Experiences of Latinx Individuals Hospitalized for COVID-19: A Qualitative Study

Key Points

Question  Can experiences of Latinx adults hospitalized with coronavirus disease 2019 (COVID-19) inform improvements to public health and health care?

Findings  In this qualitative study of 60 Latinx adults, participants reported COVID-19 misinformation, felt COVID-19 compounded existing social disadvantage, and risked infection because of the need to work. Participants hesitated to seek hospital care because of immigration and economic concerns.

Meaning  These findings suggest that to contain community spread and reduce unnecessary morbidity, immigration, employment, and economic distress must be addressed through tailored public health messaging and public policy interventions that improve economic conditions.

Importance  Latinx individuals, particularly immigrants, are at higher risk than non-Latinx White individuals of contracting and dying from coronavirus disease 2019 (COVID-19). Little is known about Latinx experiences with COVID-19 infection and treatment.

Objective  To describe the experiences of Latinx individuals who were hospitalized with and survived COVID-19.

Design, Setting, and Participants  The qualitative study used semistructured phone interviews of 60 Latinx adults who survived a COVID-19 hospitalization in public hospitals in San Francisco, California, and Denver, Colorado, from March 2020 to July 2020. Transcripts were analyzed using qualitative thematic analysis. Data analysis was conducted from May 2020 to September 2020.

Main Outcomes and Measures  Themes and subthemes that reflected patient experiences.

Results  Sixty people (24 women and 36 men; mean [SD] age, 48 [12] years) participated. All lived in low-income areas, 47 participants (78%) had more than 4 people in the home, and most (44 participants [73%]) were essential workers. Four participants (9%) could work from home, 12 (20%) had paid sick leave, and 21 (35%) lost their job because of COVID-19. We identified 5 themes (and subthemes) with public health and clinical care implications: COVID-19 was a distant and secondary threat (invincibility, misinformation and disbelief, ingrained social norms); COVID-19 was a compounder of disadvantage (fear of unemployment and eviction, lack of safeguards for undocumented immigrants, inability to protect self from COVID-19, and high-density housing); reluctance to seek medical care (worry about health care costs, concerned about ability to access care if uninsured or undocumented, undocumented immigrants fear deportation); health care system interactions (social isolation and change in hospital procedures, appreciation for clinicians and language access, and discharge with insufficient resources or clinical information); and faith and community resiliency (spirituality, Latinx COVID-19 advocates).

Conclusions and Relevance  In interviews, Latinx patients with COVID-19 who survived hospitalization described initial disease misinformation and economic and immigration fears as having driven exposure and delays in presentation. To confront COVID-19 as a compounder of social disadvantage, public health authorities should mitigate COVID-19–related misinformation, immigration fears, and challenges to health care access, as well as create policies that provide work protection and address economic disadvantages.

Coronavirus disease 2019 (COVID-19) has magnified preexisting health and social inequities stemming from long-standing poverty, structural racism, and immigration status. As a result, certain racial and ethnic groups, including Latinx individuals (who are a part of the largest ethnic minority in the US at 60 million), are overrepresented among COVID-19 infections.^1-7 Current data show that compared with White individuals, Latinx individuals are more likely to become infected, hospitalized, and die from COVID-19.^8-14 In addition, Latinx individuals have had some of the highest rates of excess mortality compared with other racial and ethnic groups and have not been shown to benefit from shelter-in-place policies.¹⁴

Multiple factors drive excess COVID-19 risk among Latinx individuals.¹⁵ Much of the excess risk in Latinx communities is concentrated among immigrants.¹⁶ Compared with other racial and ethnic groups, Latinx immigrants are more likely to work in low-wage, service industries and be uninsured.^17-22 A recent analysis of individuals with COVID-19 found that compared with non-Latinx groups, Latinx individuals were more like to report working while ill, exposure to someone with COVID-19 in the household, and having more persons in the household.¹³ These circumstances have contributed to an excess burden of COVID-19 morbidity and mortality in the Latinx community.^13,14

In this study we describe Latinx patient perspectives on COVID-19 before, during, and after hospitalization in 2 cities where Latinx people are disproportionately affected. Learning from the experiences of Latinx people who had been hospitalized with COVID-19 can inform local and national interventions to reduce avoidable COVID-19 infections and decrease COVID-19 morbidity and mortality in the Latinx community.

We conducted semistructured interviews with Latinx adults (age ≥18 years) who had been hospitalized for COVID-19 in public hospitals located in Denver, Colorado, and San Francisco, California, between March and July 2020. Both cities reported excess COVID-19 cases among Latinx individuals compared with their numbers in the overall population (54% vs 32% in Denver and 51% vs 15% in San Francisco).^23,24 We identified participants via a data query that provided the contact information for individuals who self-identified as Latinx and had been hospitalized for COVID-19, and had an interviewer call them. Purposive sampling captured a diverse sample in terms of gender and city. Participants provided informed consent that included permission to publish deidentified quotations. The institutional review boards of the University of Colorado, Denver, and the University of California, San Francisco, approved this study. We report our study using the Consolidated Criteria for Reporting Qualitative Research (COREQ) reporting guideline.

Six authors conducted semistructured interviews in English or Spanish by telephone. The interview guide (eTable 1 in the Supplement) was based on a literature review of race disparities and the COVID-19 pandemic, with a particular focus on Latinx communities.^1,2,21,25-28 Interviews were audio recorded and transcribed verbatim, and were conducted until thematic data saturation, defined as when a thorough understanding of the perspectives of Latinx respondents were obtained with few or no new concepts emerging in subsequent data collection.^29,30

Interview transcripts were imported into HyperRESEARCH³¹ version 4.0.1 (ResearchWare Inc). Five authors (L.C., M.M., M.F., J.F.F., and A.F.) identified initial themes. Coding and analysis was performed according to the principles of grounded theory and thematic analysis.^29,30 Authors A.T. and A.M.G. performed line-by-line coding to inductively identify initial concepts, then grouped similar concepts into themes and subthemes, and identified conceptual links among themes. They reached consensus on themes with authors L.C., M.M., M.F., and A.F. Investigator triangulation ensured that the themes reflected the full range and depth of the data.

Sixty individuals (36 [60%] men; mean [SD] age, 48 [12] years) participated. All lived in low-income areas, 47 participants (78%) had more than 4 people in their home, and 44 participants (73%) were classified as essential workers.^32,33 Four participants (9%) could work from home, 12 (20%) received sick leave, and 21 (35%) lost their job due to COVID-19. Fifty-four interviews (90%) were conducted in Spanish. Participants were hospitalized for a mean (SD) 8 (10) days and 17 participants (28%) required treatment in the intensive care unit (Table 1; eTable 2 in the Supplement). Of those called to participate, 77 (78%) agreed. The most common reasons for declining were fear of sharing personal information, lack of time, and fatigue. Mean (SD) interview duration was 42 (12) minutes.

We identified 5 themes and subthemes with public health and clinical care implications, which are provided in Table 2 with illustrative quotations. Conceptual links are shown in the eFigure in the Supplement.

Some participants felt they would not contract COVID-19, and if infected they would not become ill (“I initially ignored it and did not use a mask because I thought I wouldn’t get sick.”). Many felt the media was “overexaggerating,” and dismissed preventive recommendations. One participant highlighted the indifference among his coworkers, saying, “We’d make fun of COVID. If one of us coughed, we’d say, ‘you have COVID,’ and laugh.” They regarded some measures, including workplaces closing, to be too extreme: “When they told us to stop working at the restaurant I asked, ‘Why, if we’re fine?”

Many participants relied on social media for COVID-19 recommendations and described a lack of information and misinformation. Some recounted that before they became ill, they thought COVID-19 “was a bunch of lies.” Others were suspicious of the government. “I thought the government invented COVID,” said a participant. “It wasn’t until I got sick that my family and I believed it.” A few believed that COVID-19 was aimed at identifying undocumented individuals (“There is lack of information and understanding about COVID. Some of us see it as a tactic for the government to access our documentation status and deport us.”).

Some participants found it difficult to physically distance because of cultural norms. “As a community, we demonstrate affection,” said a participant. “When we see someone we know, we give a firm handshake, a strong hug, and some even get a kiss.” Participants described gatherings during shelter in place: “Latinos get together for graduations and birthdays and that’s where we get sick. …I understand now because I ignored it and that’s how I got sick.”

Participants were terrified they would become unemployed because of business closures or a COVID-19 diagnosis. “I wasn’t sure if we would all be fired [because of shelter in place],” said a participant. “After getting sick, my main concern still was being fired.” Some participants supported relatives abroad and worried about their family’s well-being (“How will I support my family in Mexico if I can’t send money?”). Even with COVID-19 symptoms, individuals felt a need to work (“I worked on days that I was feeling terrible, because I had no money and I knew more difficult times were coming.”). Participants were also concerned over losing their home (“I worried that once discharged from the hospital, I wouldn’t have a home, and the landlord would tell me that I couldn’t live there anymore because I couldn’t pay the rent.”).

Undocumented participants reported anxiety because of their lack of resources because of their immigration status. “I’ve been worried about unemployment, because in this country, if you don’t have rent money, you are thrown into the streets,” said a participant. “You don’t have food either … as an undocumented person, there are no benefits.” Undocumented immigrants described feeling dehumanized and unprotected (“We are just surviving … in this country, I have nowhere I can go for help.”).

Because of financial constraints, many participants continued working despite shelter in place. “Regardless of the risk, we have to work to make money,” said a participant. “Other people can work from home but our jobs don’t allow us to.” Those who continued working found it difficult to protect themselves because of their occupation. One participant said, “We couldn’t protect ourselves because our work requires heavy lifting and we are breathing hard, and the mask doesn’t help.”

Many participants live in multigenerational housing and were anxious about infecting their families. “I was worried about my family,” said a participant. “I would arrive home from work and hope I was not sick. I didn’t want to spread it to my whole family.” Others lived in small, crowded settings where they found it impossible to practice physical distancing (“The rent is expensive. We do not have the necessary space to safely isolate from others … there are 6 or 7 people living in each bedroom.”). They had to trust that others in the same home were taking precautions (“We all have to work. We don’t know who has COVID.”).

Many avoided seeking health care because of concerns about cost—“The cost was the main reason I didn’t want to go to the hospital. If I die, how will my family pay for it?” Participants waited until symptoms were advanced. “I couldn’t take the severe pain in my eyes,” said one. “I didn’t want to go to the hospital because I was afraid of the bills … I have no benefits.”

Uninsured participants were uncertain they would receive care even in the safety net. One participant said, “When I got sick, I feared going to the hospital because I was scared that they wouldn’t provide me care.” Some felt that because of being undocumented, they would be withheld care (“When I got sick, I worried about being an immigrant, and not receiving medical care.”).

Patients were afraid that if hospitalized, their immigration status would be assessed and reported. As one participant said, “If I don't have my legal documents, they can report that information.” Individuals feared being deported if they were hospitalized (“When they realize that you do not have papers after arriving at the hospital, they can deport you.”).

While hospitalized, participants felt loneliness and wanted more direct contact with their clinicians. “I didn’t understand at first,” said one participant. “They would come in but not touch me … they were coming in with lots of precautions … it hurt my feelings … its dehumanizing.” Individuals described being treated differently because of COVID-19. “They wouldn’t take me to the bathroom when I needed to go,” said a participant. “They wouldn’t give me physical therapy even though I needed it.” Participants with previous hospitalizations acknowledged that COVID-19 hospital care was different (“They never cleaned my room during the weeks that I was there, they never took out the trash. This time was different.”). Restrictive visitor policies led to loneliness for those that had family in the area. Others expressed anxiety about being sick in a country far from family (“What am I going to do if I get very sick because my family is in a different country? I’m scared.”).

Many participants reported relief and gratitude for clinicians that motivated them to cope with COVID-19. “The nurses and doctors, they encouraged me,” said a participant. “They would say, ‘Have hope because you will get over this.’” Participants also described readily available language services (“The doctors were always using interpreters and thank God that many of the nurses spoke Spanish.”). Many expressed preferences for Spanish-speaking clinicians (“I think it is important that nurses speak Spanish because we spend a lot of time with them.”). And participants expressed gratitude for their care (“They took excellent care of me…I am so grateful.”).

On discharge, some participants described leaving without follow up treatment and care, including oxygen and physical therapy, because they lacked health benefits. “I had low oxygen and when they took it off, it went below low,” said a participant. “They could not offer oxygen at time of discharge because I couldn’t pay for it.” After discharge, many worried about how long they were supposed to continue isolation, and wondered if they could become re-infected (“Once I was extubated, they didn’t tell me that I had already passed the [isolation time] and that I wouldn’t be contagious anymore.”).

Some participants described relying on their faith and prayed that they would not die (“I would say, my dear God, take this illness away from me.”). Many prayed for their families because they did not want them to be sick. They also prayed that if they died, their families would be okay (“Dear God, if I die, what will happen to my child? It’s God’s will as he watches us.”).

Worried that COVID-19 would continue spreading in the Latinx community, participants urged their friends and family to protect themselves (“I tell all of my family that they need to understand COVID-19 because it is awful.”). Many felt that the Latinx community was more likely to believe that COVID-19 is real if they received information from a Latinx community member. Said one participant, “It’s important to tell people that you were sick because then they can believe … for example, out of every 100, there may be 10 that know someone with COVID-19, and they can say, ‘Yes, this virus is real because I know someone who is sick,’ … Sometimes the television commercials show people from a different race or language and we think, ‘Oh, that’s not real.’ But if we see people that we know with COVID, then we will believe what is happening.”

This report on the experiences of Latinx adults who survived COVID-19 before, during, and after hospitalization identified multiple themes with public health implications. Common themes included the prevalence of COVID-19 misinformation, COVID-19 as compounding socioeconomic disadvantage, and a reluctance to seek medical care. We also identified themes with implications for health care systems including experiences of social isolation during hospitalization, difficulty with discharge planning, and the role of faith and community in recovery. These findings have implications for COVID-19 control both within and outside the Latinx community, because Latinx individuals have the highest rates of employment in jobs that require essential workers, in grocery stores and restaurants, general merchandise stores, pharmacies, and other businesses where they come into contact with the community at large. Our findings suggest a number of avenues to improve prevention and treatment of COVID-19 (Table 3).

A key finding across multiple themes was economic anxiety. The fear of losing wages or becoming unemployed because of COVID-19, coupled with poverty and risk of being evicted, pushed some individuals to work even when symptomatic, thereby contributing to viral spread. Containing COVID-19 infections means mitigating the economic pressures forcing Latinx individuals to work even when infected, regardless of documentation status. One intervention supporting Latinx individuals to isolate when infected, regardless of documentation status, is Right to Recover, a program launched in San Francisco in July 2020 that provided $1285 to workers needing to isolate because of a COVID-19 infection.³⁷ The effect of this and similar policies needs evaluation. Individuals also expressed a fear of eviction, despite a moratorium on evictions. Programs that provide rent relief may offer households the economic ability to isolate without fear of unpaid rent.^38,39

Participants described a reluctance to seek COVID-19 testing and hospital care because of concerns over cost, access to care, and immigration repercussions. A recent analysis among individuals hospitalized with COVID-19 in Colorado found that the median onset between symptom onset and hospitalization was 4 days among Latinx individuals vs 3 days among non-Latinx individuals.¹³ These findings are striking, as Denver and San Francisco both have public hospitals and health care programs that extend care to undocumented individuals at a sliding scale based on income. Both cities also have longstanding sanctuary policies prohibiting the health care system from sharing information with immigration authorities. However, despite these policies, immigration related concerns, including concerns for public charge, remain rampant.^35,36,40-44

Our study found that Latinx individuals need more and more effective public health messaging to decrease testing fears, improve contact tracing, and encourage symptomatic individuals to seek medical care. Many individuals who recovered expressed the desire to serve as COVID-19 ambassadors within their community. Former patients may be excellent messengers to communicate COVID-19 information and reduce COVID-19 spread. Mobilizing Latinx individuals and community health workers to disseminate culturally specific and language-concordant COVID-19 information may be a powerful strategy to reduce COVID-19 infections (Table 3).^20,45 Previously hospitalized individuals might also help disseminate COVID-19 vaccine information.^46,47

The experience of isolation described by participants is common across hospitalized patients with COVID-19.^48-50 However, it may be further exacerbated for Latinx individuals with limited English proficiency and for those with family abroad. To avoid mistrust and confusion, COVID-19–related changes to hospital procedures should be disclosed to patients at admission. For example, Zuckerberg San Francisco General developed an admission Spanish COVID-19 video. It addresses COVID-19 hospital policies and common COVID-19 care practices, such as proning. Technology is also key to linking patients with family members in the US and abroad. Additionally, because participants identified spirituality as important to their recovery, routinely offering spiritual support services to patients could reduce isolation.

Finally, cities and states need policies that improve access to outpatient services at discharge. Outpatient COVID-19 care remains uncovered in many states. This was the case early on in Colorado, and patients paid out of pocket for oxygen or medical care at discharge. Some states, including now Colorado, have included COVID-19 outpatient care as a qualifying condition for emergency Medicaid.^51-53

Our study has limitations. We recruited patients from 2 academic public hospitals in cities where the Latinx population had predominantly immigrated from Mexico and Central America. Transferability of the findings to other health systems or to Latinx immigrants from other regions is unknown. However, as both Denver and San Francisco have extensive public health care systems and a long history of sanctuary policies, the concerns reported by our participants may be even more widespread in other cities. Another limitation is that we only captured the experiences of patients who survived. However, we included individuals who received ICU level care, which captures the experience of those who were severely ill. This study’s strengths include the rich description from participants in 2 cities where Latinx people have faced high COVID-19 infection rates.

Latinx communities have suffered disproportionately from COVID-19. The pandemic has amplified preexisting inequities in health care and created new disparities in health, economic, and social well-being. Our findings underscore the urgent need for a public health response specific to the Latinx community that harnesses community strengths and offers tailored and specific messages. Results also highlight the need for economic and housing policies that support Latinx individuals’ ability to isolate and protect the safety of essential workers and also the broader community. For “Immigrants (We Get the Job Done)”⁵⁴ to continue to ring true, the US must enact policies to protect workers from COVID-19 and allow them and their household contacts the economic wherewithal to isolate and quarantine.

Accepted for Publication: January 13, 2021.

Published: March 11, 2021. doi:10.1001/jamanetworkopen.2021.0684

Open Access: This is an open access article distributed under the terms of the CC-BY License. © 2021 Cervantes L et al. JAMA Network Open.

Corresponding Author: Lilia Cervantes, MD, Denver Health, 777 Bannock, MC 4000, Denver, CO 80204 (lilia.cervantes@dhha.org).

Author Contributions: Dr Tong and Ms Gonzalez had full access to all of the data in the study and take responsibility for the integrity of the data and the accuracy of the data analysis. Drs Powe and Fernandez contributed equally to the study.

Concept and design: Cervantes, Martin, Tong, Powe, Fernandez.

Acquisition, analysis, or interpretation of data: All authors.

Drafting of the manuscript: Cervantes, Martin, Kearns, Camacho, Mundo, Powe, Fernandez.

Critical revision of the manuscript for important intellectual content: Cervantes, Frank, Farfan, Kearns, Rubio, Tong, Matus Gonzalez, Collings, Mundo, Powe.

Statistical analysis: Cervantes, Rubio, Collings.

Obtained funding: Fernandez.

Administrative, technical, or material support: Cervantes, Martin, Frank, Farfan, Kearns, Camacho, Mundo, Powe.

Supervision: Cervantes, Martin, Powe, Fernandez.

Conflict of Interest Disclosures: None reported.

Funding/Support: Dr Cervantes is funded by National Institute of Diabetes and Digestive and Kidney Diseases grant K23DK117018 and Robert Wood Johnson Foundation Clinical Scholars Program grant 77887. Dr Fernandez is funded by National Institute of Diabetes and Digestive and Kidney Diseases grant K24DK102057. Funding was provided by an internal grant made available from the University of California Office of the President.

Role of the Funder/Sponsor: The funders had no role in the design and conduct of the study; collection, management, analysis, and interpretation of the data; preparation, review, or approval of the manuscript; and decision to submit the manuscript for publication.