrust of the healthcare system (Armstrong et al. 2006; Armstrong et al. 2008; Gamble 1997), or their distrust of health providers (Schnittker 2004; Sewell 2015). Much of this scholarship suggests that knowledge of the Tuskegee experiment⁹ leads to Black Americans’ distrust of medicine (Braunstein et al. 2008; Corbie-Smith, Thomas, and St. George 2002; Gamble 1997).

While the literature on racialized distrust in medicine takes structural context into account, as well as patients’ relationships to doctors and medical researchers, it also tends to paint Black Americans with broad cultural brushstrokes (Benjamin 2011, 2014). Black Americans are represented as fatalistic and noncompliant, suggesting there is something inherent in Black culture that leads to distrust. Attributing distrust to historical cases of abuse (like Tuskegee, Marion Sims, or Henrietta Lacks) discounts the fact that these abuses were embedded in a racist system of knowledge and control (Wasserman, Flannery, and Clair 2007) and obscures how normalized abuse and discrimination are (Benjamin 2014; Jamison et al. 2019; Washington 2006). It is much more likely that everyday forms of mistreatment and lack of adequate health infrastructure fuel distrust (Bajaj and Stanford 2021; Washington 2006). Each time racially marginalized subjects have their pain denied or explained away as drug abuse (Ciribassi and Patil 2016) or have their conditions misdiagnosed or treatment withheld (Bajaj and Stanford 2021), they experience the everyday racism of U.S. healthcare. These “daily subtle mental assaults” are more salient in explaining distrust than isolated historical events (Bajaj and Stanford 2021:e12-1).

Further, racial inequities in healthcare treatment and access also inhere in racialized epistemologies that undergird contemporary medicine. The ways in which racial logics constitute scientific consensus and the “best practices” of Western medicine are often unexplored in theories of medical distrust. But our case exploring Somalis’ “outsider” experiences with the U.S. healthcare system illuminate these epistemic causes of medical distrust. We uncover both epistemic and corporeal dimensions of medical distrust and specify how they accumulate over time in interactions with healthcare systems.

Importantly, Somalis are racially marginalized but also immigrants. Medical sociologists who study immigrant health increasingly recognize that healthcare delivery systems are determinants of health disparities (Wright and Perry 2010) and that lack of trust in abstract systems fuels this dynamic. Immigrants face barriers in access, obstacles with translation and cultural competence, as well as discrimination and devaluation by health providers (Mechanic and Meyer 2000; Portes, Light, and Fernandez-Kelly 2009). Immigrants regularly confront adversarial relationships with their doctors, many of whom also fail to recognize the structural realities that immigrants face (Light 2012; Searles 2012). Indeed, literature on immigrant health is adept at identifying “mismatches” between healthcare system capacities and immigrants’ health needs. Our accumulated distrust framework illustrates this “mismatch,” by revealing the distinct types of harm that Somali immigrants experience in interactions with health systems.

In our account, then, distrust stems from experiences of epistemic and corporeal harm in interactions with providers and public health experts. Unlike existing accounts – which tend to homogenize distrust and/or depict marginalized groups as generally “distrustful” – we highlight layered and heterogeneous processes that lead to distrust. For people experiencing ongoing discrimination and structural disadvantage, distrust is overdetermined by harmful experiences with institutional actors, compounding beliefs and ingraining distrust over time.

ACCUMULATED DISTRUST

Somali Americans are not undereducated about vaccines, they do not experience significant barriers to access, and they reject the idea that they have been unwittingly “pulled in” to the broader vaccine-hesitant movement. Further, Somali Americans do not engage in individualistic, neoliberal practices in their parenting or health decision-making. How, then, can we explain Somalis’ vaccine hesitancy? The sociological literature on trust and distrust explains that trust and distrust are relational and racialized, but highlights the need for a temporal approach to racialized interactions and their effects on knowledge and embodiment. Our framework builds on core findings in medical sociology and immigrant health to show that healthcare decision-making is a complex, embodied process that depends on social positionality.

Our accumulated distrust framework focuses on Somalis’ interactional and experiential routes toward vaccine hesitancy, which involve a series of epistemic and corporeal harms with U.S. public health experts and medical providers. While epistemic harms reference authorities’ dismissal of Somalis’ lived realities and knowledge of their children, corporeal harms refer to experiences that Somalis perceive as literally injurious to their children, regardless of provider intention. Epistemic harm might occur when a doctor does not believe a Somali parent’s description of their children’s symptoms. Corporeal harm might occur when a Somali child receives a late diagnosis of autism, resulting in more disabling expressions of autism.

These shared experiences lead some Somalis to close ranks, trusting in their compatriots’ knowledge rather than in public health and medical experts, engaging in what Raudenbush (2016) would call “selective solidarity.” Building on Ruha Benjamin’s (2011) analysis, we illustrate how distrust emerges not only from one’s marginalized position in a racialized social order, but also from embodied experiences of discrimination in medical and public health encounters. Our model, then, argues that distrust accumulates over time based on social location, interactional networks, and embodied experiences with experts.

We also show that the accumulation of distrust does not produce uniform outcomes. “Vaccine hesitancy” is not a single phenomenon, but a diverse cluster of behaviors and beliefs that is context specific. Distrust tends to endure when social experiences reinforce it. Trust is not simply forged through tacit inculcation – it is made and unmade over time in purposeful ways. Our model insists that medical trust and distrust are relational, contextual, and cumulative – rather than static and immutable. Experiences of epistemic and corporeal harm are varied, felt at different intensities, and result in divergent outcomes.

DATA AND METHODS

This paper draws on data from a broader project that analyzed Somalis’ experiences with autism, vaccines, and healthcare systems in Minneapolis and Toronto. In Minneapolis, we interviewed 67 professionals, 32 of whom were Somali. Most of the professionals we interviewed worked in public health, disability services in schools, refugee resettlement, and healthcare. Professionals were recruited through purposive sampling based on their experiences working with Somali families of children with autism. At least one of the Somali professionals we interviewed was also a parent who expressed fears about vaccination.

We interviewed 34 Somali parents about autism, vaccines, and healthcare experiences in Minneapolis. Most Somalis began immigrating to North America after the outbreak of civil war in Somalia in 1991. Approximately 25,000 people in Minnesota have Somali ancestry (U.S. Census Bureau 2013a). About one in three Somali Americans live in the state (Dunbar 2010). The Somali diaspora in Minnesota is bifurcated by education and social class. Many Somali professionals in our sample had earned higher education degrees and worked as social workers, researchers, and liaisons with Minnesota Departments of Health and Education. On the other hand, most of the parents we interviewed were poor and had limited English proficiency. The median household income for Somalis in Minnesota in 2010 was $15,902, and 60.8 percent of all Somali descent families in Minnesota live below the poverty line (U.S. Census Bureau 2013a). In addition, 46.1 percent have less than a high school diploma, and 95.1 percent do not speak English at home (U.S. Census Bureau 2013b).

Somali parents were recruited through a support group about autism and through networks of personal care attendants who work for families of children with autism. As we indicate in Table 1, 27 parents had children with autism and seven did not, though two of them had children with other disabilities. We also conducted two focus group discussions with 14 Somali parents who were fearful of vaccines but did not necessarily have a child with autism. We were not expecting parents of non-autistic children to participate, but they expressed interest in talking to us about vaccines and autism. The inclusion of these community members in our data offers an important point of comparison between parents of autistic children and other parents. Their interest in participating also demonstrates the hyper-visibility of these topics in the community.

Table 1.

Demographics of Interviewees and Children with Autism

Minneapolis Parents (N=34)
	N
Parents born in Somalia	34
Male	8
Female	26
Child with autism	27
Child without autism	7
	Age
Mean Child Age	10
Mean Child Age at first notice of autism	1.75
Mean Child age at diagnosis	3

Because parents were recruited for a project about autism, interviews focused on the challenges of resource navigation. Parents were often reticent to provide details on their migration history, educational background, and work conditions because these were either traumatizing and/or stigmatizing experiences. Starting interviews with these questions made it more difficult to build trusting relationships. This information often emerged over time and through ethnographic observation. The majority of parents we interviewed were lower income, and a small handful were very highly educated. In general, our sample reflects educational and income patterns among Somali migrants to the United States. All interviewees were asked to sign a consent form if they wanted their name used. All other interviewees gave verbal consent and remained anonymous. Parents were compensated $20 per interview. We used an interpreter during interviews when requested. Well over half of the parents in Minneapolis stated directly that they believed vaccines were a causal factor in autism, and when probed, even more indicated that they suspected this to be the case.

In addition to interviews and focus groups, the ethnographic work for this project included attending support groups, sitting in on vaccine information sessions, attending meetings of Somalis with Minnesota Department of Health employees, spending time with parents and children in their homes, and talking with interpreters and personal care assistants for children with autism. Both authors recorded fieldnotes. At times, authors’ positionalities as highly educated white women created barriers to access in these spaces. This is in part because public health officials often share our demographic profiles and parents were accustomed to receiving pressure from officials to vaccinate their children. The ethnographic and interview data for this project were collected over several years in part because it took time to build trusting connections with community partners and parents. Trust and distrust were constant features of Somalis’ experiences with health and educational experts in the United States, and this theme emerged repeatedly throughout the years the research was conducted.

Based on multiple readings of transcripts, we wrote memos about interviews and ethnographic observations to grasp emergent themes. Memos were then clustered around themes, which translated into concrete codes. Decoteau coded all interviews, combing through interview data systematically to understand parents’ sense-making about the etiology of autism. Examples of codes used for this analysis are, “vaccines as causal factor,” “use of alternative therapy,” and “no autism in Somalia.” We read interview transcripts and fieldnotes alongside relevant literature to generate theoretically informed codes. Memoing and coding were ongoing over the course of three years of data collection (2013–2016). Codes were then refined and compared across interviews and fieldnotes.

We also discussed emergent findings with Somali parents, activists, and community leaders – both to ensure validity and to share information. Parents and community members talked repeatedly of the importance of trust/distrust as we discussed findings with them, which shaped the theoretical focus of this paper. Nevertheless, findings from this research are limited by the fact that participants were mostly parents of children with autism, which means that their fears around vaccination were politicized and highly visible to health experts. While we argue that “accumulated distrust” may apply beyond this case, the way distrust accumulates for the parents we interviewed is specific to their migrant status and experiences with a childhood condition that felt bewildering and frightening to them. Future research should explore the accumulation of corporeal and epistemic harms in groups whose vaccine hesitancy is shaped by different sociological factors.

FINDINGS

Trust and distrust in Western medicine was a constantly recurring theme throughout our research with Somali communities in Minneapolis. Somalis’ experiences with the Western healthcare system start before they even migrate. According to Fartun Weli, a Somali director of a health literacy organization in Minneapolis, Somalis are primed for racism in the U.S. healthcare system before they set foot in the United States. Fartun explains: “The mistrust did not start with their experiences of the U.S. health system.... The biggest mistrust existed before they even came to the clinic. There is this story that if you are Black, the system has treated you badly throughout history. They come in with fear, mistrust” (Interview 9/17/15). Fartun told us that a woman she met, who had just arrived in the United States from a refugee camp in Ethiopia, cited the Tuskegee experiment: “This woman did not speak one word of English, but knew the term guinea pig... the mistrust starts with stories being told, even in the camps, about minorities being treated differently” (Interview 9/17/15). Much like Black Americans, Somalis cite a history of racial discrimination as proof that the U.S. medical establishment cannot be trusted. Somalis’ diasporic positioning allows them to merge a postcolonial analysis with a critique of U.S.-based scientific racism – both of which inform their distrust of Western medicine (Decoteau 2021). While knowledge of the legacy of scientific racism in the United States is a backdrop for distrust among Somalis, experiences of racism and neglect within U.S. healthcare institutions confirm this distrust, which accumulates over time.

For Somalis, these foundational expressions of distrust intensify and accrue over time as they navigate autism diagnosis and care in the United States. Somalis in Minneapolis first noticed a high prevalence of autism in their community in 2007 and early 2008 (Lerner 2008). In 2009, the Minnesota Department of Health (MDH) published a study revealing that children of Somali parents in Minneapolis were 2–7 times more likely to receive Autism Spectrum Disorder special education services than children of non-Somali parents (MDH 2009). The Centers for Disease Control (CDC) conducted an Autism and Developmental Disabilities Monitoring (ADDM) surveillance study from 2011–2012 in Minneapolis, which confirmed a high prevalence rate of ASD in the Somali population (Hewitt et al. 2016). One hundred percent of the Somali children identified with autism in the CDC study had an accompanying intellectual disability, confirming Somalis’ sense that they suffer from more disabling forms of autism.

Local public health experts responded quickly and intensely to Somalis’ fears about vaccines when the high rate of autism in their community was first documented. MDH held a community forum with members of the Somali community, in which information about autism and vaccines were discussed (Interview 11/14/15; MDH 2009). Public health officials also launched a series of educational programs targeting Somali residents. They employed Somali Americans as translators and cultural liaisons to run workshops and door-to-door campaigns. And yet, experts involved in these programs regularly failed to acknowledge Somalis’ experiences and knowledge of autism. Somali Americans call autism the “Western disease,” because they claim autism does not exist in Somalia. Minneapolis-based public health experts often deny these beliefs, insisting that autism did exist in Somalia but was not recognized as such. To Somalis, this is a rejection of their reality and distinctive experience with autism. As such, Somalis regularly experience expert health advice as epistemic harm.

Further, Somalis feel that doctors and health officials evade questions about their distinctive experiences with high rates of disabling autism. Jama Duale who attended the community forums held by MDH explained, “they’re talking to us to convince us that shots [are] not causing the autism... [but] why is every sixth family in the Somali community [affected by autism].... Some families, they have four or three [kids with autism]” (Interview 11/7/15). Somalis’ beliefs about vaccination are intimately tied to their belief that they are particularly vulnerable to disabling forms of autism. When public health experts deny Somalis’ interpretations of this vulnerability, Somalis experience epistemic harm and a disparagement of their strategies for coping with an illness overwhelming their community. To compound epistemic harm, many Somalis also believe that their children have been injured by vaccines and/or that epistemic harms have negatively affected their children’s health, resulting in bodily suffering. In what follows, we illustrate epistemic and then corporeal harm.

Epistemic Harm

We first met Fatuma because she was working as a community educator for the Immunization, Tuberculosis and International Health department of MDH in 2013 and 2014. When we returned to conduct interviews in 2015, she had stopped working, married, and had two children. Even though she had worked for MDH convincing other Somalis of the safety of immunizations, she was having doubts about whether or not to give the MMR to her own son who was six months old at the time.

I had my son, and then, at two months... he got some shots.... He got sick, a fever... I told the doctor, “I’m not gonna give him all the shots at one time”... they say the MMR doesn’t cause... autism... I believed the information... what I learned from MDH, but then, now, I don’t know. I’m conflicted... this small voice in the back of my mind says, “What if it’s true? What if this vaccine is problematic? Who do I trust?” (Interview 9/18/15)

This is not a knowledge deficit. Fatuma is highly educated and taught vaccine safety to others. Nonetheless, there is a “small voice in the back of her head,” and she does not know whom to trust. Even though Fatuma is aware of the broader narratives in both the Somali community and among public health experts and doctors, this experience of injury with her son made her question medical authority and knowledge, turning toward community explanations.

We found that many Somalis trust their compatriots’ knowledge over that of their doctors. In fact, many felt intimidated by their doctors around their vaccine choices. One mother told us, “I refused the 18-month shots.... My doctor is intimidating. He drives you crazy... and he really intimidates you a lot.... One lady I know refused to go back to him because he intimidated her about the vaccines and she has a son with autism (Interview 6/11/14). Experiences of intimidation often cement Somalis’ beliefs in vaccine explanations for autism, compounding medical distrust.

Somalis also have broader negative experiences in seeking out healthcare. In a community participatory study of Somalis in Kansas City, Filippi and colleagues found that cost, insurance, language, transportation, and navigating the system were all barriers Somali faced in accessing healthcare, leading many to avoid doctors unless they faced a medical emergency (2014). Another study conducted among Somalis in Minnesota found that Somalis are poorly served by the state’s health and social service systems (Pavlish, Noor, and Brandt 2010). Somalis reported lack of health insurance, difficulty navigating the system, experiences of racism, communication barriers, and being treated by health professionals as ignorant. Fartun told us: “because they are immigrants and Black, there is a perception that they will be met with racism in the healthcare industry... they think [they will get] substandard care.... Doctors also see them [immigrants] as really needy – they look at immigrants and all they see is need... it debilitates peoples’ self-esteem (Interview 9/17/15). Fartun explains that doctors often implicitly employ a deficit model when engaging with immigrant groups.

In addition, there are real barriers to establishing trusting doctor-patient relationships due to language barriers. Some Somalis use interpreters, which can jeopardize the establishment of trust. As one interpreter explained, “It’s hard for them to build trust with doctors because they have to go through an interpreter. You know? You can only build trust when you’re directly communicating with someone” (Focus Group Discussion 9/12/15). Cultural and language barriers can exacerbate the social distance between Somali patients and their doctors and can compound Somalis’ distrust in the medical system. In this way, Somalis’ own theories of embodiment and knowledge of their families and histories is ignored or simply inaccessible, leading to a lack of recognition of what Janet Shim terms their “cultural health capital,” the socially transmitted and embodied set of health skills and knowledge (2010). These clashes of cultural understanding are common for immigrant and refugee communities and can fuel growing distrust.

The narrative that vaccines cause autism circulates widely among Somalis, and Somalis tend to trust their compatriots over Western health officials, engaging in “strategic solidarity” rather than generalized trust in experts (Raudenbush 2016). For them, vaccine delay is a pragmatic choice – it is something they can do to protect their children (see Brownlie and Howson 2005). When doctors and public health professionals deny that Somalis are uniquely vulnerable to autism and insist that their experiential knowledge of vaccine dangers are not real, Somalis experience epistemic harm. Because Somalis already experience a series of “subtle mental assaults” accessing healthcare as immigrants, epistemic harm related to vaccines and autism compounds their negative perceptions, leading to an accumulation of distrust.

Many Somalis have also joined forces with existing vaccine hesitant movements. The media has represented Somalis’ relationships with these movements as exploitative rather than collaborative. The portrayal of Somali parents as being “duped” by vaccine hesitant movements is another instance of epistemic harm.

The dominant explanation public health advocates provide for vaccine hesitancy among Somali Americans in Minneapolis is that they were hoodwinked by local “anti-vaccine” activists and by national leaders such as David Kirby (journalist and author of a book about vaccine harm) and Andrew Wakefield (author of a Lancet article linking MMR and vaccines that was later retracted), who both met with the Somali community on multiple occasions beginning in early 2009 (Gahr et al. 2014; Karnowski 2011). Many local public health workers blame Wakefield personally for significant drops in MMR coverage within the Somali community (Fieldnotes 6/13/13, 10/23/15; MDH Interview 11/4/15).

One parent who attended the meetings with Wakefield in 2009–2011 describes how the media distorted his first visit.

Wakefield didn’t tell us [that vaccines cause autism]. The fact was there... [For] some people, their kid was speaking, and when they took them [for the vaccine]... they had fever for days.... Wakefield was around the city, all the news was there. All of them blaming him. “Oh, these poor Somalis, they don’t speak English, blah, blah. He’s just telling them all this.” It was nasty. (Interview 10/7/15)

Somalis invited Wakefield and Kirby to Minnesota because of what they were seeing in their children: “the fact was there.” These parents see themselves as active knowers seeking out allies and resources. When public health experts suggest Somalis were “duped” by social movement actors, this causes additional epistemic harm.

Another source of epistemic harm, for Somalis who have children with autism, stems from their experiences with autism experts. Somali parents’ experiences with autism are alienating: the diagnosis is unfamiliar to them, and many have more than one child with disabling symptoms, including intellectual disabilities, lack of communication skills, and self-injurious and aggressive behavior. They also face barriers to accessing educational and therapeutic supports for their children (Decoteau 2021). Therefore, when public health and medical professionals insist that autism exists in Somalia, and that Somalis are not vulnerable to more disabling forms of autism, these parents are especially aggrieved. They experience a more profound sense of epistemic harm than other parents because professionals negate their lived experience of raising a child with a disability. The knowledge Somalis have built in the United States, based on their collective experiences of autism risk and care, contributes to their cultural health capital (Shim 2010), but is denied by public health experts and caregivers. For example, one parent we interviewed attended educational forums sponsored by the MDH about the prevalence of autism in the Somali community because she wanted to learn more about autism. Quickly, however, her anxieties were eclipsed by a narrow focus on vaccine compliance. The experts told her: “immunize, immunize, immunize.... I’m not saying the doctors are prejudice[d], but the doctors are not giving advice to the parents.... They are not helping us understand the system or giving us information. What they’re giving you is, ‘your kid, he needs the booster’” (Interview 10/7/15). This mother faced obstacles finding resources for her son with autism, but public health experts made her feel as if they cared only whether she vaccinated him. Distrust therefore accumulates when parents’ concerns about autism are eclipsed by a strong public health focus on vaccination.

The concept of epistemic harm identifies what Somalis perceive as interactive assaults on their knowledge systems and lived realities. Further, representations of Somalis’ engagements with vaccine hesitant movements have robbed them of agency as knowledge producers about their children’s health. These epistemic harms accumulate, resulting in stronger expressions of distrust for parents of children with autism who have even more dismissive interactions with doctors and other authorities. To compound epistemic harm, many parents also believe that their children have been physically injured by their interactions with Western health systems. Next, we show how experiences of perceived corporeal harms compound epistemic harms.

Corporeal Harm

Most of the parents we interviewed who experienced corporeal harm have children with autism, whom they believe were vaccine injured. Many of these parents also believe their children have experienced additional corporeal harm as they navigate autism treatment and services, such as delayed diagnosis. Many become active vaccine resisters, and others simply decide not to vaccinate their other children. The first mother we interviewed in Minneapolis told us her story of vaccine harm, which we later discovered was a prevalent account.

There was a lot of shots we received when we came [to the United States]. After them, my son lost his words.... The specialist says, “It looks like autism.” We got shots in Kenya [in the camps]... Kenya, the shots he got, I brought it to them, all that list... I believe they might gonna give the shots he already got.... I think that might be the problem. (Interview 10/16/14)

Even though this parent brought documentation of the vaccinations her son had received in the refugee camp, the doctors in the United States gave him the full battery of vaccinations, which she believes caused her son’s autism. All refugees who lack proper documentation of previous vaccines are given all recommended immunizations once they are resettled (Lifson, Thai, and Hang 2001). Many Somalis believe that they and their children were given “double vaccinations,” but these concerns are regularly ignored. Public health experts often overlook the biographical experiences of migrants that inform decisions about their health.

Another mother told us that she thinks the MMR caused her son’s autism because after getting the vaccine, he slept so long that she was unable to get him up. After that, he started showing symptoms of autism. For her second son, she waited to give him the MMR until he was three and “talking talking talking.” She said that was the only vaccine she avoided (Interview 10/24/15). This example reveals important heterogeneity among those who refuse vaccination, since this mother refused vaccines only temporarily for her second son.

Another mother thinks that the number of vaccines given to children at such a young age is problematic, whereas vaccines are more spread out and given at older ages in Somalia. One of her sons is autistic and the other has food allergies. This mother believes the combination of poor food quality, pollution, and vaccines accumulates in children’s bodies to cause a series of sensitivities (10/7/15). Several parents explained that their children had immediate responses to the vaccines – seizures, high fevers, blank stares (11/7/15, 11/1/15) – and then symptoms of autism began to emerge. In this way, parents who refuse vaccination have mostly experienced epistemic injuries in combination with witnessing bodily suffering in their children, which they attribute to vaccines.

Another form of corporeal harm occurred when American doctors failed to diagnose Somali children with autism in a timely fashion. This was more likely to occur in the first ten years that Somalis were resettled in Minneapolis. A Somali physician, Dr. Osman Harare, who practiced in Somalia but serves as a community liaison and health educator in Minneapolis, told us that Somalis distrust their doctors because of this failure to diagnose. Many of these children did not get a diagnosis until they started school at five years old. Late diagnosis often leads children to express more disabling forms of autism because of delayed treatment. Because doctors may lack knowledge about autism or fail to recognize symptoms in Somali children, these parents stopped trusting their doctors’ knowledge – and it was therefore easier to believe autism was caused by the MMR (Interview 10/15/14). In fact, several parents we interviewed were irate that their doctors failed to diagnose their children at early ages, when they knew something was amiss in their children’s developmental trajectory. One mother told me she fought to get a proper assessment for her child and was told again and again that nothing was wrong, and when his diagnosis was finally confirmed, she explained: “First, I am crying. Very pissed off. I was so upset with the doctor... very upset to the point that I complained. I said, ‘This was very unprofessional’” (Interview 10/6/15). Because of their own presumptions about race, nationality and autism diagnosis, U.S. health experts misrecognized autism in Somali refugee children. Racial disparities in diagnosing autism have been documented among Black American children as well, because autism is racially coded as white in the public imaginary (Decoteau 2021).

The concept of corporeal harm refers to parents’ perception that their children’s bodies have been injured by interactions with American medical systems and experts, especially via vaccine injuries and late diagnosis. Experiences of corporeal harm are more common for parents of children with autism, and this type of harm compounds epistemic harm in a way that leads to more intense distrust of vaccines. With corporeal harm, Somalis still have their knowledge systems and lived realities questioned in health-related interactions, but they also experience what they perceive to be bodily injury to their children. Corporeal harms not only feel frustrating and exclusionary like epistemic harms, but violent and more permanently alienating, leading to a greater sense of distrust. The concept of corporeal harm is important for understanding vaccine hesitancy more generally, since it reveals that parents are not simply responding with hesitancy to knowledge claims in health/medicine, but also to the sense that their children have been physically harmed through interactions with American medicine.

CONCLUSION

In this paper, we have shown that distrust emerges from accumulated experiences of epistemic and corporeal harm in interactions with local public health and medical experts. Existing accounts of vaccine hesitancy tend to underestimate the knowledge and strategies marginalized communities employ in their decision-making. This portrays vaccine hesitancy as a deficit. But Somalis have also forged explanations for their vulnerability to autism and have crafted strategies of resistance that draw on their situated knowledge as racialized migrants. We have shown that scholarship on vaccine hesitancy regularly fails to recognize experiences of epistemic harm in interactions with public health and medical professionals. When peoples’ knowledge and sense-making is denied, they tend to distrust the source. The power of this epistemic harm is under-explored in the current literature.

The framework of accumulated distrust may be useful for explaining other cases of medical distrust among marginalized communities. For example, in the case of the COVID-19 vaccine, it is commonly suggested that Black Americans have a general distrust of the medical establishment, leading to their hesitancy (Rusoja and Thomas 2021). And yet, there is evidence to suggest this explanation may be too general. Others have illustrated it is not simply Black Americans’ knowledge of historic abuse, but their continued mistreatment by physicians in clinical interactions, that fuels distrust (Bajaj and Stanford 2021). One Kaiser Family Foundation report found that adults without health insurance had the lowest vaccine uptake (Hamel et al. 2021). This finding points to the failures of health infrastructure to meet the needs of the most vulnerable. A recent study on COVID-19 found that many Black Americans who were vaccine hesitant discussed how a singular focus on COVID-19 by healthcare professionals and city officials obscured other ongoing crises communities were facing, such as opioid overdose (Decoteau et al. 2021). Interviewees cited a kind of misrecognition, by medical professionals, of the real issues communities were facing. Beyond simply a generalized sense of medical distrust stemming from knowledge of historic abuse, these examples suggest that COVID-19 vaccine hesitancy among racially marginalized groups may also result from experiences of accumulated distrust based on both epistemic and bodily harm enacted within health systems and by experts.

Therefore, accumulated distrust may be a useful framework beyond the case of Somalis in Minneapolis. We do not intend to suggest that all racially marginalized groups are distrustful of medicine. Rather, we wish to highlight the fact that medical distrust is highly variable and contingent on groups’ unique social locations and patterns of expert interactions. Articulations of accumulated distrust tend to be overdetermined by a series of pedantic and injurious experiences. Taking these experiences seriously has led us to the empirical findings outlined in this paper, which offer new explanations of vaccine hesitancy amongst marginalized groups. Further, we offer theoretical contributions that highlight the fact that social location, alongside epistemic misrepresentation and embodied experience are important in the construction of trust and distrust. As such, this research suggests that vaccine policy should address marginalized groups’ ongoing, alienating interactions with experts in combination with broader, sustained efforts to dismantle structural racism and xenophobia in biomedicine.

Footnotes

1

As Yaqub et al. explain, people may be “hesitant” long before they delay or stop vaccinating, so “[f]ocusing only on vaccine uptake rates and neglecting underlying attitudes is likely to underestimate the challenge of maintaining vaccination coverage in the future” (2014:6).

2

Autism spectrum disorder (ASD) is diagnosed based on developmental delays in three areas: language, social skills and behavioral repertoire, often presenting as repetitive play, object fixation and self-stimulation. https://www.cdc.gov/ncbddd/autism/hcp-dsm.html.

3

The notion of the social contract is that individuals must share collective responsibility for the prevention of disease (Lakoff 2015).

4

Parental decision-making about vaccinating children is often distinct from decisions to vaccinate themselves. Here we focus on parental decision-making around childhood vaccination.

5

Herd immunity is a condition in which a high proportion of the population obtains immunity to a disease, which in turn protects those who cannot be immunized.

6

One research team has explored the reasons behind Black Americans’ resistance to the flu vaccine but their work focuses on adult vaccinations (Freimuth et al. 2017; Jamison et al. 2019; Quinn et al. 2017).

7

Social capital is typically defined as access to resources through social networks.

8

Eyal (2019) and Brownlie and Howson (2005) link literature on trust to vaccine hesitancy, but do not focus on racial differences. One research group links vaccine hesitancy among Black communities to medical distrust (Freimuth et al. 2017; Quinn et al. 2017).

9

Tuskegee refers to a U.S. Public Health Service experiment (1932-1972), wherein 600 Black men were recruited to study the long-term effects of untreated syphilis.

Author notes

This project was funded by the National Science Foundation (1423025) and the Institute for Research on Race and Public Policy at the University of Illinois at Chicago. We would like to thank all of the Somali parents who opened up their lives to us and shared their vaccine concerns and their stories of parenting children with autism. We would also like to thank Andy Clarno and Jonah Stuart Brundage for comments on earlier drafts of the article. Three reviewers and the editors also made helpful comments that improved our analysis. Please direct correspondence to the first author at the Behavioral Sciences Building, 4150B, 1007 West Harrison St. (MC 312), Chicago, IL 60607-7140, USA; email: [email protected].