Perceptions of Safety and Fear of Seeking Emergency Care After 2025 Anti-Immigrant Policies

Abstract

This cross-sectional study examines safety concerns and fear of seeking emergency care among English- and Spanish-speaking Latino patients following implementation of immigration policy changes in 2025.

Introduction

Anti-immigrant rhetoric and policies have been associated with reduced health care utilization in immigrant communities, a phenomenon described as a chilling effect.¹ In 2017, federal anti-immigrant rhetoric was associated with safety concerns and delays in seeking emergency care among Latino populations, including both immigrants and citizens.² Following federal executive orders and other immigration policy changes issued in 2025,³ we conducted a multisite emergency department (ED) survey to assess whether these concerns differed between Spanish-speaking and English-speaking Latino patients, with English-speaking non-Latino patients as a comparator.

Methods

We performed a cross-sectional survey study of adult ED patients at 3 hospitals (2 sites in San Francisco, California, and 1 site in Houston, Texas) from July 1, 2025, to January 30, 2026. Patients with critical illness, altered mental status, incarceration, psychiatric hold, temporary visitor status, or inability to participate were excluded. Each site obtained approval from its respective institutional review board, and we followed the STROBE reporting guidelines. Each participant verbally consented to participation in their preferred language (English or Spanish).

Using time-block convenience sampling, staff approached eligible patients during weekday shifts and administered surveys verbally in the participant’s preferred language (English or Spanish). We adapted and piloted a previously published instrument² with members of immigrant communities to improve clarity and comprehension (Supplement 1).

Using race, Latino origin, and preferred language extracted from the electronic health record, participants were categorized as Spanish-speaking Latino, English-speaking Latino, or English-speaking non-Latino. Prior work has demonstrated greater concerns about immigration enforcement among Latino populations, including both immigrants and US citizens²; therefore, we hypothesized that perceptions of safety and fear of seeking emergency care would differ across these patient groups.

The primary outcome was fear of seeking ED care due to anti-immigrant rhetoric and policies, assessed by the question, “Did any of these [anti-immigrant rhetoric or policies] make you afraid to come to the emergency department today?” Responses were dichotomized as fear (“Yes, a little,” “Yes, a moderate amount,” or “Yes, a lot”) vs no fear (“No, not at all”). Secondary outcomes included awareness of anti-immigrant rhetoric and policies, feeling unsafe living in the US because of these policies and rhetoric, self-reported delays in seeking ED care, and knowledge of others avoiding the ED because of fear of being discovered as undocumented.

Survey responses were summarized as proportions with 95% CIs. Multivariable logistic regression adjusted a priori for age, sex, belief that anti-immigrant statements would occur, insurance status, and study site based on prior literature² and hypothesized associations with the primary outcome. We assessed multicollinearity among covariates and conducted analyses in R version 2025.09.2 + 418 (R Project for Statistical Computing). Significance was defined as a 2-sided P < .05.

Results

Among 345 participants (mean [SD] age, 47.7 [16.3] years), 139 (40%) were female, 100 (29%) were Spanish-speaking Latino, 101 (29%) were English-speaking Latino, and 144 (42%) were English-speaking non-Latino (Table). Awareness of anti-immigrant rhetoric and policies was high across groups (Spanish-speaking Latino: 91% [95% CI, 83%-96%]; English-speaking Latino: 86% [95% CI, 77%-92%]; English-speaking non-Latino: 88% [95% CI, 81%-92%]). Feeling unsafe living in the US was highest among Spanish-speaking Latino (64% [95% CI, 53%-73%]), followed by English-speaking Latino (52% [95% CI, 41%-62%]) and English-speaking non-Latino (42% [95% CI, 33%-51%]) participants. Fear of seeking ED care was more common among Spanish-speaking Latino (34% [95% CI, 25%-45%]) than English-speaking Latino (4.6% [95% CI, 1.5%-12%]) or English-speaking non-Latino (3.2% [95% CI, 1.0%-8.4%]) participants (Figure). Among 39 participants reporting fear, 16 (41%) described delaying ED care (median [IQR], 2.5 [0.2-37.3] days); 12 (75%) were Spanish-speaking Latino (median [IQR], 5.0 [1.5-37.5] days). Knowledge of others avoiding the ED due to fear of being identified as undocumented was reported by 37 Spanish-speaking Latino (38%), 33 English-speaking Latino (34%), and 28 English-speaking non-Latino (29%) participants.

Table. Patient Characteristics by Study Group.

Characteristic	Patients, No. (%) (N = 345)
	Latino		English-speaking, non-Latino (n = 144)
	Spanish-speaking (n = 100)	English-speaking (n = 101)
Age, mean (SD), y	46.6 (15.8)	44.6 (15.3)	50.7 (17.0)
Sex
Female	48 (48)	46 (46)	45 (31)
Male	52 (52)	55 (55)	95 (66)
Othera	0	0	4 (3)
Uninsured	34 (34)	14 (14)	16 (11)
Unhoused	8 (8)	13 (13)	12 (8)
Has primary care physician	60 (60)	71 (70)	102 (71)
Study site
Zuckerberg San Francisco General Hospital	53 (53)	65 (64)	60 (42)
UCSF Parnassus	5 (5)	17 (17)	46 (32)
Ben Taub General Hospital (Baylor University)	42 (42)	19 (19)	38 (26)

Abbreviation: UCSF, University of California, San Francisco.

^a Includes transgender and nonbinary individuals.

Figure. Bar Graph of Survey Responses Regarding Immigration-Related Perceptions and Emergency Department Care by Study Group.

Bars represent the percentage of respondents answering yes to each survey question. Error bars indicate 95% CIs.

^aParticipant responses were categorized by study staff into predefined categories. Responses indicates one of the following: (1) President Trump wants to deport immigrants (eg, to Panama or Guantanamo Bay); (2) President Trump wants to deny services to immigrants; (3) President Trump wants to prevent immigrants from working in the US; (4) President Trump wants to end birthright citizenship; (5) President Trump has allowed immigration enforcement agents to act in previously safe spaces such as churches, hospitals, and schools; (6) President Trump wants to change how asylum seekers can live in the US; (7) President Trump wants to prevent immigrants from getting health care in the US; or (8) other.

After adjusting for age, sex, belief that anti-immigrant statements would happen, insurance status, and site, Spanish-speaking Latino participants had higher odds of reporting fear of seeking ED care than English speaking non-Latino participants (adjusted odds ratio, 17.38 [95% CI, 5.47-78.75]). No evidence of multicollinearity was identified (all variance inflation factors <2).

Discussion

In this multisite cross-sectional ED study, we found high awareness of anti-immigrant rhetoric and policies across Spanish-speaking Latino, English-speaking Latino, and English-speaking non-Latino groups. However, Spanish-speaking Latino participants had substantially higher fear of seeking ED care compared with English-speaking non-Latino participants, whereas English-speaking Latino participants did not differ significantly. Moreover, nearly one-half of participants who reported fear seeking ED care delayed their emergency care, which is associated with higher acuity illness on presentation.⁴

Limitations include the inability to directly assess immigration or documentation status, precluding assessment of whether observed differences reflected immigration-related experiences within the study groups. Self-reported fear of seeking ED care, weekday convenience sampling that may have underrepresented patients presenting after hours or delaying care, residual confounding, and the small number of outcome events may have reduced the precision of the effect size estimates; subgroup comparisons should therefore be interpreted cautiously. Unmeasured structural factors, including local immigration policy environments and barriers to health care access unrelated to immigration concerns, may also have contributed to the observed associations.

Ensuring that EDs are perceived as safe and accessible may reduce fear-driven avoidance of care among Spanish-speaking Latino populations. Future studies should evaluate interventions to improve ED access.

Supplement 1.

eMethods. Survey

Supplement 2.

Data Sharing Statement