The Impact of Local Immigration Enforcement Policies on the Health of Immigrant Hispanics/Latinos in the United States

Abstract

Objectives. We sought to understand how local immigration enforcement policies affect the utilization of health services among immigrant Hispanics/Latinos in North Carolina.

Methods. In 2012, we analyzed vital records data to determine whether local implementation of section 287(g) of the Immigration and Nationality Act and the Secure Communities program, which authorizes local law enforcement agencies to enforce federal immigration laws, affected the prenatal care utilization of Hispanics/Latinas. We also conducted 6 focus groups and 17 interviews with Hispanic/Latino persons across North Carolina to explore the impact of immigration policies on their utilization of health services.

Results. We found no significant differences in utilization of prenatal care before and after implementation of section 287(g), but we did find that, in individual-level analysis, Hispanic/Latina mothers sought prenatal care later and had inadequate care when compared with non-Hispanic/Latina mothers. Participants reported profound mistrust of health services, avoiding health services, and sacrificing their health and the health of their family members.

Conclusions. Fear of immigration enforcement policies is generalized across counties. Interventions are needed to increase immigrant Hispanics/Latinos’ understanding of their rights and eligibility to utilize health services. Policy-level initiatives are also needed (e.g., driver’s licenses) to help undocumented persons access and utilize these services.

Federal immigration enforcement policies have been increasingly delegated to state and local jurisdictions, leading to increased enforcement activities by local police. This shift has resulted largely from the implementation of 2 federal initiatives: section 287(g) of the Immigration and Nationality Act and the Secure Communities program. Section 287(g) authorizes Immigration and Custom Enforcement to enter into agreements with state and local law enforcement agencies to enforce federal immigration law during their regular, daily law enforcement activities. The original intention was to “target and remove undocumented immigrants convicted of violent crimes, human smuggling, gang/organized crime activity, sexual-related offenses, narcotics smuggling and money laundering.”¹ Added to the Immigration and Nationality Act in 1996, section 287(g) was not widely used in its first decade, but its use accelerated in the mid- to late 2000s.^2,3

The Secure Communities program differs from section 287(g) in that it does not authorize local enforcement bodies to arrest individuals for federal immigration violations. Instead, when individuals are arrested for nonimmigration matters, the Secure Communities program facilitates the sharing of local arrestees’ fingerprints and information with Immigration and Custom Enforcement and the Federal Bureau of Investigation, which checks them against immigration databases. If these checks reveal that an individual is unlawfully present in the United States or otherwise removable because of a criminal conviction, Immigration and Custom Enforcement takes enforcement action.⁴

Some evidence suggests that both section 287(g) and the Secure Communities program contribute to Hispanic/Latino immigrants’ general mistrust of local law enforcement and fear of utilizing a variety of public services, such as police protection and emergency services.^2,5–7 Although many immigrant Hispanics/Latinos in the United States experience barriers to care because of a lack of bilingual and bicultural services, low health literacy, insufficient public transportation, and limited knowledge of available health services,^8–12 studies have suggested that individuals lacking legal status may have more difficulty obtaining health services and may experience worse health outcomes than do individuals with legal status.^13–18 Among immigrant Hispanics/Latinos, the fear of deportation, a lack of required forms of documentation, interaction with law enforcement personnel, and racial profiling are factors also associated with reduced utilization of health services and worse health.^6,19–22 Such fears lead to incomplete sequences of care,^19,20,23,24 promote the use of nonstandard and unsafe contingencies for care,^16,25–27 and contribute to public health hazards, as immigrants delay preventive care or treatment.^13,22,28 These fears further affect long-term health outcomes as immigrant Hispanics/Latinos alter their physical activity, food purchasing behaviors, and food consumption because of concerns about being in public.²⁹ They may withhold information from health care providers¹⁹ and experience high levels of stress, leading to compromised mental health.^20,30,31

The Patient Protection and Affordable Care Act bars undocumented or recent legal immigrants from receiving financial assistance for health insurance³²; thus, many will continue to remain uninsured and dependent on public health services and free clinics for a significant portion of their care. Because these services are associated with government authority, there is the potential that increasing immigration enforcement policies will deter noncitizens from seeking needed care, not only to their detriment but also to the detriment of public health.

Currently there is little research examining the impact of recent immigration enforcement policies on the access to and utilization of health care, and there has been a call to better understand the public health impact of current immigration policies and their enforcement.²⁹ Through mixed methods, we explored the effect of local immigration enforcement policies on access to and utilization of health services among immigrant Hispanics/Latinos in North Carolina. We analyzed vital records data to determine whether there were differences in utilization of prenatal services by Hispanic/Latina mothers pre- and postimplementation of section 287(g), and we conducted focus groups and in-depth interviews with Hispanics/Latinos living in counties that had implemented section 287(g) and in “sanctuary” counties, counties in which leaders, including politicians and clergy, have spoken out against the program.

METHODS

We analyzed publicly available vital records data from the North Carolina Vital Statistics System. The North Carolina State Center for Health Service collected the data, which contained birth certificate data for all infants born in the state.

Dependent Variables

We used 2 standard indicators of prenatal care: late entry into care and inadequate care. For analysis, we created dichotomous indicator variables for each measure. We defined late prenatal care as entering care after the first trimester of pregnancy or not receiving any prenatal care.

We determined inadequate prenatal care using the Kotelchuck Index, which classifies prenatal care utilization into 4 categories: (1) adequate-plus, comprising care begun by the fourth month and 110% or more of recommended visits attended; (2) adequate, comprising care begun by the fourth month and 80% to 109% of recommended visits attended; (3) intermediate, comprising care begun by the fourth month and 50% to 79% of recommended visits attended; and (4) inadequate, comprising care begun after the fourth month or less than 50% of recommended visits attended.^33,34 We defined inadequate prenatal care as not receiving adequate or adequate-plus care as measured by the Kotelchuck Index.

Both indicators are calculated on the basis of the month prenatal care began. The month prenatal care began was missing for 0.96% of birth records. We coded missing data as late or inadequate prenatal care, as recommended, because missing month indicates no prenatal care or not following prenatal care guidelines.³⁵

Independent Variables

Ethnicity of mother.

Hispanic/Latina mothers were mothers who self-identified as Hispanic/Latina or of Mexican or Central or South American origin. We coded all other mothers as non-Hispanic/Latina.

Section 287(g) implementation.

If a county implemented 287(g), we coded it as 1 (yes).

Pre- and Postimplementation of 287(g)

Because we wanted to capture the potential effect of section 287(g) during the entire length of each pregnancy, we used birth data from the 9 months preceding implementation (coded as preimplementation) and birth data collected between 9 and 18 months after implementation (coded as postimplementation). Preimplementation and postimplementation periods differ in each county because each county did not implement section 287(g) at the same time.

We matched the 7 counties in North Carolina that had implemented section 287(g) with 7 nonimplementation counties using propensity scores with nearest-neighbor matching.³⁶ We matched communities on the percentage below the poverty level, uninsured, and Spanish speaking using data from the 2010 American Community Survey for the years 2005 to 2009.

For the primary county-level analysis, we used a difference-in-difference approach³⁷ to compare changes in prenatal care among Hispanic/Latina mothers in the 7 counties that implemented section 287(g) with those in 7 matched counties that did not implement section 287(g). Difference-in-difference is a method commonly used in econometrics to estimate the effect of a treatment on an outcome by comparing the average change over time in an outcome variable for a treatment group with a control group.^37,38 Our treatment group was composed of counties that implemented section 287(g), and the control group was composed of counties that did not implement section 287(g).

We compared vital records data from preimplementation to postimplementation in each county. We used the corresponding pre- and postimplementation period as defined for a specific section 287(g) county for its matched county. Table 1 provides section 287(g) implementation and matched nonimplementation counties and periods of data used in analyses. The preimplementation sample had 7663 and the postimplementation sample had 7593 births to Hispanic/Latina mothers.

TABLE 1—

Counties Implementing Section 287(g), Matched Counties, and Period of Data: North Carolina, 2012

		Preimplementation Period (9–0 Mo Before 287(g) Implementation)			Postimplementation Period (9–18 Mo After 287(g) Implementation)
287(g) County	Matched County	From	To	Implementation of 287(g)	From	To
Alamance	Caldwell	4/10/2006	1/9/2007	1/10/2007	10/11/2007	7/10/2008
Cabarrus	Dare	11/2/2006	8/1/2007	8/2/2007	5/3/2008	2/2/2009
Durham	Burke	5/1/2007	1/31/2008	2/1/2008	11/2/2008	8/1/2009
Gaston	Moore	5/22/2006	2/21/2007	2/22/2007	11/23/2007	8/22/2008
Henderson	Buncombe	9/25/2007	6/24/2008	6/25/2008	3/26/2009	12/25/2009
Mecklenburg	Carteret	5/27/2005	2/26/2006	2/27/2006	11/28/2006	8/27/2007
Wake	Forsyth	9/25/2007	6/24/2008	6/25/2008	3/26/2009	12/25/2009

Note. Section 287(g) of the Immigration and Nationality Act and the Secure Communities program authorizes the US Immigration and Custom Enforcement to enter into agreements with state and local law enforcement agencies to enforce federal immigration law during their regular, daily law enforcement activities.

We also conducted a secondary individual-level analysis to compare the prenatal care of Hispanic/Latina mothers to non-Hispanic/Latina mothers in all counties in North Carolina during the 9-month period before implementation of section 287(g) (May 2005–February 2006) with the 9- to 18-month period after the last county implemented section 287(g) (March–December 2009), adjusting for county and whether the mother lived in a county that had implemented section 287(g). There were 39 200 births during the preimplementation and 28 984 during the postimplementation periods.

We conducted all analyses using the DIFF procedure in Stata version 12.1.³⁹

Focus Groups and In-Depth Interviews

In fall 2012, we conducted focus groups and individual in-depth interviews to explore the extent to which local immigration enforcement policies were related to access to and utilization of public health services among Hispanics/Latinos. Standardized focus group and interview guides outlined the process and were adhered to in the discussions. We crafted the guides in Spanish with careful consideration to wording, sequence, and content, and we included open-ended questions related to immigration experiences, utilization of health care services, and perceptions of immigration enforcement policies. We obtained participants’ written informed consent.

We conducted 3 focus groups and 9 interviews in the 3 North Carolina counties in which section 287(g) had been implemented for the longest period. We conducted 3 focus groups and 8 interviews in 3 sanctuary counties. Local community-based organizations in each county helped to recruit participants for both the focus groups and the in-depth interviews through purposive snowball sampling approaches to ensure a broad spectrum of participants.⁴⁰ Eligibility criteria were being aged 18 years or older, self-identifying as Hispanic or Latino, speaking Spanish, and living in the county in which we held the focus group or interview. Local community-based organizations provided facilities to conduct focus groups and interviews.

We collected participant demographic data using a brief Spanish-language assessment that included age, gender, country of origin, educational attainment, employment status, languages spoken, health status, and amount of time having lived in North Carolina. We have successfully used these items with immigrant Hispanics/Latinos previously.^41,42 Because of the sensitive nature of directly asking whether a participant has documentation, we used a valid state-issued driver’s license as a proxy for documentation; at the time of data collection, the North Carolina Department of Motor Vehicles had not issued driver’s licenses to anyone without social security numbers since 2006.⁴³

Bilingual, bicultural, and trained staff audio recorded each focus group and interview with participant permission and conducted them in Spanish. Staff were experienced in qualitative cross-cultural health research. Three were native Spanish speakers; 2 were women, and 2 were men. Two moderators were present during each focus group. One or 2 interviewers were present during each interview. We compensated participants $40 for their time. Focus groups and interviews averaged 90 and 60 minutes, respectively.

A professional transcriptionist transcribed the audio-recorded discussions verbatim. We analyzed transcripts to identify themes using constant comparison, an approach to grounded theory.⁴⁴ The analysis team, which included community members, organizational representatives, and academic researchers, read and reread transcripts, developed themes, and worked together to reconcile and interpret themes.

We explored demographic characteristics of focus group and interview participants using descriptive statistics using SPSS version 19.0 (IBM, Somers, NY).

RESULTS

Using difference-in-difference analysis of the matched-county data (n = 14), we found no significant differences in entry into and adequacy of prenatal care among Hispanic/Latina mothers pre- and postimplementation of section 287(g) by implementation status. Rates of late and inadequate prenatal care increased among Hispanic/Latina mothers in counties implementing section 287(g) and decreased among Hispanic/Latina mothers in nonimplementing counties; however, these differences were not significant (Table 2).

TABLE 2—

County-Level Late and Inadequate Prenatal Care at Pre– and Post–Section 287(g) Implementation, by County Status Among Hispanic/Latina Mothers: North Carolina, 2012

	Late Prenatal Care, %			Inadequate Prenatal Care, %
Section 287(g) Status	Pre-287(g)	Post-287(g)	Difference-in-Difference	Pre-287(g)	Post-287(g)	Difference-in-Difference
287(g) county (n = 7)	33.82	34.28		35.62	37.28
Non-287(g) county (n = 7)	26.78	29.98		31.08	26.96
Differences	7.04 (P = .52)	8.30 (P = .45)	1.30 (P = .94)	4.54 (P = .65)	10.33 (P = .31)	5.79 (P = .68)

Note. Section 287(g) of the Immigration and Nationality Act and the Secure Communities programs authorizes the US Immigration and Custom Enforcement to enter into agreements with state and local law enforcement agencies to enforce federal immigration law during their regular, daily law enforcement activities.

In the individual-level analysis that compared Hispanic/Latina with non-Hispanic/Latina mothers throughout North Carolina, we found that Hispanic/Latina mothers were more likely to have late and inadequate prenatal care both before and after section 287(g) implementation than were non-Hispanic/Latina mothers, after adjusting for county and whether it had adopted section 287(g). Moreover, for late prenatal care, the difference between Hispanic/Latina and non-Hispanic/Latina mothers was significantly greater during the postimplementation period of section 287(g) than during the preimplementation period after adjusting for county and section 287(g) implementation status. The magnitude of the difference was 1.5% (P = .03). For inadequate prenatal care, the difference between Hispanic/Latina and non-Hispanic/Latina mothers was greater during the postimplementation period of section 287(g) after adjusting for county and section 287(g) implementation status. The magnitude of the difference was 1.1%, which is marginally significant (P = .1; Table 3).

TABLE 3—

Individual-Level Late and Inadequate Prenatal Care at Pre– and Post–Section 287(g) Implementation, by Ethnicity Among Hispanic/Latina and Non-Hispanic/Latina Mothers: North Carolina, 2012

	Late Prenatal Care,a %			Inadequate Prenatal Care,a %
Race/Ethnicity	Pre-287(g)	Post-287(g)	Difference-in-Difference	Pre-287(g)	Post-287(g)	Difference-in-Difference
Hispanic/Latina (n = 15 256)	28.2	29.8		29.1	30.6
Non-Hispanic/Latina (n = 62 928)	9.9	10.1		7.9	8.4
Differences	18.3 (P < .01)	19.7 (P < .01)	1.5 (P = .03)	21.1 (P < .01)	22.2 (P < .01)	1.1 (P = .13)

Note. Section 287(g) of the Immigration and Nationality Act and the Secure Communities program authorizes the US Immigration and Custom Enforcement to enter into agreements with state and local law enforcement agencies to enforce federal immigration law during their regular, daily law enforcement activities.

^aAdjusted for county and section 287(g) implementation status.

Characteristics of Focus Group and Interview Participants

We held 1 focus group and 2 to 4 interviews in each of the 6 study counties (i.e., Alamance, Buncombe, Chatham, Gaston, Mecklenburg, and Orange counties). We enrolled 66 focus group and 17 interview participants. Focus groups ranged in size from 8 to 12 participants. The mean age of participants was 37 (range = 18–86) years, three fifths were women, the majority were from Mexico, most had a high school diploma (or equivalent) or less, and slightly more than one third reported being employed full time. Participants, on average, had lived in North Carolina for more than a decade. Only 13% reported having health insurance, and 40% reported having a valid driver’s license. Participant characteristics are provided in Table 4.

TABLE 4—

Select Characteristics of Focus Group and In-Depth Interview Participants: North Carolina, 2012

	Mean ±SD (range) or No. (%)
Characteristic	Focus Group (n = 66)	Interview (n = 17)	Total (n = 83)
Age, y	37.52 ±12.90 (19.00–86.00)	36.00 ±12.60 (18.00–64.00)	37.20 ±12.80 (18.00–86.00)
Country of origin
Mexico	36 (54.5)	13 (76.5)	49 (59.0)
El Salvador	10 (15.2)	1 (5.9)	11 (13.3)
Dominican Republic	5 (7.6)	0 (0.0)	5 (6.0)
Guatemala	4 (6.1)	1 (5.9)	5 (6.0)
Cuba	2 (3.0)	0 (0.0)	2 (2.4)
Othera	9 (13.6)	2 (11.8)	11 (13.3)
Years living in North Carolina	11.84 ±7.90 (0.60–43.70)	13.03 ±7.40 (1.70–38.20)	12.08 ±7.80 (0.60–43.70)
Current county
Alamance	12 (18.2)	4 (23.5)	16 (19.3)
Buncombe	12 (18.2)	4 (23.5)	16 (19.3)
Chatham	12 (18.2)	2 (11.8)	14 (16.9)
Gaston	8 (12.1)	2 (11.8)	10 (12.0)
Mecklenburg	12 (18.2)	3 (17.6)	15 (18.1)
Orange	10 (15.1)	2 (11.8)	12 (14.4)
Gender
Female	42 (63.6)	11 (64.7)	53 (63.9)
Male	24 (36.4)	6 (35.3)	30 (36.1)
Current marital status
Single	14 (21.9)	4 (23.5)	18 (22.2)
Married or partnered	47 (73.4)	9 (52.9)	56 (69.1)
Separated or divorced	3 (4.7)	3 (17.6)	6 (7.4)
Widowed	0 (0.0)	1 (5.9)	1 (1.2)
Household composition
Spouse	44 (66.7)	9 (52.9)	53 (63.9)
Children	27 (40.9)	8 (47.1)	35 (42.2)
Other family members	15 (22.7)	4 (23.5)	19 (22.9)
Friends	2 (3.0)	1 (5.9)	3 (3.6)
Alone	3 (4.5)	2 (11.8)	5 (6.0)
Educational attainment
< high school diploma or equivalent	21 (32.8)	9 (52.9)	30 (37.0)
High school diploma or equivalent	20 (31.3)	4 (23.5)	24 (29.6)
> high school diploma or equivalent	23 (35.9)	4 (23.5)	27 (33.3)
Current employment
Full time	22 (34.4)	8 (47.1)	30 (37.0)
Part time	7 (10.9)	4 (23.5)	11 (13.6)
Multiple jobs	2 (3.1)	1 (5.9)	3 (3.7)
Unemployed	33 (51.6)	4 (23.5)	37 (45.7)
Forms of employment in US, past 12 mo
Food services	9 (13.6)	5 (29.4)	14 (16.9)
Child care	8 (12.1)	3 (17.6)	11 (13.3)
Construction	8 (12.1)	2 (11.8)	10 (12.0)
Manufacturing	8 (12.1)	2 (11.8)	10 (12.0)
Clerical	7 (10.6)	3 (17.6)	10 (12.0)
Odd jobs	8 (12.1)	2 (11.8)	10 (12.0)
Furniture	4 (6.1)	3 (17.6)	7 (8.4)
Janitor	3 (4.5)	1 (5.9)	4 (4.8)
Cashier	3 (4.5)	1 (5.9)	4 (4.8)
Animal processing	2 (3.0)	1 (5.9)	3 (3.6)
Lawn care	3 (4.5)	0 (0.0)	3 (3.6)
Hairstylist	2 (3.0)	0 (0.0)	2 (2.4)
Management	2 (3.0)	0 (0.0)	2 (2.4)
Sales	2 (3.0)	0 (0.0)	2 (2.4)
Other	17 (25.8)	2 (11.8)	19 (22.9)
No work	9 (13.6)	2 (11.8)	11 (13.3)
Language spoken most comfortably
Only Spanish	25 (38.5)	6 (35.3)	31 (37.8)
More Spanish than English	24 (36.9)	3 (17.6)	27 (32.9)
Both equally	13 (20.0)	8 (47.1)	21 (25.6)
More English than Spanish	2 (3.1)	0 (0.0)	2 (2.4)
Only English	0 (0.0)	0 (0.0)	0 (0.0)
Other	1 (1.5)	0 (0.0)	1 (1.2)
Language written most comfortably
Only Spanish	27 (43.5)	8 (47.1)	35 (44.3)
More Spanish than English	19 (30.6)	3 (17.6)	22 (27.8)
Both equally	14 (22.6)	5 (29.4)	19 (24.1)
More English than Spanish	1 (1.6)	1 (5.9)	2 (2.5)
Only English	1 (1.6)	0 (0.0)	1 (1.3)
Currently has health insurance	8 (12.7)	2 (11.8)	10 (12.5)
Self-perceived general physical health
Excellent	10 (15.9)	2 (11.8)	12 (15.0)
Very good	13 (20.6)	3 (17.6)	16 (20.0)
Good	19 (30.2)	6 (35.3)	25 (31.3)
Fair	19 (30.2)	4 (23.5)	23 (28.7)
Poor	2 (3.2)	2 (11.8)	4 (5.0)
Currently has valid driver’s license	27 (42.2)	5 (29.4)	32 (39.5)

^aOther are Ecuador, Nicaragua, Panama, and Peru.

Themes From Focus Groups and Individual Interviews

Seven themes emerged related to immigration enforcement policies and health (Box 1), and are discussed below.

Qualitative Themes and Select Quotations From Focus Groups and In-Depth Interviews: North Carolina, 2012

1. Immigration enforcement policies are not well understood.

“I have heard of it [section 287(g)], but not by name or the details of it. We all know something is happening and the laws are changing, but I don’t really know what it is.” -Focus group participant from a section 287(g) county

“I thought it was only when driving that they could ask you for your license. But now I know that that’s not the way it is. Even if a person is just walking on the street, someone can ask you for your documents.” -Interview participant from a section 287(g) county

2. Immigration enforcement policies compound existing distrust of services.

“That’s why they don’t want to go [to the doctor]. They are scared that [staff] will call the police.” -Interview participant from a section 287(g) county

“Health [services are] not the only services I do not trust. You cannot call the police to report a crime; using the [public] library is a risk for us Latinos.” -Focus group participant from a non–section 287(g) county

3. Immigration enforcement policies are perceived as condoning racism.

“It’s like [some people] feel that they have the support of the US government. These programs give them the feeling that they have the right to say, ‘Latinos have no business being here.’” -Focus group participant from a section 287(g) county

“Police see that someone is Hispanic and stop them even though everything is fine. Why? Because they say that there are now laws that give them the right to do anything. Now, they just see that you are Hispanic, and the police won’t leave you alone.” -Focus group participant from a section 287(g) county

“And in this town you can see it now because years ago it did not used to be like that. There is racism everywhere. We went to the store to exchange a pair of shoes and to do so, they asked for my [driver’s] license. Yes, to exchange a pair of shoes, they asked for my license. They only did it because I was Hispanic. They did not want to serve me.” -Focus group participant from a section 287(g) county

4. Immigration enforcement policies create practical barriers to accessing and utilizing health services.

“Recently, someone told me there is a very good clinic where they have comprehensive services, interpreters, and have more health programs including dental services. However, it is in [a neighboring county]. So, if I go there, I take the risk that a police officer will stop me. These are extra challenges to go to that place.” -Interview participant from a non–section 287(g) county

“A woman was sick, and the husband went to pick up her prescription medication for her. And [the hospital pharmacy] asked him, ‘Do you have identification?’ He answered, ‘I don’t have any [identification].’ So the pharmacist wouldn’t give the medicine to him. He had to leave and get a relative who is a citizen or resident or whatever so that [the relative] could get the medication for him.” -Focus group participant from a non–section 287(g) county

5. Immigration enforcement policies reduce physical and mental health.

“When my sister-in-law was pregnant, she had not gone to the doctor because she was undocumented. When she started feeling bad, my brother took her to the hospital. At that point, she was 7 months pregnant!” -Focus group participant from a section 287(g) county

“Many of us no longer have that desire to fight. We had that strength before. I did get into a depression. It was an ugly depression, and I didn’t know when it would be over.” -Interview participant from a section 287(g) county

“I wouldn’t, I couldn’t go beyond the door [of my house] for over a year. I was too fearful of what might happen.” -Focus group participant from a section 287(g) county

6. Immigration enforcement policies profoundly compromise child health.

“I used to take my kids to [a hospital in a neighboring community], but I can no longer take them because there are many [driver’s license] checkpoints.” -Focus group participant from a non–section 287(g) county

“There was a 4-year-old girl who had a nervous attack. She couldn’t sleep well because she had overheard that her dad was going to be deported because the police stopped him. That girl already knew that the family could fall apart. Children understand what it means to be stopped by the police, and they fear that their parents will be deported.” -Focus group participant from a section 287(g) county

“These policies affect children a lot. The 3 months that they are on vacation [from school], they spend at home. We cannot take them to the zoo. We take them to the movies because it is nearby. Even when we do that, we are scared. It wasn’t like before when I had a [driver’s] license. When they go back to school, what do they ask them? ‘Where did you go on vacation?’ [They must reply] ‘Nowhere.’” -Focus group participant from a non–section 287(g) county

7. Immigration enforcement policies lead to use of nonstandard and risky contingencies for care.

“That’s the reason why many people bear it or look for medicine in the Mexican stores. If a person is vomiting, what is the person going to do? It’s because they are afraid to go to the doctor. They look for alternatives. That’s the reason.” -Focus group participant from a non–section 287(g) county

“It’s difficult; I am going to tell you but I won’t mention names or anything, but here in Charlotte and in many parts of the US, dentists from other countries come here and in your home they will do everything and charge you really cheap.” -Focus group participant from a section 287(g) county

Note. Section 287(g) of the Immigration and Nationality Act and the Secure Communities program authorizes the US Immigration and Custom Enforcement to enter into agreements with state and local law enforcement agencies to enforce federal immigration law during their regular, daily law enforcement activities.

Understanding of immigration enforcement policies.

Participants in both implementation and nonimplementation counties perceived a growing anti-immigration sentiment in North Carolina, and they perceived that policies were becoming more hostile and restrictive. However, participants were not clear about the specific details of policies, and they lacked an understanding of their rights.

Compounding of existing distrust of services.

Participants reported that they did not trust staff at agencies providing services, including staff at public and private agencies providing health care, public safety, other social services, and libraries, as examples. They did not access or utilize health services for which they were eligible, including preventive services (e.g., reproductive health services). They worried that when visiting public health and free clinics their lack of documentation (e.g., social security numbers and valid driver’s licenses) would put them at risk for detention and deportation.

Condoning racism.

Participants expressed concern that immigration enforcement policies were exacerbating anti-immigrant sentiments and promoting racial profiling and discrimination, including within health care settings. Among participants who reported having utilized health care, some perceived discrimination by providers and other staff, including interpreters, which they linked to not receiving necessary, timely, and quality care. Moreover, participants reported experiencing racial profiling by law enforcement while driving and indicated that some immigrant Hispanics/Latinos avoid driving even in the case of a medical emergency.

Creating practical barriers to accessing and utilizing health services.

Ineligibility for state-issued driver’s licenses and the threat of police enforcement while driving were 2 key barriers that participants identified as preventing Hispanics/Latinos from accessing and utilizing health services. Participants indicated that, although immigrant Hispanics/Latinos who do not have driver’s licenses may drive to their jobs out of necessity, they may not “risk” driving to access health services. Some participants described relying on rides from others such as friends and family to seek care when seeking care was unavoidable (e.g., childbirth).

Reducing physical and mental health.

Participants reported that they and others they know often do not seek care when needed. As a result, they delay important preventive services and remain sicker for longer periods of time. Participants also described reduced mental health for family members resulting from high levels of fear associated with immigration enforcement policies. Mental health impacts highlighted included isolation, frustration, decreased self-esteem, anxiety, and depression. Some participants also reported reduced opportunities for social support because of fearing to leave their home to spend time with others.

Compromising child health.

Participants emphasized the impact of enforcement policies on children. They feared being asked to present identification when seeking services at a public health or free clinic and feared that they could be discovered as undocumented while trying to obtain prescription drugs for sick children. Participants also provided examples when children did not obtain necessary diagnoses, care, and treatment because their parents feared being identified as undocumented at a checkpoint while driving to a provider.

Participants reported that many children of immigrant Hispanics/Latinos are aware of the limitations their families face because of immigration policies and share the fear of enforcement that their parents feel. This fear manifests in mental health consequences. Participants described being reoccupied with avoiding interactions with systems, suspicious of those in positions of power (including health care providers), and fearful of being detained and deported. Because of these profound concerns, they reported engaging in few recreational activities and seeing extended family less frequently. The consequences of this isolation included emotional distress.

Using nonstandard and unsafe contingencies for care.

Participants reported that they often rely on nonmedical sources of care, such as self-diagnosing and self-treating and using medications purchased from Latino stores, brought from their home country, or left over from others’ prescriptions.

DISCUSSION

Our research provides a picture of how immigration enforcement policies affect access to and utilization of public health services among Hispanics/Latinos. Although differences in entry into or adequacy of prenatal care among Hispanic/Latina mothers pre- and postimplementation of section 287(g) were not significant in the matched-county analysis, when we compared all Hispanic/Latina mothers to all non-Hispanic/Latina mothers, we found that Hispanic/Latina mothers were more likely to have both late and inadequate prenatal care.

In the matched-county analysis, our failure to find utilization of prenatal services differing by local immigration enforcement policies can be attributed to several factors. First, measurement may have lacked sufficient power, precision, or sensitivity. There was temporal overlap of implementation of section 287(g) and the Secure Communities program. By December 2010, 75% of North Carolina counties had also implemented the Secure Communities program; thus, the implementation of the Secure Communities program may have obscured the effect of section 287(g).

Moreover, fear of immigration enforcement policies may generalize across county lines. Heyman et al.¹⁸ found that barriers to care on the basis of immigration status cross jurisdictional boundaries. Berk and Schur found that California’s anti-immigration Proposition 187 (adopted in 1994) appeared to deter care seeking in Texas as well as in California, perhaps because of the national or regional publicity it received.⁴⁵

Participants in our study reported knowing that enforcement was increasing, but many lacked information about actual programs; many did not know whether the county that they lived in implemented section 287(g). Overall, participants reported fearing immigration enforcement policies, avoiding health services, and thus sacrificing their own health and the health of members of their families.

Implications for Intervention

Interventions that increase knowledge of immigrant rights and demystify the process to access health care services are needed. Immigrant Hispanics/Latinos would benefit from guidance about their rights, what they are obligated to say when approached by law enforcement officials, and what documentation is necessary to access health services. For example, organizations such as the American Civil Liberties Union have created pamphlets to educate immigrants about their rights, but there is a need to get these types of materials, and materials like them, into the hands of those who need them.

There remains a need for lower literacy materials as well. Similarly, educating immigrant Hispanics/Latinos about the procedures of accessing health services (e.g., how to make an appointment, what documentation is necessary, and what financial assistance is available) and building their efficacy to overcome the challenges faced when accessing a public health department or free clinic (e.g., the security guard at the front door and the lack of interpreters) may assist in appropriate and timely utilization of services. Reaching immigrant Hispanics/Latinos will require creativity. Using Spanish-language media, including radio, may be a strategy to increase knowledge of rights, eligibility, and health care access. A structural intervention may include training providers, including frontline staff, to provide linguistically and culturally congruent and immigrant-friendly services.

Policy-level interventions may also be needed. Sixty percent of participants did not have valid state-issued driver’s licenses. Much of the fear expressed by participants was related to their worries of driving without licenses because local police who enforce traffic laws may also be involved in immigration enforcement. The hesitation to drive delays diagnosis and treatment and utilization of prevention services, such as prenatal care. A policy-level intervention designed to improve public health may be increasing access to driver’s licenses among immigrants who are undocumented. Such interventions have been proposed to increase the numbers of drivers who are aware of traffic rules and regulations and decrease the number of uninsured drivers.⁴⁶ Future research should determine how access to driver’s licenses affects access and utilization of health services and overall community health.

Conclusions

Our research suggests that immigration policies affect the health of immigrant Hispanics/Latinos, including those with and those without documentation. Hispanics/Latinos are less likely to utilize prenatal services and reported profound fear about accessing and utilizing services for which they are eligible.

Prenatal care is a commonly needed and often understood service. If immigrant Hispanic/Latino uptake of prenatal services is late or inadequate, accessing and utilizing other types of health care services may be even more influenced by immigration policies. For example, health needs that are less common or even stigmatized, such as testing and treatment of sexually transmitted infections and HIV, may be more delayed, resulting in disease transmission and further infections.

Public officials should consider these possibilities in deciding whether, and to what extent, they should undertake local enforcement of federal immigration policies. There is a need to seek concrete solutions to address immediate health needs in the context of these policies and to work toward long-term solutions.