1. Left out under Federal Health Reform: Undocumented immigrant adults excluded from ACA Medicaid expansions Jessie KemmickPintor, MPH Graduate Research Assistant State Health Access Data Assistance Center University of Minnesota, School of Public Health 17th Annual NRSA Trainees Research Conference Seattle, WA June 11th, 2011 Funded by a grant from the Robert Wood Johnson Foundation
2. Acknowledgments SHADAC Co-Authors Sharon Long Senior Health Economist Lynn Blewett Professor, Director – SHADAC Michel Boudreaux Doctoral Student, RA Peter Graven Doctoral Student, RA 2
5. Medicaid Expansion in ACA New mandatory eligibility group for low-income individuals – 2014 implementation date Includes all persons with family incomes up to 133% (effective 138%) of the FPL who are not: Age 65 and older Eligible for Part A Medicare or enrolled in Part B Legal residents who have resided in the U.S. < 5 years Undocumented immigrants or legal immigrants subject to 5-year ban 4 Newly covered include children 6-19, parents of covered children, childless adults
6. What does national health reform mean for undocumented immigrants? Continuation of Medicaid/CHIP exclusion Exceptions: Emergency services for income-eligible undocumented immigrants States have option of covering prenatal care for pregnant women including undocumented under CHIP Exemption from individual mandate Cannot participate in new federal or state health insurance exchanges 5
7. Who are undocumented immigrants? Individuals who enter the country without approval by immigration authorities Individuals who violate the terms of a temporary admission i.e. overstaying tourist/student visa without adjusting immigration status Source: Congressional Budget Office. (2007). The Impact of Unauthorized Immigrants on the Budgets of State and Local Governments. Washington DC. 6
8. How many undocumented immigrants reside in the U.S.? (1) Lack of data No direct estimates as immigrant status not asked in national surveys Need for estimation methods to fill gaps in data Few sources of indirect estimates Jeff Passel & colleagues at Pew Hispanic Center Dept. of Homeland Security (Hoefer et al.) Census Bureau (estimates not available) Sensitivity of data that is available 7
9. How many undocumented immigrants reside in the U.S.? (2) 11.2 million people in 2010 1 million children 4% of total U.S. population 5.2% of workforce 8 million workers Source: Passel & Cohn. (2011). Unauthorized immigrant population: National and state trends, 2010. Washington, DC: Pew Hispanic Center. 8
10. Why should we care about immigrants’ access to care? High rates of uninsurance A substantial number of immigrants are excluded from ACA expansions Access to coverage and care likely to worsen under ACA Remaining uninsured immigrants are likely to continue to seek care at CHCs and public hospitals 9
11. 10 Research objectives Using 2008 American Community Survey (ACS) data, we estimate: The number of low-income (FPG<=138%) immigrants excluded from 2014 Medicaid expansions (undocumented immigrants) Characteristics of excluded immigrants Distribution of excluded immigrants across states
12. Data and Approach 2008 American Community Survey (ACS) Large national survey with state-representative samples Provides data on demographic, socioeconomic, and geographic characteristics of U.S. population We restrict sample to civilian/non-institutionalized, low-income (<=138% FPG), non-elderly adults (18-64) (N=280,130) 11
13. Immigration Status ACS asks about citizenship, country of origin, and years in the U.S., but does not ask immigration status SHADAC assigns immigration status to non-citizens whose legal status is unknown 12
14. U.S. Population by Citizenship Status 13 Source: ACS, 2008. *Non-citizens include legal, non-, and undocumented immigrants
19. are likely legal under the Temporary Protected Status program, given country of birth and year of entry into the US Remainder of sample assigned “unknown” legal status Source: Passel, J. (2006). The Size and Characteristics of the Unauthorized Migrant Population in the U.S. Estimates Based on the March 2005 Current Population Survey. Washington DC: Pew Hispanic Center. 14
20. Three-Step Method for Assigning Immigration Status Predict legal vs. undocumented status for those in ACS using external data source Predictive model from Urban Institute based on 2004 Current Population Survey file that includes estimates of legal status based on PHC framework Use model coefficients to predict probability of legal status for ACS sample Benchmark to PHC results for share of undocumented in each state 15
21. Preliminary Findings How many non-elderly adults are estimated to be eligible for Medicaid under 2014 income rules? 33.6 million low-income non-elderly adults How many of those low-income adults are likely to be undocumented immigrants? 11% or 3.7 million 16
23. 18 Number of low-income undocumented adults by state
24. 19 Proportion of low-income adults who are undocumented withineach state
25. Limitations Preliminaryestimates Do not address the 5-year waiting period for Medicaid eligibility for legal residents Assignment of legal foreign-born population likely includes legal residents who have resided in U.S. for less than 5 years Our estimate is conservative But legal residents will be able to purchase through exchanges 20
26. Conclusions First estimates of the number of low-income (per FPG) non-elderly adults excluded from 2014 Medicaid expansions because of their immigration status A substantial number of income-eligible undocumented immigrants will not be covered Up to an estimated 3.7 million adults In three states, more than 1 in 5 low-income non-elderly adults will not be eligible 21
27. Policy Implications Need for safety net care will not be evenly distributed across states States with disproportionate number: CA, TX States with disproportionate share: CA, AZ, NV Understanding the likely scope of the population without coverage will help states and safety-net providers prepare for the safety net under reform. Implications for states will depend on the share of undocumented immigrants among low-income adults and capacity of the safety-net within states. 22
28. Next steps Currently working with demographer to update model Further examine health care access, use, and expenditures of undocumented immigrants Assess capacity of safety-net clinics/providers to fill this gap 23
First of all, I’d like to acknowledge my SHADAC co-authors
To start off, I wanted to frame our project within the larger picture of health reform – the policies that have been implemented so far and what we can expect. In the past year a few of the major components that have been implemented: Small employer (less than 25 employees) tax creditsEarly Medicaid: giving states option of cover more on Medicaid (federal matching funds)Early retiree reinsurance program: preserving employer coverage for gap between retiring and reaching 65 when eligible for Medicare (until exchanges)Pre-existing condition insurance plansCovering dependents up to age 26 regardless of school statusThen on the horizon in 2014 we see some of the more substantial policies for reducing uninsurance: establishing state and federal health insurance exchanges and the individual mandateAnd finally the focus of our presentation today: Medicaid expansions
+ 5% income disregard Individuals who are income-eligible for Medicaid under 2014 income rules will be referred to as low-income adults here
Why should we care about the impact of health reform on undocumented immigrants?As was true prior to ACA, not eligible for federal funding under Medicaid or CHIP; some states have covered with state dollarsEmergency services provisions may now cover more undocumented since income-eligibility standard has increasedStates continue to have option of covering undocumented pregnant women. As of 2009, 17 states do so (Kaiser Commission on Medicaid and the Uninsured)Exemptions will be granted for financial hardship, religious objections, American Indians, those without coverage for less than three months, incarcerated individuals, those for whom the lowest cost plan option exceeds 8% of an individual’s income, and those with incomes below the tax filing threshold, undocumented immigrants Cannot participate in exchange even if using their own dollarsLimitation on participation in exchanges may affect those who purchase coverage on their own as well as those who work for small firms that enter the exchange
We typically hear more about the first case when in fact of all the undoc immigrants currently residing in the U.S. about ½ entered the country each way
So we know that any discussion of immigrants, particularly undocumented immigrants, is highly contentious, and both the consequence and cause of this political contentiousness has been a dearth of data on the number of and characteristics of undocumented immigrantswe don’t ask about immigrant status (for both practical and ethical reasons), and the data that is available is highly sensitive – so for example there is one national survey (SIPP) and one state survey (CHIS) that does ask about legal status, but not clear how reliable this question is due to nonresponse and social desirability biasPrecarious state of undocumented immigrantsBUT despite this political sensitivity, policymakers need this data and thus reliable estimation methods are necessary
1. PHC work developed by Jeff Passel and colleagues over many years. Uses variety of logical edits and administrative data sources (Dept. of Homeland Security, census) to project size of population. I’ll talk a bit more about this later.- There are a few larger surveys that do ask immigration status beyond citizenship (i.e. CHIS) and smaller studies distinguishing between undocumented and legal immigrantsOf the 11.2 million, most are adults: 10.2 million or 90%There are estimated 8 million workers (IRS estimates that 6-7 million file taxes)
So why should we care about their access to coverage? High rates of uninsuranceSubstantial number excluded from expansions and combined with undoc immigrants not being allowed to purchase through the exchange it is likely that uninsurance among immigrants will grow under ACA
Replaced decennial census long-formMandatory, 98% response rateIncludes those living in non-institutionalized group quarters (such as dormitories, group homes, migrant worker camps, etc.)
We know the immigration/citizenship status of 93% of the total U.S. population. In this analysis, we are assigning either legal status or undocumented status to the 7% of the U.S. population who are non-citizens
Builds off PHC framework-Jeff Passel’s work and Urban Institute. We use a residual approach that subtracts the estimated legal-immigrant population from the total foreign-born population and treats the residual as a source of data on the unauthorized migrant populationOccupations: public workers, police officers, lawyers, judges, magistrates, physicians, past/present military Approximately 94% of the sample is assigned to legal status based on these logical edits, 6% “unknown”If someone asks about 1980 rule: I would say that it is a rule that we borrowed from Passel and our assumption is that it's motivated by the fact that the longer you are here, the more likely you are to be legal. (this could be because the immigration laws prior to 1980 were more liberal or because the more chance you have to get a green card). We can get back to questioners about specific details at a later date.
10.9% of all low-income non-elderly adultsOr almost 40% of all undocumented adults
So once again when I’m referring to low-income adults, I’m referring to those at or below 138% of povertyRelative to low-income adults who are US citizens and those who are likely legal immigrants, among undocumented immigrants we see:-young working famliesYounger adultsMore likely married and parents—although not shown here, many have citizen children; Passel & the Pew Hispanic Center estimate that 73% of children of undocumented immigrants were born in the United States (are U.S. citizens)Twice the uninsurance rate, about 1/3 with coverage17% with private (either employer-sponsored or purchased on individual market)13% with public (likely pregnant women and immigrants in states like NY that offer coverage)All significant differences
The following 2 maps present our findings for our last question on the distribution of low-income undoc. Adults across the country: I want to note that this is only for 4 million of the 10 million adults, the maps would look different if we were looking at ALL undoc. adultsAs you can see, over ¼ of all low-incomeundoc. Adults live in CA and TX alone (with over 500,000 each)Map displays# of undocumented in the total income-eligible non-elderly population within states 6 states with 100-500,0008 states with 50-100,00022 states with 10-50,00013 states (including D.C.) with less than 10,000
This map really gets to our question, not just looking at raw numbers but what % of all low-income adults are undoc. w/in each state3 states with 20% +Texas--% not among the highest, but total number of people is13states with 10-19%15 states (including D.C.) with 5-9%20 states with less than 5%
So combined with their not being allowed to purchase through the exchange and being excluded from Medicaid, it is likely that undoc. Immigrants will continue to rely on the safety net for their health care needs AND
Whether this will present an issue for states will depend on the share of undoc. immigrants among low-income adults AND the capacity of safety-net clinics/providers within states