Date of Award


Document Type

Open Access Dissertation


College of Social Work


Social Work

First Advisor

Miriam M Johnson


Background: The intended result of the Personal Responsibility and Work Opportunity Reconciliation Act of 1996 (Public Law 104-193: PRWORA) was to conserve public funds while addressing welfare deficits. To achieve this end, the PRWORA (1) requires immigrants who came to the United States after the law took effect to show proof of U.S. citizenship to be eligible for federally funded public benefits, including Medicaid, unless the states where they lived provide state-funded benefits; (2) reinforced the `public charge law,' indicating that once the Immigration and Nationalization Service deems a post-welfare reform immigrant a public charge, this could result in the denial of a green card, denial of readmission to the United States after a trip abroad, or in deportation (Buff, 2008; Kandula et al., 2004); and (3) deems the income and resources of the sponsors of immigrants to judge the immigrant's eligibility for public assistance including Medicaid, which is considered `a unique obstacle' to immigrants to be judged as being poor (Dordeski & Steffens, 2010). As a result, to be eligible for public assistance `Public Law 104-193' and `Affidavit of Support Under Section 213A of I-864' require elderly immigrants (a) becoming U.S. citizens or (b) completing 40 quarters (i.e., 10 years' worth) of work requirement unless they live in a state which provides state-funded assistance.

On the other hand, some factors related to elderly immigrants raise some concerns about elderly immigrants' health. The factors may include that (1) most commonly, elderly immigrants enter the United States at 60-79 years of age through the family reunification program at the invitation of their naturalized adult children, implying that these immigrants have little or no U.S. work history and that they are more likely than their U.S.-born counterparts to live in poverty (Leach, 2009); (2) elderly immigrants, in particular, those from non-English speaking countries, are limited in learning a second language at their later age to pass the citizenship test which asks their listening, writing, and speaking ability in English: in 2006 almost 71 percent of newly arrived elderly immigrants had little or no English proficiency (Leach, 2009) and among the total elderly immigrants, as much as 56 percent reported limited command of English in 2010 (Batalova, 2012); and (3) they are more likely than their U.S.-born counterparts to rely on Medicaid due to (a) their socioeconomic status (Batalova, 2012; Leach, 2009), (b) limitation in their access to Medicare due to their lack of the required work history (Quadagno, 2005; Nam, 2011), (c) limitation in their access to employment-based private health insurance due to age-based discrimination in the labor market (Quadagno 2005; Nam, 2011), (d) limitation in working adults' access to direct-purchase health plans covering their older adults without Medicare coverage (Yelowitz, 2000; Choi, 2009), and (e) partly, little market for that age group due to the almost universal Medicare coverage among older adults - if anything, they often do not cover preexisting conditions (Choi, 2009)

Consequently, the PRWORA's citizenship requirement for access to public assistance has caused great concern about the health of elderly immigrants, considering that the number of foreign-born adults 65 years or older in 2010 had doubled since 1990, from 2.7 million to 5 million, accounting for about 12 percent of the 40 million foreign-born population and 12 percent of the 40.4 million older adults in the United States (Batalova & Lee, 2012). However, there has been a paucity of studies on the effects of the PRWORA on elderly immigrants' use of healthcare services and their health.

Objectives: By comparing two different U.S. welfare regimes (i.e., the pre-PRWORA era and post-PRWORA era), the present study examined the impact of PRWORA (1) on elderly immigrants' healthcare service utilization by adopting Andersen's Behavioral Model of Health Services Use (Andersen, 1968, 1995, 2008) and (2) on their health by testing "healthy immigrant effect" theories applicable to elderly immigrants in the United States. The study population is defined as immigrants aged 65 or older who were in the United States at the time of the interview.

Methods: To analyze data from the 1993-1996 (for the pre-PRWORA) and 2002-2008 (for the post-PRWORA) National Health Interview Survey collected by the Centers for Disease Control and Prevention, the present study used multilevel random intercept models with logit link function to address violation of independent observations within states with similar adaptation of PRWORA and within a racial/ethnic group sharing similar health beliefs, culture, and language. The multilevel models were estimated by adopting Markov Chain Monte Carlo (MCMC) method in MLwiN 2.20, which allows Bayesian models to be fitted. The MCMC is the best with discrete response models: it has no requirement of normality assumption in making inferences for variance parameters (Browne, 2009). The dependent variables used to test elderly immigrants' healthcare service use behaviors are (1) "doctor visits during the past 2 weeks" as a discretionary behavior and (2) "short-stay hospital use during the past 12 months" as a non-discretionary behavior. To test elderly immigrant's health, the dependent variables are (1) "self-assessed health status" and (2) "activity limitation status due to one or more chronic diseases." They are all binary variables.

Results: Discretionary Healthcare Service Use: During the pre-PRWORA period, only age and health status significantly explained discretionary service use behavior. However, during the post-PRWORA period, education, citizenship, and the length of residence in the U.S. became important indicators in elderly immigrants' discretionary service use behavior. In addition, before the PRWORA, racial/ethnic minority groups' use of discretionary healthcare service use was not significantly different from that of non-Hispanic whites. However, after the PRWORA, all racial/ethnic groups but Cubans were much less likely than non-Hispanic whites to use discretionary healthcare services.

Non-discretionary Healthcare Service Use: Elderly immigrants' health status and age substantially and significantly explained their non-discretionary health service use both before and after PRWORA. However, during the post-PRWORA era, race/ethnicity and health insurance coverage status became significant contributors, in addition to educational achievement level. Health Status: As Jasso, Massey, and Rosenzweig (2004) postulated, the findings show a `reversed healthy immigrant effect' before welfare reform: newly arrived immigrants were more likely than immigrants who had lived longer in the United States to report poor health and activity limitation due to chronic disease, with the best perception of good health and the least reports of activity limitation due to chronic disease among those with 15 or more year residents. The opposite pattern in the period after welfare reform was observed: newcomers were more likely than immigrants who had lived longer in the United States to report good health, with the worst perception of poor health and the most reports of activity limitation due to chronic disease among those 5- to 14-year residents.

Conclusions: Prominent Inequitable Access Indicators in Use of Healthcare Services: Andersen (1968, 1995, 2008) postulated that individuals' healthcare service use that was explained by the social structure and enabling factor can be an indicator of `inequitable' access. This study documented that after the PRWORA, elderly immigrants' discretionary health service use behaviors are affected by social structural differences (= education, race/ethnicity, citizenship, and the length of residence) and enabling factor (= health insurance coverage). Based on Andersen's definition of inequitable access, the findings of this study may suggest that the implementation of the PRWORA has led to inequitable access in elderly immigrants' healthcare service use.

Greater Variations at the Culture Level than the State Level: The multilevel analysis made it possible to simultaneously incorporate both individual and group level models within the contexts of changing policies at the federal and state level. Increasingly greater variations at any level suggest the need for further investigation within that level (Carle, 2009). The present study found the variations at the culture level of which the variations were greater those of the state level both before and after PRWORA in elderly immigrants' use of healthcare services and their health. The variations at the culture level became statistically significant during the post-PRWORA era.

Increased Health Disparities among Racial/Ethnic Groups: Elderly immigrants' response to the PRWORA in relation to their self-assessed health and activity limitations differed depending on their racial/ethnic identity. After the PRWORA, when non-Hispanic whites were referenced, almost all racial and ethnic minority groups reported increased poor health. In addition, Chinese, Filipinos, APIs, and Mexican-Americans reported their activity limitation also increased after the PRWORA. The groups with the highest odds of activity limitations were Filipinos and APIs among Asian groups and Puerto Ricans and Mexican-Americans among Hispanic groups.

Marginalized Immigrant Cohort: 5-14 Year Residents in the United States: In line with many studies, the present study found a significant relationship of duration of residence to elderly immigrants' health (Angel et al., 2010; Choi, 2012; Gee, Kobayashi, & Prus, 2004; González et al., 2009; Kobayashi & Prus, 2012) and their healthcare service use (Leclere, Jensen, & Biddlecom, 1994). The duration of residence variable became a contributing factor in explaining elderly immigrants' health and their utilization of discretionary healthcare services during the post-PRWORA era. In particular, elderly immigrants with 5-14 years of residence in the United States had the worst perceptions of health and the most activity limitation due to chronic disease.

Healthy Immigrant Effect due to Restrictive Welfare Reform: In support of the postulation by Jasso et al. (2004), during the pre-PRWORA period when the United States allowed elderly immigrants and refugees to enjoy the same access to medical services as citizens, elderly immigrants' initial health on arrival was poor, but the disadvantage disappeared due to quality and availability of good healthcare. However, during the post-PRWORA period when elderly immigrants' access to healthcare services was limited, their initial health was good, but the advantage disappeared over time due to limited availability of healthcare while their demand for medical care will increase due to aging.

Implications: The present study casts doubt on the long term effectiveness of welfare reform's eligibility restrictions on immigrants to achieve the goals of reserving more public financing and addressing welfare deficits. This is because, after the PRWORA, of elderly immigrants' reported greatly increased perception of poor health and of activity limitation due to chronic conditions among those with 5-14 years of residence and among racial/ethnic minority groups. This is a trend that requires the U.S. society to pay increased costs for elderly immigrants' medical care by using Emergency Medicaid more frequently. Many entities including National Institutes of Health and federal agencies are trying to reduce the health disparities among population because access to the healthcare system is also an issue of social justice and moral obligation. Therefore, three considerations suggest that welfare reform requirements related to elderly immigrants need to be revoked: deteriorating health among these populations, expectations of increased costs for their medical care, and American values of fairness.

Included in

Social Work Commons