A Comparative Study of Two Welfare Regimes within the U.S. with Respect to Elderly Immigrants' Primary Healthcare Service Use Behaviors

A comparative study, “a substitute for experimentation in the natural sciences” (Mabbett & Bolderson, 1999, p. 35), allows for identifying and gaining a better understanding of different structures and institutions (Jones, 1985). To examine elderly immigrants (EIs)’ use of primary healthcare services (PHS) in light of Andersen’s equitable healthcare access model (1995), this study compared the pre-welfare reform era and post-welfare reform era. Andersen defined “equitable access” as occurring when demographic characteristics and healthcare need variables account for most of the variance in utilization; “inequitable access” is when social structures and enabling resources determine who receives medical care.

There is a growing concern about EIs’ health given that EIs living as green-card holders are limited in securing health insurance because (1) companies do not sell policies to them and (2) welfare reform restricts them from obtaining health benefits including Medicaid, unless they pass the citizenship test in English or live in immigrant-friendly states (Choi, 2012; Greene, 2011). However, 71% of EIs who entered the U.S. in 2006 had little or no English proficiency (Leach, 2009). The Affordable Care Act also excludes them from the universal health coverage plan (Aliferis, 2013). Before welfare reform, however, the U.S. granted legal EIs eligibility for the same health benefits as citizens (Fix & Laglagaron, 2002).

A logical assumption is that EIs’ healthcare service use behaviors changed after welfare reform. However, few studies examined the changes.

Methods. Data on immigrants aged 65 or older from the National Health Interview Survey (1992-1996 for pre-welfare reform [N=40,479]) and 2002-2008 for post-welfare reform [N=65,995]) were analyzed, employing multilevel models with logit link function. Guided by multilevel analysis methods for studies of individuals and cultures (see Van de Vijver, Van Hemert, & Poortinga, 2008), variations in service use among racial/ethnic groups were calculated at level-2 while variations among states were calculated at level-3. The PHS variable (=DV) asks whether respondents visited doctors during the past 2 weeks.

Results. Before welfare reform, being insured (OR=3.82, p<.001) and being ages 75 or older (OR=1.27, p<.05, compared to 65-69) significantly explained EIs’ PHS use.

However, after welfare reform, the indicators for PHS use are:

• being insured (OR=2.72, p<.001)
• living 15 or more years in the U.S. (OR=1.74, p<.01, compared to <5)
• having at least some college education (OR=1.35, p<.001, compared to <high school).
• Moreover, explained variance in PHS use among race/ethnicity was not significant before welfare reform (variance=0.046); however, it became doubled and significant after welfare reform (variance=0.086, p<.01).

Conclusion: The results indicate that healthcare use behaviors before welfare reform was more equitable compared to the post-reform era. According to Andersen (1995), inequitable healthcare access indicators were pronounced after welfare reform: both social structure (education, race/ethnicity, duration of residence) and enabling resources (health insurance) explained EIs’ PHS use. It is important to note that ‘15 years or more residents’ immigrated to the U.S. before welfare reform (they were not affected by welfare reform). Hence, the duration of residence variable reflects “eligibility for public benefits” as well as “acculturation.”