Erin R. Hamilton, UC Davis Jo Mhairi Hale, Max Planck Institute for Demographic Research Robin Savinar, UC Davis

Transcription

1 Immigrant Legal Status and Health: Legal Status Disparities in Chronic Conditions and Musculoskeletal Pain among Mexican-Born Farm Workers in the United States Erin R. Hamilton, UC Davis Jo Mhairi Hale, Max Planck Institute for Demographic Research Robin Savinar, UC Davis Thanks to the Western Center for Agricultural Health and Safety for funding the research. Jose Bedia

2 Immigrant legal status as a fundamental cause of health There is a strong expectation in the literature of a legal status disparity whereby the unauthorized have worse health than the authorized (Asad and Clair; Castañeda et al. 2015; Martinez et al. 2013;Torres andyoung 2016) Research shows that unauthorized immigrants are disadvantaged in terms of: Self-reported health Mental health Access to health care The reviews note a lack of research on immigrant legal status disparities in physical health Very few population health data sources with immigrant legal status measured We conducted a systematic review of the literature on this topic

3 11 studies on immigrant legal status and physical health random survey data Publication Data source Legal status Health outcomes Findings Bitler and Shi 2006 LAFANS (n=2500) Self report by Asthma elimination Diabetes Wallace et al 2013 Holmes and Marcelli 2012 De Trinidad Young and Pebley 2017 CHIS 2009 (n= 40,000) Boston Metro Immigration Health and Legal Status Survey (n=307) LA FANS Latino sample (n=1396) Self report by Asthma Unauth better elimination Heart disease Diabetes High blood pressure Self-report High C-reactive protein Unauth better Self report by Systolic blood pressure elimination Hypertension Unauth better

4 11 studies on immigrant legal status and physical health non-random data Publication Data source Legal status Health outcomes Findings Marshall et al Achkar et al Iten et al women in Ft. Worth, TX Medical records of active TB patients, NYC hospital, (n=194) 401 Mexican diabetes patients in SF and Chicago Self report by elimination Notes by doctors or social workers Self report by elimination Backache Vision problems Flu/colds Allergies High blood pressure HIV positive Fever Night sweats Weight loss Cough Blood in cough Duration of symptoms Poor A1C control High blood pressure High low-density lipoprotein Comorbidities Unauth worse Unauth worse Unauth worse Unauth better

5 11 studies on immigrant legal status and physical health administrative data Publication Data source Legal status Health outcomes Findings Kelaher and Jessop 2002 Flores et al 2012 Wen and Maloney 2014 Swartz et al Hospital intake data, NYC, (n=4,975) UT birth records (n=196,000) UT driver licenses data (n=740,000) Insurance claims for births in Oregon, (n=213,000) No resident or SS card No SSN Use of an ITIN on application Use of Emergency Medicaid History of low birthweight Preterm birth Low birthweight Small for gestational age Overweight women Overweight men Low birth weight Very low birth weight Ext. low birth weight Preterm birth Infant mortality Unauth worse Unauth better Unauth better Unauth better Unauth better Unauth better Unauth better

6 Summary of existing literature on immigrant legal status disparities in physical health outcomes 11 studies using different data sources, different means for identifying unauthorized immigrants, and different health outcomes 36 total comparisons on 27 different physical health outcomes On 4 unauthorized immigrants fared worse than their authorized counterparts On 9 the unauthorized fared better On 23 there was no difference Why is there no support for the fundamental cause perspective in the existing empirical literature on immigrant legal status disparities in physical health?

7 Three possible reasons why the existing literature does not find legal status disparities in physical health: 1. Study limitations: a. Immigrant legal status is not directly measured b. Many of the studies use non-random samples; none use nationally representative data c. Very few of the studies control for important confounders of the relationship between immigrant legal status and health: time in the United States 2. There is no disparity: a. Unauthorized immigrants are (more) positively selected on health than authorized immigrants 3. Models need to consider the life course (Torres and Young 2016)

8 What we do in this study Examine legal status disparities in two physical health outcomes: the prevalence of lifetime, doctor-diagnosed health conditions reports of recent experiences of musculoskeletal pain In data on 18,666 Mexican-born farm workers in the NAWS Control for time spent in the United States and access to health care Attempt to account for selectivity Control for period and time of migration to test for role of selectivity into migration flow Use inverse probability weights to account for selection on observables Interact legal status with gender & indigenous status

9 Study details: Measures Legal status: respondents asked directly, What is your current legal status in the United States? Citizen (5% of our analytic sample) Legal Permanent Resident (31%) Temporary Visa Holder (1%) Unauthorized (63%) Health Chronic conditions: Has a doctor or other health professional ever told you that you have... asthma, heart disease, diabetes, or high blood pressure? Pain: In the past twelve months have you had pain or discomfort in your back, shoulder/neck, elbow/arm, hand/wrist/finger, legs/feet/toes, or other?

10 Study details: Measures Time in U.S. 0-4 years, 5-9, 10+; Also: English language ability & property ownership in U.S. Controls Health insurance coverage Age, gender, indigenous, region in US, follow the crop, type of crop, education, income, year of interview

11 Results: Chronic conditions and MSK Pain of unauthorized and authorized farm workers Chronic Conditions Musculoskeletal Pain Predicted Probability OR =.54*** OR =.64*** OR =.80** OR =.89 Base Full Base Full Unauthorized Authorized Predicted probabilities from logistic regression models of chronic conditions & pain; base model controls for age and year of survey; full model has all controls

12 Results: Chronic conditions and MSK Pain of unauthorized, temporary visa holder, LPR, and citizens farm workers Chronic Conditions Musculoskeletal Pain OR =.72*** OR =.40*** OR =.66*** OR =.59*** Citizen LPR Temp Unauthorized Predicted probabilities from logistic regression models of chronic conditions & pain (full models)

13 Results: Chronic Conditions and MSK Pain of unauthorized and authorized women and men Chronic Conditions Musculoskeletal Pain Predicted Probability Men Women Men Women Unauthorized Authorized Predicted probabilities from logistic regression models of chronic conditions & pain (full models). Gender is interacted with authorization status.

14 Results: Chronic conditions and MSK pain of authorized and unauthorized indigenous and non-indigenous farm workers Chronic Conditions Musculoskeletal Pain Non-Indigenous Indigenous Non-Indigenous Indigenous Unauthorized Authorized Predicted probabilities from logistic regression models of chronic conditions & pain (full models). Indigenous status is interacted with authorization status.

15 Summary of findings Unauthorized Mexican farm workers report fewer chronic conditions and a lower incidence of pain than authorized Mexican farm workers, on average This is net of measured socioeconomic, geographic, and other differences This is true for men and women, indigenous and non-indigenous farm workers There appears to be a legal status gradient: citizens are the worst off, followed by LPRs, and the unauthorized Test for selectivity (controlling for time, region of origin) made no difference to general pattern

16 Three causes for concern specific to our population 1. Differences in doctor-diagnosis may be confounded by differences in access to health care We controlled for health insurance coverage 2. Health problems may be under-reported NAWS is collected on the worksite Multiple studies report that undocumented farm workers fear reprisal from employers or field managers (Flynn, Eggerth, and Jacobson 2015; Lee et al. 2014; Leigh, McCurdy, and Schenker 2001; Liebman et al. 2016) 3. Negative selectivity on who remains in farm work However, our results are consistent with 11 studies using different data sources and measures of health

17 If there is no disparity, why not? A second epidemiologic paradox The first refers to the good health of immigrants/hispanic immigrants compared to non-immigrants This one refers to the good health of unauthorized immigrants compared to authorized immigrants Positive health selectivity into unauthorized status The larger costs and risks of unauthorized entry/residence means that more fit, more ambitious, more risk-taking, and more self-efficacious people do it The disadvantages of being unauthorized unfold over the longterm/across generations (Torres and Young 2016) More research, better data are needed

18 Thanks for your attention! Questions or comments? Jose Bedia