Original Research | Published: 22 April 2024
open-url

Persistent effects of legal status on healthcare access and outcomes: findings from a state-wide representative cross-sectional survey in California

https://doi.org/10.1136/bmjph-2023-000800

Request reuse permissionopen-url
This is an open access article distributed in accordance with the Creative Commons Attribution Non Commercial (CC BY-NC 4.0) licenseopen-url

Abstract

Background To examine how legal status and past undocumented status are associated with healthcare access and health outcomes.

Methods Data were collected between 2018 and 2020 as a follow-on, cross-sectional survey to the California Health Interview Survey (N=2010). We used multiple regression models to assess associations between past and current legal status and usual source of care, delayed medical care and psychological distress.

Results Overall, 26.2% of the sample had ever been undocumented. Compared with citizens who have always held lawful status (CLS), non-citizens who were previously undocumented (adjusted OR (aOR)=0.48, p<0.05) and non-citizens who have always held lawful status (NLS) (aOR=0.41, p<0.01) were less likely to have a usual source of care. Citizens who were previously undocumented were more likely to delay medical care (aOR=2.32, p<0.05) compared with CLS. NLS were more likely to have moderate and above distress (aOR=2.31, p<0.01) compared with CLS.

Conclusion Public health efforts are needed to address the burden of trauma and disadvantage among those experiencing persistent effects of undocumented status.

What is already known on this topic

  • There is increasing recognition that legal status can change across the life of an immigrant; however, no studies exist on the general immigrant population that examine how having a history of undocumented status may be associated with healthcare access and outcomes.

What this study adds

  • This is the first quantitative study to examine how having a history of undocumented status influences access to care, delayed medical care and psychological distress using the largest state-level, representative survey.

  • The study found that having a history of being undocumented was associated with less access to usual source of care and delayed care compared with citizens who always held lawful status. Among those who always held lawful status, non-citizens were more likely to report moderate psychological distress compared with citizens.

How this study might affect research, practice or policy

  • Our findings are in line with past qualitative immigrant studies that suggest that legal status may have lasting effects on health.

  • Public health efforts are needed to address the persistent effects of ever having undocumented status.

Introduction

In 2022, 45.5 million immigrants lived in the USA, including 24 million naturalised citizens and 21 million non-citizens.1 Non-citizens include those who are lawfully present as well as 11 million undocumented immigrants, of which 15% and 67% represent those from Asia and Mexico/Central America, respectively.2 Citizenship and legal status are critical determinants of health stratification across this population.3 Non-citizens, including undocumented, may also experience elevated stress due to fear of immigration enforcement actions and anti-immigrant sentiments.4–6 Undocumented immigrants, in particular, are more likely to get sick, yet less likely to have a regular healthcare provider, visit a doctor or use emergency department services compared with those who are documented,7–9 and experience high levels of psychological distress, depression and anxiety.10 11

Restrictive immigrant policies contribute to health inequities through restricted access to social benefits (eg, employment opportunities, education, etc), material conditions (eg, housing and income) and health-related services (eg, decreased healthcare access and insurance).12 Availability of health insurance options likely contribute to these inequities as naturalised citizens qualify for employer-based insurance as well as federal programmes such as Medicaid and the Children’s Health Insurance Program. Non-citizens face a number of eligibility restrictions such as a 5-year waiting period to qualify for federal programmes, while those undocumented are excluded from public insurance options completely except for state-level exceptions. California, for example, expanded the state’s Medicare programme to all low-income undocumented residents in 2024, yet it is projected that over 500 000 will remain uninsured due to income requirements. Non-citizens are significantly more likely to report being uninsured than citizens. Approximately 50% of undocumented report being uninsured compared with 18% of lawfully present non-citizens and 6% of naturalised citizens.1 Even when eligible for services, non-citizens in particular may be fearful of accessing needed healthcare due to fear of public charge,13 barriers to language access1 and mistrust of the healthcare system.14

Although immigration status profoundly impacts various health inequities, few studies have examined how past undocumented status may influence their health. The conceptualisation of how immigration status affects health has been limited with extant research examining immigration status as a binary (legal/‘illegal’) and static variable.15 However, there is increasing recognition that legal status can change across the life of an immigrant,16 17 and that the implications of legal status may have lasting influences across immigrants’ life course—even after obtaining a permanent residence status (ie, lawful permanent resident) or becoming a US citizen through naturalisation.17 The examination of the recipients of the Deferred Action for Childhood Arrivals (DACA) Programme is perhaps the only body of literature we have that focuses on immigrants who were previously undocumented who gain a more protected status. Studies find that transitioning to a more protected status improves psychological distress,18 labour market and socioeconomic outcomes19 20; however, given that the status is temporary, there continues to be uncertainties about immigration enforcement and financial stress.21–23 While DACA points to potential ways in which changing legal status may impact health, given the temporary nature of the programme, it is critical to take into account the role of past undocumented status in health across immigration statuses including those who are currently non-citizens or naturalised citizens.

To examine the persistent effects of legal status, this study uses population-based data on foreign-born adult Latinos and Asians in California to examine how past undocumented and current legal status is associated with healthcare access (eg, having a usual source of care, delays in care) and health outcomes (eg, psychological distress). We hypothesise that citizens who always had lawful status will report greater healthcare access, including usual source of care and less delays in care, and lower psychological distress compared with non-citizens and those who were previously undocumented.

Methods

Study design and sample

The authors collected population-based data as part of the RIGHTS Study to understand the experiences that Asian and Latinx immigrants in California have as they seek healthcare and engage in their communities. The RIGHTS Study was a follow-up survey to the 2018, 2019 and 2020 California Health Interview Survey (CHIS), the largest state-level, representative, population health survey in the USA. CHIS is a mixed-mode (telephone and web) survey of a representative sample of people using an address-based and phone-based (including cell phones) sampling frames in California and is administered in multiple languages including English, Spanish, Chinese (Mandarin and Cantonese), Vietnamese and Korean. CHIS collects information about demographics, immigration, healthcare access and health status.24 Approximately 1–3 months after completing CHIS, respondents were recruited to complete the RIGHTS survey if they were age 18 years or over and were foreign-born from a country of Latin America (including the Caribbean) or Asia (including South Asia and Southeast Asia but excluding the Middle East). The RIGHTS survey was administered by phone. 6974 participants were eligible based on CHIS, of which 5959 agreed to be recontacted with a final sample of 2010 participants. Non-response was due to not agreeing to be recontacted or not picking up the phone. There were no missing responses for the variables analysed. The sample size was powered to assess differences in healthcare among the study population. Weights were calculated to account for CHIS and RIGHTS follow-up survey complex survey designs and to produce population-based estimates. In addition, study data were protected under a National Institutes of Health Certificate of Confidentiality which protects participants’ privacy by prohibiting disclosure of identifiable or sensitive information.25

Patient and public involvement

The study included a Community Advisory Board (CAB) of 15 individuals who were recruited because of their direct work with immigrant community members and immigrant-serving organisations. During survey development, the CAB informed the survey measures and recruitment materials. During data collection and analysis, CAB members met twice each year of the study to ensure that study materials reflected community perspectives, reviewed and contextualised study results, and ensured information was disseminated to community and key stakeholders.

Study measures

Healthcare access and outcomes

Outcome measures are previously used measures from CHIS. Usual source of care was measured by the question: ‘Is there a place that you usually go to when you are sick or need advice about your health?’ and a follow-up question if participants responded ‘yes’: ‘What kind of place do you go to most often—a doctor’s office, a clinic or hospital clinic, an emergency room or some other place?’ Those who reported going to a doctor’s office/other health maintenance organisations or clinic/health centre/hospital clinic were coded as having a usual source of care. Those who responded ‘no’ to the former question or those who reported going to the emergency room or having no one place for care were coded as not having a usual source of care.

Delayed medical care in the past 12 months was measured by the question, ‘During the past 12 months, did you delay or not get any other medical care you felt you needed—such as seeing a doctor, a specialist or other health professionals?’

Psychological distress was measured by the validated Kessler K6 non-specific distress scale to identify severe mental illness (primarily depression and anxiety) which includes six survey items that are calculated into a measure of moderate mental distress (using threshold of K6 ≥5).26 We used the score as a binary variable to assess mental distress at a moderate and above, yet clinically relevant, level. This approach has been previously used and validated with CHIS data.26

Past and current legal status

Participants were asked to report their current citizenship status (citizen or non-citizen). After a careful piloting process, the RIGHTS Study developed measures of past undocumented status using the question, ‘While living in the USA, was there ever a time you did not have a valid visa or other documents which permitted you to stay in the USA?’ A response of ‘yes’ was coded as previously undocumented and a response of ‘no’ was coded as always held lawful status. Using these two questions, we constructed a categorical variable that combined current citizenship and past undocumented status into four groups: citizens who have always held lawful status (CLS); citizens who were previously undocumented (CU); non-citizens who have always held lawful status (NLS); and non-citizens who were previously undocumented (NU).

Covariates included sociodemographic characteristics that were associated with the outcomes of interest: race/ethnicity (Latinx or Asian), age (continuous), gender (male or female), education level (high school and above or below high school), employment status (employed, unemployed or not in labour market), income level (≥200% federal poverty level or <200% federal poverty level), cohabiting (married/partnered or not), language of interview (English or non-English language), health insurance (private, public or no insurance) and self-rated health (good or poor).

Analyses

We examined univariate statistics of the outcomes, legal status and sociodemographic characteristics. We also examined differences in outcomes and sociodemographic variables by legal status categories. We used logistic regression models to assess associations between legal status and usual source of care and delayed medical care in the past 12 months, adjusting for covariates. We used a linear regression model to assess the association between legal status and psychological distress as a continuous variable, controlling for covariates. For each model, we used post-estimate pairwise comparisons to assess differences between all levels of legal status. All analyses were weighted to produce population-based estimates. We conducted sensitivity analyses using a binary variable of severe psychological distress (cut-off of 13+ vs not) and a continuous variable (range 0–24). No significant results were found between past and current legal status and this outcome, controlling for covariates, for the binary severe psychological distress measure. However, similar results were found for the continuous measure (results not shown).

Results

Sociodemographic characteristics

Table 1 shows the sample characteristics of immigrants in 2018–2020 RIGHTS Survey by current citizenship status and previously undocumented (N=2010). Overall, about 36.8% of participants were CLS, 10.9% were CU, 37.1% were NLS and 15.3% were NU.

Table 1
|
Sample characteristics of immigrants in 2018–2020 RIGHTS Study, by citizenship status and previously undocumented (N=2010)

Healthcare access and outcomes

In regard to different health outcomes, 79.7% of participants had a usual source of care, 15.6% had delayed medical care, and 14.4% had moderate and above distress. Among CLS, 91.8% of them had usual source of care, 11.9% had delayed medical care, and 11.2% had moderate and above distress. Among CU, 85.2% of them had usual source of care, 19.7% had delayed medical care, and 10% had moderate and above distress. Among NLS, 70.8% of them had usual source of care, 15.7% had delayed medical care, and 19.5% had moderate and above distress. Among NU, 68.0% of them had usual source of care, 21.2% had delayed medical care, and 13.1% had moderate and above distress.

Multivariate associations between past and current legal status with health outcomes

We assessed the associations between participants’ past and current legal status and their usual source of care (table 2). NU were less likely to have a usual source of care (OR=0.48, p<0.05) compared with CLS, controlling for covariates. NLS were less likely to have usual source of care (OR=0.40, p<0.01) compared with CLS, controlling for covariates. Being older, being female and cohabiting with others were associated with increased odds of having a usual source of care. Having no insurance was associated with decreased odds of having a usual source of care compared with those having private insurance. No other covariates were associated.

Table 2
|
Unadjusted and adjusted odds of reporting usual source of care by citizenship status and previously undocumented, RIGHTS Study 2018–2020 (N=2010)

We assessed the associations between participants’ past and current legal status and whether they delayed medical care in the past 12 months. In the logistic regression model (table 3), CU were more likely to delay medical care (OR=2.34, p<0.05) compared with CLS, controlling for covariates. Being younger, being female, living below 200% federal poverty level, receiving the interview in English and reporting poor self-rated health were associated with increased odds of delaying medical care. No other covariates were associated.

Table 3
|
Unadjusted and adjusted odds of delaying medical care in past 12 months by citizenship status and previously undocumented, RIGHTS Study 2018–2020 (N=2010)

We also assessed the associations between participants’ past and current legal status and their psychological distress. In the logistic regression model (table 4), NLS were more likely to have moderate and above distress (OR=2.31, p<0.01) compared with CLS, controlling for covariates. Being a female, not cohabiting with others, receiving the interview in English and reporting poor self-rated health were associated with increased odds of having moderate and above distress. No other covariates were associated.

Table 4
|
Unadjusted and adjusted odds of moderate and above psychological distress by citizenship status and previously undocumented, RIGHTS Study 2018–2020 (N=2010)

Discussion

This study found that there were persistent effects of legal status on health. We found non-citizens were less likely to have a usual source of care regardless of a past undocumented status; moreover, CU were more likely than their citizen counterparts without such history to have delayed care.

Overall, this research not only reinforces the importance of current legal status but points to the potential ways in which being previously undocumented may influence health. This research provides evidence of the ways in which legal status trajectories—or the processes of living through, adjusting and changing statuses—may influence different facets of well-being. For some immigrants, a more permanent status does not erase uncertainty and vulnerability to immigration enforcement nor persistent financial and social stressors. Our findings likely reflect continual challenges faced by such immigrants in accessing care and benefits of being a citizen that are critical to healthcare access.27 28 These challenges may be due to difficulties of obtaining affordable health insurance without subsidies or other systemic barriers that prevent non-citizens from tapping into broader sets of social resources. Differences may also be due to continued influence of past experiences that limit non-citizen immigrants’ ability to navigate the healthcare delivery system or limit linguistic or cultural competency of providers who can provide care to them.29 30

Our findings demonstrate that being CU was associated with delays in care. This corroborates other studies that have found that those who are undocumented were less likely to access care and experience delays in needed care.7 31 32 For example, undocumented immigrants may experience economic and occupational threats and violations which result in sustained isolation and limited social networks.33 34 Similar differences were not found among non-citizens, suggesting that other predictors of delay such as discrimination, length of undocumented status or policy context should be considered.

The study also found that among those who always had lawful status, non-citizens had a higher psychological distress score compared with citizens. This is corroborated by substantial literature that citizenship, not just documentation status, confers a number of benefits, including freedom from fear of deportation and the associated stress and uncertainty that result from criminalising policies.11 12 35 Future studies should explore the mechanisms for why and how changes in legal status may be associated with improved mental health.

There are a number of limitations to the study. First, this study uses cross-sectional data that do not allow for analyses across time. While we examined bivariate analyses with years in the USA to account for time in the USA, we left this out in the final models because it was highly correlated with age of the sample. With cross-sectional data, we are unable to examine causal associations or how immigration status may impact health over time. Our measures of undocumented status and history of statuses are also limited. We are unable to discern how long individuals were undocumented and therefore unable to examine the cumulative impacts of undocumented status on health. Relatedly, the category of NU may include both those who are currently documented or undocumented; therefore, other studies should include a more robust sample of those who are currently undocumented. Given that our data were collected through self-reported surveys, our outcome measures of interest may be prone to recall bias. Additionally, while we use population-based data, our findings were limited to California and may not represent states with more restrictive immigrant environments. Lastly, the respondents in the RIGHTS sample may differ from non-respondents from the overall CHIS sample by under-representing the previously undocumented group. However, among the immigrants in the CHIS sample, there were 59.2% naturalised citizens vs 40.8% non-citizens, while the RIGHTS sample included 49% naturalised citizens vs 51% non-citizens. This suggests that it is unlikely that those who were previously undocumented are under-represented in this sample. One strength of this study is that it captures a sizeable enough population to capture the diversity of current and previous legal status experiences.

Our findings highlight the importance of re-examining restrictive immigrant policies or systemic barriers that may deter eligible immigrants to access benefits. Efforts such as disseminating knowledge of available benefits and increased outreach to immigrants with experiences of previously being undocumented who may continue to harbour fears or distrust of authorities could promote access and subsequent health of these groups. Our findings also indicate the need for further research to highlight the cumulative burden of trauma and disadvantage among those with a history of being undocumented and how such experiences may impact their future care-seeking behaviours and interactions with the healthcare system.

Article metrics
Altmetric data not available for this article.
Dimensionsopen-url