Impact statement
This study partnered with community members to study the impact of the COVID-19 pandemic on stress in Latinx immigrants in an area without significant social or medical support for Latinx immigrants. Our results confirmed that the pandemic and stress were experienced differently between subgroups of Latinx immigrants. Guatemalans were more likely to report experiencing stigma or discrimination and were less likely to report an increase in stress during the pandemic. Community conversation participants suggested that this was related to their high levels of pre-pandemic stress and emphasized the need for more resources serving local Guatemalan immigrants. Guatemalans also reported different uses of stress management techniques and were more likely to prefer learning about stress management through text messaging groups or YouTube videos. Previous work from our group showed that stress and stress management are priority health concerns within the Latinx immigrant community in our city, and that peer-led stress management interventions are effective and acceptable. Despite this, only about half of the surveyed participants wanted to learn more about stress management 9–14 months after the start of the COVID-19 pandemic, and most participants preferred to learn about stress management through watching YouTube videos and preferred professional-led groups over peer-led groups. These results are important for showing that when designing programs to serve Latinx immigrants, a one-size-fits-all approach is unlikely to be effective, particularly in areas with limited social and medical support for these communities. In addition, asynchronous methods of teaching about stress management are likely to be preferred.
Introduction
As of 2020, Latinx individuals are the second-largest racial/ethnic minority group in the U.S., only outnumbered by non-Hispanic White individuals (Funk and Lopez, Reference Funk and Lopez2022). Due to sociopolitical and economic factors (including changes to U.S. immigration law), there has been a consistent and rapid growth in the number of immigrants living in the U.S. since 1970 (Valentín-Cortés et al., Reference Valentín-Cortés, Benavides, Bryce, Rabinowitz, Rion, Lopez and Fleming2020). In 2018, about half of U.S. immigrants came from Latin America (Budiman et al., Reference Budiman, Tamir, Mora and Noe-Bustamante2020). States with the largest number of immigrants include California, Texas, New Jersey, New York, Florida and Nevada (Funk and Lopez, Reference Funk and Lopez2022).
Geographic areas outside of these high-volume states, called non-traditional immigration destinations (NTIDs), have experienced an exponential increase in the Latinx immigrant population only in recent years. Because of this, these locations often lack the infrastructure, Spanish language services, and community support to effectively meet the needs of Latinx newcomers (Jacquez et al., Reference Jacquez, Vaughn, Zhen-Duan and Graham2016). Latinx immigrants living in NTIDs are more likely to struggle with access to health services compared to peers in traditional immigration destinations (Gresenz et al., Reference Gresenz, Derose, Ruder and Escarce2012; Topmiller et al., Reference Topmiller, Zhen-Duan, Jacquez and Vaughn2017; Zhen-Duan et al., Reference Zhen-Duan, Jacquez and Vaughn2017; Esterline and Batalova, Reference Esterline and Batalova2022). Additionally, Latinx individuals living in NTIDs are at greater risk of social and economic disadvantages and exclusion (Crowley et al., Reference Crowley, Lichter and Turner2015; Jacquez et al., Reference Jacquez, Vaughn, Pelley and Topmiller2015, Reference Jacquez, Vaughn, Zhen-Duan and Graham2016).
Stress among Latinx individuals
Stress is a physiological or psychological response in which a person feels tense, restless, nervous, or unable to relax because their mind feels troubled. Latinx adults have higher stress levels than those reported by other ethnicities (American Psychological Association, 2016; Reference LR, Okoro, Strine, Verlenden, Hollis, Njai, Mitchell, Board, Puddy and ThomasMcKnight-Eily et al., 2021). Documented sources of stress among immigrants include ethnic discrimination, concerns over immigration status and family separation, and socioeconomic inequalities (American Psychological Association, 2016; Reference LR, Okoro, Strine, Verlenden, Hollis, Njai, Mitchell, Board, Puddy and ThomasMcKnight-Eily et al., 2021). Stress associated with racism and discrimination has been shown to negatively affect mental and physical health in Latinx individuals (Carvajal et al., Reference Carvajal, Rosales, Rubio-Goldsmith, Sabo, Ingram, McClelland, Redondo, Torres, Romero, O’Leary, Sanchez and de Zapien2013; Cobb et al., Reference Cobb, Salas-Wright, John, Schwartz, Vaughn, Martínez, Awad, Pinedo and Cano2021; Paradies et al., Reference Paradies, Ben, Denson, Elias, Priest, Pieterse, Gupta, Kelaher and Gee2015; Valentín-Cortés et al., Reference Valentín-Cortés, Benavides, Bryce, Rabinowitz, Rion, Lopez and Fleming2020). The health implications of high-stress levels are particularly concerning given that Latinx immigrants in the U.S. have poorer access to health services that could offer treatment (Ortega et al., Reference Ortega, Rodriguez and Vargas Bustamante2015; Ramos, Reference Ramos2022).
COVID-19 pandemic
Compounding the existing levels of stress, Latinx communities in the U.S. were disproportionately affected by the COVID-19 pandemic (Webb Hooper et al., Reference Webb Hooper, Nápoles and Pérez-Stable2020). Compared to non-Hispanic White individuals, Latinx individuals were more likely to be diagnosed with and die due to COVID-19, experience financial or job loss, and experience stress due to housing and food instability, especially in NTIDs (Webb Hooper et al., Reference Webb Hooper, Nápoles and Pérez-Stable2020; Collins Niesz, Reference Collins Niesz2021; Hibel et al., Reference Hibel, Boyer, Buhler-Wassmann and Shaw2021; Reference LR, Okoro, Strine, Verlenden, Hollis, Njai, Mitchell, Board, Puddy and ThomasMcKnight-Eily et al., 2021; Holden et al., Reference Holden, Simon, Arnold, Halloway and Gerardin2022; Martin et al., Reference Martin, Castano, Geraghty, Horner, McCann, Beck, Xu, Gomez, O’Dea, Jacquez, Clark and Rule2022; Bovell-Ammon et al., Reference Bovell-Ammon, Ettinger de Cuba, Lê-Scherban, Rateau, Heeren, Cantave, Green, Frank, Cutts, Ochoa and Sandel2023). During the pandemic, many people in the U.S. required assistance with basic needs (e.g., healthcare, food), but anti-immigrant legislation such as the “public charge rule” (which stipulated that use of public services by U.S. immigrants could revoke green card eligibility) made immigrants hesitant to access available services (Miller et al., Reference Miller, Ripepi, Ernstes and Peguero2020; Page et al., Reference Page, Venkataramani, Beyrer and Polk2020; Wang et al., Reference Wang, Glied, Babcock and Chaudry2022). Unsurprisingly, mounting evidence suggests these structural and systemic stressors, which were exacerbated during COVID-19, negatively impacted the mental health of Latinx individuals (Swaziek and Wozniak, Reference Swaziek and Wozniak2020; Garcini et al., Reference Garcini, Rosenfeld, Kneese, Bondurant and Kanzler2021; Gomez-Aguinaga et al., Reference Gomez-Aguinaga, Dominguez and Manzano2021; Hibel et al., Reference Hibel, Boyer, Buhler-Wassmann and Shaw2021; Serafini et al., Reference Serafini, Powell, Frere, Saali, Krystal, Kumar, Yashaswini, Hernandez, Moody, Aronson, Meah and Katz2021).
Current setting
Cincinnati, Ohio, the location for this project, is an NTID. Since 2010, the Latinx population in Cincinnati has grown 48.5%, with Central American immigrants outnumbering other Latinx groups (George Mason University Institute for Immigration Research, 2018; Esterline and Batalova, Reference Esterline and Batalova2022). In fact, the yearly growth rate of the Latinx population is nearly double the national rate (4.0 vs. 2.2%, respectively) (Hispanic Chamber Cincinnati, 2015). A higher proportion of Latinx individuals in Cincinnati are immigrants than in other similar metro areas due to this recent influx (Hispanic Chamber Cincinnati, 2015). In Cincinnati, the median household income for Latinx households was 85% that of all households in 2020, and the unemployment rate is higher than the overall unemployment rate (Economics Center Research and Consulting, 2022). Based on available data, in 2020, 91.5% of the total Cincinnati population had at least a high school diploma, while 76.8% of the local Latinx population did. The three industries within which the Latinx population is most likely to be employed are manufacturing (14.0%) and administrative, support, waste management, and remediation services (14.0%), and accommodation and food services (11.1%) (Economics Center Research and Consulting, 2022).
Within this context, we aimed to build on prior work that showed that stress is a health priority for the Cincinnati Latinx community (Jacquez et al., Reference Jacquez, Vaughn and Suarez-Cano2019). Given that Latinx immigrants in NTIDs face barriers to accessing health services, we aimed for our work to contextualize stress during the COVID-19 pandemic in Latinx immigrants in an NTID and to understand Latinx immigrants’ preferences for stress management interventions given the sociopolitical and public health context.
Methods
Research design
Community-based participatory research (CBPR) is a research approach that involves shared decision-making between community members and academic partners and has the goal of generating meaningful knowledge that is responsive to the needs of the community (Israel et al., Reference Israel, Schulz, Parker and Becker1998; Wallerstein and Duran, Reference Wallerstein and Duran2006). Latinos Unidos por la Salud (LU-Salud) is a Cincinnati-based CBPR group focused on addressing Latinx health in an NTID (Vaughn et al., Reference Vaughn, Jacquez, Marschner and McLinden2016, Reference Vaughn, Jacquez, Zhen-Duan, Graham and Marschner2017; Jacquez et al., Reference Jacquez, Vaughn and Suarez-Cano2019). LU-Salud is a group of community members that was formed in 2013 with academic partners at the University of Cincinnati (UC) who were involved in community-engaged research to study the health priorities of the local Latinx population. Stress was identified as a community priority in the original survey, and in 2019 an opportunity for additional funding emerged to continue work LU-Salud had been doing with group stress management visits. Upon funding announcement in January 2020, the network of LU-Salud co-researchers was engaged over text asking which members were interested in paid work on this initiative, and two members who had been involved in the original and subsequent studies expressed interest. In line with Israel et al.’s CBPR model for maintaining and sustaining engagement, the two LU-Salud community co-researchers contributed to the study design and were involved with survey development, data collection, qualitative data analysis, and dissemination after receiving training on each of these by the academic team (Israel et al., Reference Israel, Schulz, Parker and Becker1998). This study applied CBPR principles within a mixed methods design, the phases of which are illustrated in Figure 1.
Study design.
Procedures
This study was reviewed by the Institutional Review Board of the University of Cincinnati (UC) and determined to be exempt from ongoing oversight given its low risk, lack of collection of identifying information, and goal of informing local initiatives. After survey development, community co-researchers recruited adult Latinx immigrants in Cincinnati between December 2020 and May 2021 via social media advertisement and their social networks and verbally administered the survey during this same period in Spanish (or Mam when necessary). Regular meetings between community co-researchers and academic partners occurred to solve difficulties with recruitment and COVID-19 safety protocols. Survey responses were recorded on printed interview guides and later entered into Research Electronic Data Capture (REDCap) by academic partners for data analysis. Manual double data entry was done on 15% of surveys for data verification (Barchard and Verenikina, Reference Barchard and Verenikina2013). Surveys were completed either via telephone conversation or in person. Participants were given $5 gift cards after survey completion.
Participants
Survey participants included 121 adults, the average age of whom was 36.62 (SD = 8.90; range = 18–61). Most participants were women (80.2%) and from Guatemala (53.7%). The average time participants had been living in Cincinnati was 12.79 years (SD = 6.38; range = 1–32). Additional sample demographics are shared in Table 1.
Survey participant demographics
Note. n = Number of participants. Total unique participants in survey n = 121.
Survey development
In a planning meeting in November 2020, the co-researchers shared that a verbally administered survey would be the best way to achieve our study aim, particularly in the context of public health restrictions. We had four virtual team sessions through the end of December 2020 to develop the survey and review survey administration techniques. The order, flow, wording, and length of Likert scales were co-created with the co-researchers. The instrument was developed in Spanish, and to increase trust, all questions were optional. A co-researcher who spoke Mam would clarify questions in Mam when needed during data collection, but because this is not a written language we did not have a written instrument in Mam. See supplemental materials for English and Spanish versions of the full instrument.
Survey measures
Impact of pandemic on daily life
Experiences during the pandemic were assessed using a modified version of the second and third questions of the Pandemic Stress Index (PSI) tool, a three-question scale that was translated into Spanish using translation, back translation, and comparison/reconciliation (Harkness et al., Reference Harkness, Behar-Zusman and Safren2020). This tool was selected from among the few available existing tools during the study design phase by our co-researchers for its relevance and brevity. Our co-researchers modified response options to match local verbiage and experiences (e.g., worry about children’s education). See Table 2 for the questions and response options.
Survey responses
Note. n = Number of participants. Total unique participants in survey n = 121.
Stressors and stress levels
To assess stress, participants were asked to self-describe their stress level using a five-point Likert scale (1 = Not at all to 5 = Very much). If the participant answered that they had experienced any stress, changes to their stress level were assessed with the follow-up question, “Was your stress level higher than before the pandemic?” (Yes/No).
Stress management
The co-researchers were interested in understanding the strategies participants used to manage stress. They developed a list of common strategies based on a previous LU-Salud study and the preferences of their social networks (see Table 3) (Jacquez et al., Reference Jacquez, Vaughn and Suarez-Cano2019). In the survey, participants were asked to select each technique they used during the pandemic from this list. They were also asked whether they were interested in learning more about stress and stress management. If they indicated interest, they were asked which options they would be interested in during and after the pandemic from a list generated by the co-researchers.
Stress management preferences
Note. n = Number of participants who chose at least one response for each survey question. Stress management preference questions were only asked to those who answered “yes” to wanting to learn more about stress management. X2 = Differences between those from countries besides Guatemala and those from Guatemala: *p <. 05 **p < .01, ***p < .001.
Quantitative analytic approach
Prior to conducting our main analyses, we screened the data for missing variables. Most (74.4%) participants gave complete demographic information. Analysis of missing data across all variables indicated that 1.27% of all items for all participants were missing, and 80% of the items were not missing data for any participant. Approximately 51% of participants had no missing data. According to Little’s MCAR Test, the data appeared to be missing at random, χ2(1,371, N = 121) = 1,302.20, p = .91. Given the random nature of missing data and minimal levels of missingness, pairwise deletion was applied to all analyses, except for the mediation analysis where the full information maximum likelihood method was used (Kang, Reference Kang2013).
To complete our main analyses, Pearson chi-square tests were used to assess the differences in COVID-19’s impact on daily life (none/very mild/moderate or a lot/extreme) by the demographic variables: sex, age, partnership status (single/widowed/divorced or married/partnered), children at home (none or one or more), country of birth (Guatemala or All other countries), household language ([English and Spanish and/or Mam] or [Spanish and/or Mam]), and employment in the last 6 months (yes or no). Pearson chi-square tests were also used to analyze differences in stress levels (none/mild/moderate or a lot/extreme) and stress change (higher than pre-pandemic or not higher than pre-pandemic) by all demographic variables listed above.
Given the demographics of our sample and impression among our LU-Salud co-researchers that the experiences of Guatemalan-born Latinx immigrants, many of whom have indigenous ancestry, are different, we examined participant differences by country of origin (Guatemala vs. all other countries). Prior LU-Salud work has identified differences in experiences between Cincinnati-based Mexican and Guatemalan immigrants (Zhen-Duan et al., Reference Zhen-Duan, Jacquez and Vaughn2017; DiMascio et al., Reference DiMascio, Zhen-Duan, Rabin, Vaughn and Jacquez2020). Our sample size limited our ability to detect significant demographic differences among our Guatemala-born participants compared to other participants, but we nevertheless knew it was important to honor this request and examine differences in pandemic experiences and stress management preferences among Guatemala compared to the rest of the group. We used Pearson chi-square tests to examine significant differences in participant experiences during the pandemic, stress management techniques used during the pandemic, interest in learning stress management techniques, and preferences for techniques during and after the pandemic by country of origin.
A one-way ANOVA with Tukey’s post hoc analysis was used to examine differences in COVID-19’s impact on daily life by time spent in Cincinnati (0–5 years vs. 6–10 years vs. 11–20 years vs. 20+ years). We also completed a post hoc mediational analysis to examine the mediating effects of household language on post-pandemic stress change by country of origin. We used JASP Structural Equation Modeling and bootstrapping techniques (1,000 Bias-corrected replications) to estimate the direct and indirect effects of these associations. All statistical analyses were completed in IBM SPSS 28 and JASP. The p-value was set to p < .05 for all analyses.
Community conversation procedures and participants
Due to complications from the COVID-19 pandemic, the community conversation (CC) initially needed to be conducted virtually, but most survey participants were uninterested in or unable to do a virtual focus group. Thus, to conduct the CC, we collaborated with a local church whose congregation agreed to host the research team in the spring of 2022 when it was more safe to convene. Twenty-five Latinx adult congregation members participated in the CC and received a $25 gift card. Demographic information of CC participants was not gathered, but the research team observed that, compared to survey participants, the CC had more participants over the age of 50 and more men. No congregation members had participated in the survey. CC participants were served refreshments, shown a presentation of the survey key results and participant demographics, and asked about their impressions and interpretations of the results. The conversation was audio-recorded and transcribed in Spanish using Microsoft Word, with an academic team member checking the transcription for accuracy and correcting it when necessary.
Qualitative analytic approach
The Spanish transcript of the CC was analyzed using thematic analysis (Braun and Clarke, Reference Braun and Clarke2006). Two Spanish-English bilingual academic team members (one of whom is a Latinx second-generation immigrant) familiarized themselves with the data and generated initial codes, which were organized into a codebook. Next, those two team members and one of the LU-Salud co-researchers used the codebook to code the entire transcript. The principal investigator reconciled disparate codes through discussion. Codes were organized into themes, and representative quotes for each theme were translated into English for dissemination.
Results
Impact of COVID-19
While 31.4% of participants reported that COVID-19’s impact on their lives was “a little”, 40.7% of participants reported the impact was “a lot” or “extreme.” The majority (91.7%) of participants reported “worrying about friends, family, partners, etc.” “More anxiety” was experienced by more than half of participants (50.4%), and “stigma or discrimination from other people” was also common (46.3%). See Table 2 for additional descriptive statistics on the impact of COVID-19.
Time in Cincinnati was significantly associated with COVID-19 impact (F(3,108) = 3.77, p = .013). A Tukey’s post-hoc analysis revealed that those who spent 6–10 years in Cincinnati and those who spent 11–20 years in Cincinnati significantly varied from one another, with those living in Cincinnati for 6–10 years reporting higher impact of COVID-19 on daily life than those living in the city 11–20 years (.29622, 95% CI [.0009,.591], p = .049). No other demographic variables were associated with COVID-19 impact.
Those who identified as Guatemalans were more likely to experience stigma during the pandemic (X 2 [1, N = 119] = 8.64, p < .01) than those from all other countries. Contrastingly, those who identified as being from other countries were more likely to experience sleep changes (X 2 [1, N = 119] = 16.71, p < .001), financial loss (X 2 [1, N = 119] = 17.67, p < .001), and financial support (X 2 [1, N = 119] = 6.28, p = .01) than Guatemalans.
Stress levels during the pandemic
As seen in Table 2, many participants reported feeling “a little” (50.0%) stress during the pandemic, and nearly a quarter (24.2%) reported “a lot” or “extreme” stress. Of those who reported stress and compared their stress levels to before the pandemic, more than half (n = 52 of 84, 61.9%) reported that their stress level was higher than before the pandemic.
No demographic group (by sex, age, partnership status, parenting status, country of origin, time in Cincinnati, or household language[s]) was more likely to report “a lot” or “extreme” levels of stress than others. However, Guatemalans were less likely to report increases in stress levels during the pandemic than participants from other countries (X 2 [1, N = 83] = 12.64, p < .001). Those who did not speak English at home were also less likely to report increases in stress levels during the pandemic than those who did (X 2 [1, N = 81] = 5.53, p = .02). No other demographic variables were associated with changes in stress level during the pandemic.
To assess a post hoc hypothesis that country of origin mediates the association between household language and post-pandemic stress change, the direct and indirect effects were estimated with JASP Structural Equation Modeling and bootstrapping techniques (1,000 Bias-corrected replications). As shown in Figure 2, in the mediated model, the indirect effect of language spoken at home on COVID-19 stress through origin country was significant, β = −.19 (SE = .09; 95% CI = −.42 to −.04; p = .04). The direct effect of language spoken at home on COVID-19 stress was not significant, β = −.42 (SE = .24; 95% CI = −.83 to.07; p = .08), suggesting full mediation.
Mediation model of household language on COVID-19 stress through country of origin.Note: p < .05*; p < .01**. Standardized regression coefficients are displayed. The total effect of language spoken at home on COVID-19 stress is displayed above the arrow; the direct effect of this relationship is displayed below the arrow.
Stress management preferences
The most common responses for stress management techniques utilized during the pandemic were “cleaning or organizing the house” (84.5%), “calling close friends” (68.0%), “cooking or following recipes” (67.0%), and “listening to music” (67.0%). The least common responses were “talking to a psychologist” (1.9%) and “deep breathing exercises or meditating” (13.6%).
A series of chi-square analyses (X 2 [1, N = 119]) showed that individuals from countries besides Guatemala were more likely than Guatemalans to use the following stress management strategies: deep breathing, praying, talking to a friend, singing, dancing, exercising, watching a movie, laughing, reading, spending time with family, spending time with pets and looking for information on their phones related to fitness and/or entertainment. See Table 3 for chi-square and p-values.
Chi-square analyses (X 2 [1, N = 119]) further revealed no difference in interest in learning more about stress/stress management by country of origin. Among participants who were interested in learning more, Guatemalans were less likely to prefer learning through in-person groups led by professionals and more likely to prefer learning through text messaging groups and YouTube videos during the pandemic. After the pandemic, Guatemalans were less likely to prefer learning through in-person groups led by peers and in-person groups led by professionals, while they were more likely to prefer learning through text messaging groups and YouTube videos. See Table 3 for chi-square and p-values.
Community conversation interpretation of survey results
Qualitative data from the CC revealed that the experiences and stress levels of Latinx immigrants in Cincinnati were different based on demographics (see Table 4). Overall, many CC participants were surprised that stress levels reported by survey participants were not higher, given their personal experiences. They noted that some groups within the community were disproportionately impacted by the pandemic (e.g., older individuals with chronic medical conditions, school-age children and their parents). They also described grief over social distancing restrictions due to the cultural importance of community, family, and worshiping together. Despite the importance of community, CC participants were not entirely surprised that private ways of learning about stress management were preferred but also acknowledged that the resumption of peer-led women’s support groups was helping reduce stress.
Community conversation themes and representative quotes
CC participants suggested that survey participants did not describe overall higher stress levels because they had been reflecting on services and resources available during the pandemic, and because many Latinx immigrants having jobs considered essential was protective against job loss. They also noted that some Central Americans, particularly Guatemalans, had experienced high levels of suffering and stress prior to coming to the U.S. and within the U.S. prior to the pandemic, and therefore may not have noticed a dramatic increase from an already-high stress level. One of our co-researchers emphasized that the local Guatemalan immigrant community was deeply affected by the pandemic and cautioned that survey results showing no increase in stress levels do not suggest that service providers do not need to serve or worry about the impact of stress on Guatemalans.
Discussion
The COVID-19 pandemic has disproportionately affected the Latinx community in the U.S., and our results show that in an NTID, the pandemic impacted subgroups of the Latinx immigrant community differently. In our sample, those who did not speak English at home were less likely to report increases in their stress levels during the pandemic. Across nationalities of our sample, Guatemalans were more likely to report experiencing stigma yet less likely to report increases in their stress level since pre-pandemic. They were also less likely to report using many coping strategies commonly used by other local Latinx immigrants. Guatemalans in our sample may have had higher baseline levels of stress prior to the pandemic and experienced or described stress differently than other Latinx groups, particularly because 28.9% of survey participants reported speaking Mam (an indigenous language in Guatemala) at home. Our CC participants referred to the “suffering” of Guatemalans before migration, and while much of Central America experienced conflict, civil war, and human rights abuses in the 1960s–1990s, the war in Guatemala was the longest and had the most casualties, with indigenous Guatemalans being disproportionately affected. Today, about 42% of Guatemalans are indigenous (the highest in Latin America), are 1.7 times more likely to be living in poverty than non-indigenous Guatemalans, and also more likely to be living in poverty than other indigenous groups in Latin America (World Bank Group, 2016).
A telephone study completed during the pandemic in Guatemala showed similar levels of stress as our results (greater than 60% of their sample denied stress, while 62.5% of our participants reported no or “a little” stress. Our CC participants expressed surprise that levels of stress were not higher but acknowledged that Guatemalans may have different historical experiences and baseline stress levels compared to other Latinx immigrants. The finding that a sense of privacy around struggling was valued could additionally contribute to the lower-than-expected levels of reported stress by survey participants. Available research shows that Guatemalans exposed to war have increased rates of mental disorders, particularly when social networks are disrupted (Herrera et al., Reference Herrera and de Jesús2005; Puac-Polanco et al., Reference Puac-Polanco, Lopez-Soto, Kohn, Xie, Richmond and Branas2015; Kowal et al., Reference Kowal, Coll-Martín, Ikizer, Rasmussen, Eichel, Studzińska, Koszałkowska, Karwowski, Najmussaqib, Pankowski, Lieberoth and Ahmed2020).
More survey participants reported increased anxiety (50.4%) than moderate or higher levels of stress (37.2%). Similarly, in the Guatemalan phone survey, more people reported anxiety during the pandemic (46%) than stress (36%). Therefore, it is possible that culturally specific perceptions of “anxiety” and “stress” may have affected self-reported levels of stress in our survey. Prior research by our study team illustrates that local Guatemalan immigrants were less likely than local Mexican immigrants to report stress as a priority health concern, which may suggest different perceptions (DiMascio et al., Reference DiMascio, Zhen-Duan, Rabin, Vaughn and Jacquez2020).
Across demographic characteristics, we found that no group was more likely to report “a lot” or “extreme” levels of stress than others. These findings contrasted with prior literature, which highlights the unique susceptibility of women, those who are unmarried, and those who have children to stress and mental health outcomes during the pandemic (Kowal et al., Reference Kowal, Coll-Martín, Ikizer, Rasmussen, Eichel, Studzińska, Koszałkowska, Karwowski, Najmussaqib, Pankowski, Lieberoth and Ahmed2020; Gomez-Aguinaga et al., Reference Gomez-Aguinaga, Dominguez and Manzano2021). However, our findings may reflect limitations in our sample size and diversity, which could compromise statistical power to detect within-group differences. We are not able to compare stress levels in this group to non-Latinx Cincinnatians with our available data, but recognize that structural inequities and racism contribute to the socioeconomic disparities faced by Latinx Cincinnatians. With a larger sample size, we might be able to better explore ethnic and sociodemographic factors within the Cincinnati Latinx community that contribute to stressors and stress.
Impact of the pandemic
Most (56.8%) participants reported that the impact of the pandemic on their daily lives was “a little” or “moderate.” This is in contrast with data showing that in other NTID states with less well-established Latinx communities such as Oregon, Washington and Utah, there were higher rates of cases and less support systems in place than in states with more well-established Latinx communities such as California, Arizona or New Mexico (Jordan and Oppel, Reference Jordan and Oppel2020). The most common pandemic-related experiences reported by our participants were worrying about others, which has also been a theme in qualitative studies and may be higher than self-concern in Latinx immigrant groups (Moyce et al., Reference Moyce, Velazquez, Claudio, Thompson, Metcalf, Aghbashian, Vanderwood and Sisson2021; Quandt et al., Reference Quandt, LaMonto, Mora, Talton, Laurienti and Arcury2021). Reports of “stigma or discrimination from other people” were also common in our sample (46.3%) as has been described in other studies during the pandemic, and may reflect racism-related stress in this NTID (Hearst et al., Reference Hearst, Ekwonye, Munala, Ismail, Kennedy and Buesseler2021). Substance use was least likely to be described, in contrast with other reports that show that the pandemic may have contributed to immigrants’ stress and substance use (Romano and Sánchez, Reference Romano and Sánchez2023). As pointed out by a CC participant, it is possible that the reported level of impact on the lives of Latinx immigrants in Cincinnati is blunted because many immigrants in our NTID have low-income and essential jobs; this is supported by local labor force data showing that the industries in which most Latinx Cincinnatians work are manufacturing, administrative, support, waste management, accommodation and food services (Economics Center Research and Consulting, 2022). Lastly, given that a central theme was keeping struggles private and the stigma reported by our participants, it is possible that survey participants were endorsing the impact of the pandemic on their experience, particularly given the surprise expressed by CC participants at survey results. The result that non-Guatemalans were more likely to experience both financial loss and support may suggest the presence of community support, which was also a key qualitative theme but should be explored more to understand the experiences of those from countries other than Guatemala that confer socioeconomic disadvantage.
Stress management preferences
The current study showed that asynchronous, flexible methods of learning about stress management (watching YouTube videos or being a part of group texts) were preferred and that once public health restrictions were lifted, professional-led groups were preferred over peer-led groups. Contrary to this, one of our CC participants expressed surprise at these results given the importance of a peer-based women’s group that her church organized. Other research has shown that faith-based health promotion programs are a promising stress management approach for the Latinx community, and pre-pandemic research shows that peers or “promotoras”/“compañeras” are a preferred source of information (Schwingel and Gálvez, Reference Schwingel and Gálvez2016; Nápoles et al., Reference Nápoles, Santoyo-Olsson, Stewart, Ortiz and García-Jímenez2018; Jacquez et al., Reference Jacquez, Vaughn and Suarez-Cano2019; Moyce et al., Reference Moyce, Velazquez, Claudio, Thompson, Metcalf, Aghbashian, Vanderwood and Sisson2021). Compared to previous studies in which physical activity was a preferred method of stress reduction, cleaning the house was a preferred stress management strategy in this study, likely in part because people were spending more time in their homes (Jacquez et al., Reference Jacquez, Vaughn and Suarez-Cano2019). Connecting with others was the second-most common currently used method of stress management, which aligns with CC qualitative themes showing the importance of community and grief associated with restrictions on gathering and with other qualitative work during the pandemic (Moyce et al., Reference Moyce, Velazquez, Claudio, Thompson, Metcalf, Aghbashian, Vanderwood and Sisson2021). Given that survey participants demonstrated a preference for professional-led groups, but “talking with a psychologist” was the lowest-rated stress management tool being used, our results show that there is a critical need for psychotherapists who serve the Latinx community in Cincinnati. Given the lack of Latinx and Spanish-speaking psychotherapists in Cincinnati, the need may be better met if those professionals were supported to integrate into existing community-based initiatives such as groups hosted by churches or social service organizations, or focusing efforts on text-based support or video-based education.
Limitations and future directions
There are a few limitations of this work to note when reviewing these findings. Many survey items were generated by our research team to preserve local relevance, which resulted in a survey with limited available psychometric properties. A few questions used a “select all that apply” response system without an option for participants to indicate “none”; in these cases, missing responses were coded as denial rather than true missingness. Thus, despite the participant response rate being relatively high across survey items, results may be biased towards negative responses due to conflation of negative and missing responses. Ideally, our CC would have been hosted with survey participants and hosted close to the time of survey data collection, but despite efforts to do so there were no survey participants who expressed interest in virtual or in-person group follow-up, and so the reflections of the CC were in a different phase of the pandemic. We are grateful to the local church which hosted our CC, but recognize that their interpretation of our results is more useful than only our own, but may be particular to the religious and geographic community served by the partner church. Specifically, we are eager to offer more insight into the differential experiences of the Guatemalan survey participants compared to those born elsewhere, but we are not able to do so given our inability to re-engage survey respondents. Also, given the Guatemalan and Nicaraguan backgrounds of our community co-researchers, data collection occurred predominantly in ethnic enclaves of these subgroups, which may have limited generalizability. Finally, our predominantly female, married, and with children participant pool left us underpowered to complete extensive within-group comparisons.
Despite these limitations, this work provides important reminders about the heterogeneity of the Latinx immigrant population in the U.S. and about the vulnerability and resilience of the community. Future research should consider the use of cultural consensus modeling to explore the concepts of anxiety and stress among Guatemalans. In addition, further exploration of the differences in experiences within Latinx immigrant groups, particularly in NTIDs, would help to clarify psychosocial needs in subgroups. Specifically, future research by our group would prefer to be able to distinguish the experiences of Latinx community members based on all countries of origin and/or based on the timing of immigration given that many community members beyond Guatemalans have experienced significant and distinct hardship before, during, and after immigration.
Open peer review
To view the open peer review materials for this article, please visit http://doi.org/10.1017/gmh.2024.101.
Supplementary material
The supplementary material for this article can be found at http://doi.org/10.1017/gmh.2024.101.
Data availability statement
The dataset supporting the conclusions is available from the corresponding author on reasonable request.
Acknowledgements
We would like to express our gratitude to the participants of this study, the church that hosted our CC, and our community co-researchers from LU-Salud.
Author contribution
S.S., L.V., and F.J. conceptualized the study; J.P., J.R., K.C., S.S. collected data; S.S., J.R., K.C. were involved in qualitative analysis; S.C. and J.R. were involved in quantitative analysis; S.S., J.R., S.C. were involved in writing the manuscript; F.J., K.C., J.P., L.V. were involved in revising the manuscript.
Financial support
Funding for this work was provided by the UC Next Lives Here Urban Health Pathway 2020 Mental Health Grant. REDCap was hosted by the Cincinnati Center for Clinical and Translational Science and Training (CCTST), which is funded by the National Institutes of Health (NIH) National Center for Advancing Translational Science’s Clinical and Translational Science Award program, grant UL1TR001425. The content is solely the responsibility of the authors and does not represent the official views of the NIH.
Competing interest
The authors declare no Competing interest
Ethics statement
This study was reviewed by the Institutional Review Board of the University of Cincinnati and determined to be exempt from ongoing oversight given its low risk, lack of collection of identifying information, and goal of informing local initiatives.
Open access
Comments
Drs. Bass and Chibanda,
I hope you are well. I am delighted to submit this original research article entitled “Latinx Immigrants’ Stress Experiences and Stress Management Preferences During the COVID-19 Pandemic: Mixed-Methods Results from a Community-Academic Research Team” on behalf of my co-authors for consideration for publication in the journal Global Mental Health. We used a mixed methods design and community-based participatory research principles to guide this work, including working with community-based co-researchers with the team Latinos Unidos por la Salud.
We know that Latinx immigrants were disproportionately affected from a health and socioeconomic perspective during the COVID-19 pandemic, but this study highlights that the differences in experiences within the Latinx immigrant community is incredibly important to acknowledge. We must understand that a one-size-fits-all approach to mental health or social interventions will not meet the needs or priorities of this diverse group of individuals. This work can help guide researchers and clinicians to address stress in Latinx immigrants and to thoughtfully engage with their local community when designing interventions.
We believe that this manuscript is appropriate for publication in the journal Global Mental Health due to your journal’s reputation as a scholarly forum for community-driven research on health promotion and risk factors that compromise mental health.
This manuscript has not been previously published and is not under consideration for publication or in press with any other journal. We have no conflicts of interest to disclose.
Thank you for your consideration.
Shanna D. Stryker, MD, MPH
Assistant Professor
Family Medicine & Psychiatry
University of Cincinnati College of Medicine
Department of Family and Community Medicine
Division of Research & Division of Global Health