The Misconception of Hispanic Homogeneity: H1N1 Vaccination among US-born and Foreignborn

Transcription

1 The Misconception of Hispanic Homogeneity: H1N1 Vaccination among US-born and Foreignborn Hispanics Andrew E. Burger Eric N. Reither Erin T. Hofmann Total Word Count: 2,876 Utah State University Dept. Sociology, Social Work, and Anthropology 2016

2 ABSTRACT Background: The 2009 H1N1 influenza pandemic produced a substantial mortality and morbidity burden in the United States. As with seasonal influenza vaccine uptake, H1N1 vaccinations were reported to vary by race/ethnicity. Methods: We estimate H1N1 vaccination uptake and control for potential mediators of vaccination, such as education and income, among non-hispanic whites and Hispanics using data from the National Health Interview Survey (NHIS). Results: Our analyses show significant within-group differences among Hispanics, with foreign-born Hispanics experiencing the lowest rates of H1N1 vaccination (15.3%), followed by US-born Hispanics (17.9%) and non-hispanic whites (20.7%). By controlling for inequalities in SES and healthcare coverage, we find odds of vaccination among Hispanics similar, if not higher, to those of non-hispanic whites. Discussion: Our findings indicate disparities in income and education help explain low Hispanic vaccination in comparison to non-hispanic whites uptake during the 2009 H1N1 pandemic. Also, we find healthcare coverage a strong predictor of vaccination that appears to function outside traditional measures of SES. While efforts to reduce disparities in income and education must continue, immediate efforts to improve improving healthcare coverage may prove beneficial in preparation in increasing vaccination for a future influenza pandemic. Abstract word count: 193 2

3 BACKGROUND In 2009, the United States experienced its first pandemic influenza since the 1968 Hong Kong Flu (Tognotti 2009). With more than 60 million cases and an estimated 274,304 hospitalizations, the 2009 H1N1 influenza imposed a substantial disease burden in the United States (Shrestha et al. 2011). Because populations lack prior immune response during pandemic influenzas, developing vaccines are key to preventing infection but only if they are produced and delivered efficiently (Nichol and Treanor 2006). Whereas 40.4% of U.S. adults aged 18 and older received the seasonal flu vaccine, only 22.7% U.S. adults received the separately administered 2009 H1N1 vaccine (CDC 2011). This lower rate of pandemic vaccination is especially troubling among Hispanics, who experienced higher risks of H1N1 related complications such as hospitalization and mortality (Kwan-Gett, Baer and Duchin 2009; Miller et al. 2010; Ritger et al. 2009; Thompson et al. 2011). Excess morbidity and mortality among Hispanics were preventable, with several observational studies indicating that Hispanics experienced lower rates of H1N1 vaccination than non-hispanic whites (Quinn et al. 2011; Uscher-Pines, Maurer and Harris 2011). To prevent a recurrence of this in future pandemics, it is necessary to identify the determinants of low vaccination uptake among Hispanics so that the public health community can develop effective vaccination programs and strategies. The Misconception of Hispanic Homogeneity Often due to data limitations, self-identifying Hispanics are often grouped into a single ethnic category. Unfortunately, this practice likely produces a misconception of 3

4 Hispanic homogeneity, concealing the broad diversity within this quickly growing ethnic group (Umaña-Taylor and Fine 2001). Previous research notes significant differences among Hispanics by immigration status (Kochhar 2014), country of origin (Pabon-Nau et al. 2010), generation since-migration (Dustmann, Frattini and Lanzara 2012), and language (Wisnivesky et al. 2012). We believe separate analyses of foreign-born Hispanics are warranted, given, among others, significant differences in education and income (Suro and Passel 2003). We additionally recognize the diversity among foreignborn Hispanics, examining factors such as English language use, the length of residence in the USA, region of birth, and citizenship status. Disparities in flexible resources as a cause of low H1N1 vaccination uptake. Guiding our discussion on the causes of low vaccination uptake among Hispanics, we utilize Fundamental Cause Theory (FCT). FCT posits that deeply rooted societal inequalities are fundamental in the cause of disease as those with more resources are better able to seek health promoting services and procedures (Link and Phelan 1995). For example available healthcare services, such as the 2009 H1N1 vaccine, may be disproportionally effective since those with more flexible resources (i.e. economic, cultural, social, educational capital) can efficiently convert those resources to obtain services (Phelan and Link 2005). With the previously mentioned differences between US-born and foreign-born Hispanics, we expect disparities in flexible resources to be influential in predicting H1N1 vaccination. If disparities in flexible resources do account for differences in healthcare utilization, eliminating those disparities should produce an increase in utilization of healthcare services. We expect, for both US-born and foreign-born Hispanics, 4

5 controlling for income and education (i.e. flexible resources) will explain the lower odds of Hispanic H1N1 vaccination when compared to non-hispanic whites. Also, we expect foreign-born Hispanics to experience the greatest increase in odds of vaccination as we control for flexible resources, as they are likely to experience the greatest disadvantages. METHODS Data For our analysis, we use the 2010 National Health Interview Survey (NHIS). Designed to be representative of the US adult population, the CDC administers the NHIS annually (CDC 2010). The NHIS is advantageous for our study as it oversamples Hispanics and includes information on nativity status and other sociodemographic characteristics. Measures Dependent Variable. The NHIS records respondents 2009 monovalent H1N1 pandemic vaccination status. Since most H1N1 vaccinations occurred during the fall of 2009, using the 2010 NHIS reduces the risk of false negatives, where respondents not vaccinated at the time of the survey became vaccinated in the future (Burger and Reither 2014). Independent Variables. Our primary independent variables of interest, race/ethnicity and nativity status, are self-reported by the respondent. We limit our analysis to US-born non-hispanic whites, US-born Hispanics, and foreign-born Hispanicsvaccination status is known. In addition to common demographic variables, we include measures of flexible resources as described by FCT, including education and household income. As an added measure of economic resources, we include 5

6 homeownership status, which may be a better indicator of wealth than household income (Keene et al. 2014). A key variable of interest, healthcare coverage, is also self-reported by the respondent. We dichotomize healthcare coverage into those who have some coverage and those with no coverage. While the H1N1 vaccine was available to entire US population, individuals with pre-existing conditions were particularly at risk of H1N1 complications and were encouraged by the CDC to vaccinate. In order to account for the possibility that vaccination uptake varied because of differing proportions of those at risk of H1N1 complication within different racial/ethnic groups, we create a binary "at risk" variable which includes individuals who are pregnant, currently has asthma, ever been told they have diabetes, ever been told they have some form of heart disease (CHD, angina pectoris, heart attack, other), or told they have weak/failing kidneys during the last 12 months. Allowing us to make additional distinctions among foreign-born Hispanics, the NHIS includes self-reported measures of citizenship status, duration in the US, and the region of birth (dichotomized into Central America/South America). As a measure of English language ability, we include the NHIS field technicians' report of the language in which the respondent was interviewed, which we group into two categories, those who responded in English only (or a mixture of English and Spanish) and those who responded in Spanish only. ANALYSIS We estimate H1N1 vaccination uptake for US-born non-hispanic whites, USborn Hispanics, and foreign-born Hispanics. Also, we provide descriptive information on various sociodemographic variables which we include as measures of flexible resources. We then conduct a series of binary logistic regression models 6

7 to estimate adjusted odds ratios predicting H1N1 vaccination. As we examine H1N1 uptake specifically among foreign-born Hispanics, we present additional descriptive characteristics unique to migrants. Afterwards, we perform a series of logistic regression analyses estimating vaccination specifically among foreign-born Hispanics including migrant-specific characteristics. These analyses allow us to see how being foreign-born might influence the flexible resources in predicting odds of H1N1 vaccination. We perform all analyses using STATA 13 (StataCorp 2013a), taking into account the complex sampling design of the NHIS. RESULTS As seen in Table 1, US-born non-hispanic whites report the highest level of H1N1 vaccination (20.7%) among all groups, statistically higher (p < 0.05 level, onetailed) than either US-born Hispanics or foreign-born Hispanics. While the difference in H1N1 vaccination between US-born Hispanics and foreign-born Hispanics is modest (17.9 percent vs. 15.3%), it is significant (p < 0.05, one-tailed). Our estimates show significant sociodemographic differences among our groups of interest. Foreign-born Hispanics are more likely to be male and are older than US-born Hispanics, and they are less likely to have a high school degree or GED. Our estimates of educational attainment among foreign-born Hispanics are similar to those reported by the Pew Hispanic Center (Brown and Patten 2014). US-born Hispanics are younger than both non-hispanic whites and foreign-born Hispanics. Regarding educational attainment, USborn Hispanics report education levels more similar to US-born whites than foreign-born Hispanics. For example, while 62.8% of non-hispanic whites reported some college or 7

8 more, 50.1% of US-born Hispanics and only 27.9% of foreign-born Hispanics reported similar levels of education. US-born Hispanics also reported higher levels of income compared to foreign-born Hispanics, but slightly lower than non-hispanic whites. Another indicator of economic wealth, household ownership, was significantly higher among non-hispanic whites (75.8%) than either US-born Hispanics (58.0%) or foreignborn Hispanics (45.4%). 8

9 TABLE 1. H1N1 Vaccination and Descriptive Statistics 2010 National Health Interview Survey White Hispanic, Hispanic, Characteristics Non-Hispanic (n=8,780) US-born (n=1,142) Foreign-born (n=1,912) % 95% CI % 95% CI % 95% CI H1N1 Vaccination Yes 20.7 ( ) 17.9 ( ) 15.3 ( ) No 79.3 ( ) 82.1 ( ) 84.7 ( ) Sex Male 48.8 ( ) 48.5 ( ) 54.1 ( ) Female 51.2 ( ) 51.5 ( ) 45.9 ( ) Age ( ) 52.4 ( ) 36.8 ( ) ( ) 40.5 ( ) 54.0 ( ) ( ) 7.1 ( ) 9.3 ( ) Marital Status Married 57.4 ( ) 44.2 ( ) 61.8 ( ) Not married 42.6 ( ) 55.8 ( ) 38.2 ( ) At Risk No 78.2 ( ) 83.5 ( ) 86.6 ( ) Yes 21.8 ( ) 16.5 ( ) 13.4 ( ) Healthcare Coverage No 13.3 ( ) 24.6 ( ) 47.0 ( ) Yes 86.7 ( ) 75.4 ( ) 53.0 ( ) Education Less than HS 9.8 ( ) 17.7 ( ) 48.1 ( ) High School or GED 27.5 ( ) 32.2 ( ) 24.0 ( ) Some College/Associates 32.4 ( ) 34.5 ( ) 17.1 ( ) Bachelor s or more 30.4 ( ) 15.6 ( ) 10.8 ( ) Homeownership Status Rent, other 24.2 ( ) 42.0 ( ) 54.6 ( ) Own 75.8 ( ) 58.0 ( ) 45.4 ( ) Household Income <$20, ( ) 16.4 ( ) 25.8 ( ) $20,000-$39, ( ) 22.5 ( ) 32.7 ( ) $40,000-$69, ( ) 27.7 ( ) 24.0 ( ) >$69, ( ) 34.0 ( ) 17.5 ( ) Note: % and CI estimated accounting for complex sampling design of NHIS 9

10 In Table 2 we present the estimated adjusted odds ratios for H1N1 vaccination by race/ethnicity and nativity status, controlling for various sociodemographic characteristics. In model 1, we find significant differences in odds of vaccination between foreign-born Hispanics and their non-hispanic whites. Foreign-born Hispanics report 23% lower odds of vaccination (p<0.01) compared to non-hispanic whites after accounting for differences in the age, sex, marital status, region of residence, and at-risk status. US-born Hispanics did not experience significantly different odds of vaccination as non-hispanic whites. People with pre-existing conditions were 74% more likely to vaccinate than those with no such conditions, a finding that also remains relatively consistent through all models. Additionally, we note the significant differences in odds of H1N1 vaccination by age, sex, marital status, and at risk status through all the models. Model 2 introduces SES-related variables, factors that may be important in distinguishing differences in "flexible resources" as described by FCT. Higher levels of income and education do appear to be predictive of increased odds of H1N1 vaccination. Those with a bachelor degree or greater were 92% more likely to vaccinate than those with less than a high school education. While not as strong of an effect as education, those in the highest income bracket were 27% more likely to vaccinate than those in the lowest. Homeownership status, another indicator of wealth, is modestly linked to higher odds of vaccination. It is important to note that controlling for FCT related factors attenuates the lower odds of vaccination found for foreign-born Hispanics in model 1. Model 3 introduces healthcare coverage into the analysis. Healthcare coverage is entered apart from other FCT measures to show how it might partly mediate the effects of lower income and education. Healthcare coverage is a strong predictor of vaccination, 10

11 with those covered by any type of healthcare coverage 91% more likely to vaccinate than those with no coverage even when controlling for SES. The odds of vaccination for foreign-born Hispanics appears to increase substantially from those reported in model 1, being significant at the p<0.1 level. 11

12 TABLE 2. Odds of 2009 H1N1 Influenza Vaccination 2010 NHIS Characteristics Race/ethnicity Sex Age Model 1 Model 2 Model 3 AOR 95% CI AOR 95% CI AOR 95% CI White, non-hispanic Hispanic, US-born Hispanic, For-born 0.77** Female Male 0.68*** *** *** ** *** *** ** *** ** Marital Status Married Not married 0.75*** * Target Population No Yes 1.73*** *** *** Homeownership Status Rent, other Own 1.16* * Household Income $0-$19, $20,000-$39, $40,000-$69, $70, * Education Less than High School High School or GED Some College/Associate s 1.44*** ** Bachelor or more 1.92*** *** Healthcare Coverage No 1.00 Yes 1.91*** Valid n. 11,474 11,431 11,411 p <0.1; p <0.05; **p <0.01; ***p < AOR, adjusted odds ratio; CI, confidence interval. Unweighted sample size. 12

13 H1N1 vaccination among foreign-born Hispanics. Notable demographic differences exist among foreign-born Hispanics. Among foreign-born Hispanics, we estimate 39.7% being naturalized citizens. A minority of foreign-born Hispanics responded to the NHIS interview in Spanish only (33.3%), with most responding in either English or both English and Spanish (66.7%). The majority (58.7%) of foreign-born Hispanics report being in the US for 15 years or more with 34.8% reporting being in the US for 5 to 15 years, and 6.6% being in the US for less than five years. A clear majority of foreign-born Hispanics in our sample reported Central America (89.4%) as their region of birth compared to those born in South America (10.6%). 13

14 TABLE 4. Migrant specific characteristics, foreign-born Hispanics 2010 National Health Interview Survey Characteristics Hispanic, Foreign-born (n=1,912) % 95% CI Citizenship Status Non-US Citizen 60.3 ( ) US Citizen 39.7 ( ) Survey Language English 66.7 ( ) Spanish 33.3 ( ) Time in the US < 5 years 6.6 ( ) 5 to <15 years 34.8 ( ) 15+ years 58.7 ( ) Region of Birth South America 10.6 ( ) Central America 89.4 ( ) Note: % and CI estimated accounting for complex sampling design of NHIS 14

15 Following a similar pattern as previous analyses, Figure 1 presents the odds of H1N1 vaccination for foreign-born Hispanics focusing on healthcare coverage and migrant specific characteristics in their influence on H1N1 vaccination. Among Hispanic migrants who have resided in the United States for 15 years or more, odds of vaccination were 53% higher than those with <5 years in the United States. While not statistically significant, factors such as US citizenship and English language ability do appear to trend towards higher odds of vaccination. Respondent's region of birth has no effect on odds of vaccination in our analyses. Healthcare coverage positively influences odds of vaccination when compared to those with no coverage. All other variables included (but not shown) were found to be non-significant beyond the p<.05 level, except for those at risk of H1N1 complications (AOR 1.62) and those with some college or associate s degree (AOR 0.53). 15

16 16

17 DISCUSSION/CONCLUSION Unlike previous pandemics, rich data on vaccination uptake exists for the 2009 H1N1 pandemic. With the benefit of using large, representative data, we provide estimates of H1N1 vaccination for non-hispanic whites and Hispanics using the NHIS. Unlike other health related surveys, the NHIS allows us to incorporate nativity status into our analysis, expanding our ability to analyze some of the diversity within the Hispanic ethnic category, specifically nativity. Our analyses show significant within-group differences among Hispanics with US-born Hispanics vaccinating at higher levels (17.9%) than foreign-born Hispanics (15.3%). Both groups of Hispanics reported lower rates of H1N1 vaccination than non-hispanic whites (20.7%). As predicted by FCT, our logistic regression models show that controlling for inequalities in education and income/wealth accounts for the disparity in odds ratios for H1N1 vaccination between US-born whites and foreign-born Hispanics. Decreasing these social inequalities will likely contribute to increased utilization of healthcare services such as vaccines among all Hispanics, especially the foreign-born. While household income and homeownership status were not powerful predictors of H1N1 vaccination, increasing levels of education tended to be positively associated with higher odds of H1N1 vaccination. The importance of addressing educational disparities among Hispanics, often described as the linchpin of Hispanic social mobility, is well documented (Telles and Ortiz 2009). An important finding from our analysis is the powerful impact healthcare coverage exhibits on H1N1 vaccinations. Even when including mediating factors such as education and income, healthcare coverage remains an independent influence increasing 17

18 odds of H1N1 vaccination by 91%. While we agree that society must address deep structural disparities in income and education, in the short-term, expanding healthcare coverage appears to be more effective (and more amenable to policy change) in addressing H1N1 vaccination disparities than eliminating those socioeconomic disparities that may hinder vaccination for a future influenza pandemic. 18

19 WORKS CITED Brown, Anna, and Eileen Patten "Statistical portrait of the foreign-born population in the United States, 2012." Washington, DC: Pew Hispanic Center. Burger, Andrew E, and Eric N Reither "Monitoring receipt of seasonal influenza vaccines with BRFSS and NHIS data: Challenges and solutions." Vaccine. Centers for Disease Control and Prevention "National Health Interview Survey: The Principal Source of Information on the Health of the U.S. Population." "Final estimates for Seasonal Influenza and Influenza A(H1N1) 2009 Monovalent Vaccination Coverage - United States, August 2009 through May, 2010.". Dustmann, Christian, Tommaso Frattini, and Gianandrea Lanzara "Educational achievement of second-generation immigrants: an international comparison." Economic Policy 27(69): Kochhar, Rakesh Latino Jobs Growth Driven by U.S. Born: Immigrants No Longer the Majority of Hispanic Workers. Washington, D.C.:: Pew Research Center's Hispanic Trends Project, June. Kwan-Gett, Tao Sheng, Atar Baer, and Jeffrey S Duchin "Spring 2009 H1N1 influenza outbreak in King County, Washington." Disaster medicine and public health preparedness 3(S2):S109-S16. Link, B.G., and J. Phelan "Social conditions as fundamental causes of disease." Journal of health and social behavior: Miller, Russell R, Boaz A Markewitz, Robert T Rolfs, Samuel M Brown, Kristin K Dascomb, Colin K Grissom, Michael D Friedrichs, Jeanmarie Mayer, Eliotte L Hirshberg, and Jamie Conklin "Clinical findings and demographic factors associated with ICU admission in Utah due to novel 2009 influenza A (H1N1) infection." CHEST Journal 137(4): Nichol, Kristin L, and John J Treanor "Vaccines for seasonal and pandemic influenza." Journal of Infectious Diseases 194(Supplement 2):S111-S18. Pabon-Nau, Lina P, Amy Cohen, James B Meigs, and Richard W Grant "Hypertension and diabetes prevalence among US Hispanics by country of origin: the National Health Interview Survey " Journal of general internal medicine 25(8):

20 Phelan, J.C., and B.G. Link "Controlling disease and creating disparities: a fundamental cause perspective." The Journals of Gerontology Series B: Psychological Sciences and Social Sciences 60(Special Issue 2):S27. Quinn, Sandra Crouse, Supriya Kumar, Vicki S Freimuth, Donald Musa, Nestor Casteneda-Angarita, and Kelley Kidwell "Racial disparities in exposure, susceptibility, and access to health care in the US H1N1 influenza pandemic." American Journal of Public Health 101(2):285. Ritger, KA, RC Jones, KN Weaver, E Ramirez, S Smith, JY Morita, CJ Lohff, SB Black, JD Jones, and W Wong "2009 pandemic influenza A (H1N1) virus infections-chicago, Illinois, April-July 2009." Morbidity and Mortality Weekly Report 58(33): Shrestha, Sundar S., David L. Swerdlow, Rebekah H. Borse, Vimalanand S. Prabhu, Lyn Finelli, Charisma Y. Atkins, Kwame Owusu-Edusei, Beth Bell, Paul S. Mead, Matthew Biggerstaff, Lynnette Brammer, Heidi Davidson, Daniel Jernigan, Michael A. Jhung, Laurie A. Kamimoto, Toby L. Merlin, Mackenzie Nowell, Stephen C. Redd, Carrie Reed, Anne Schuchat, and Martin I. Meltzer "Estimating the Burden of 2009 Pandemic Influenza A (H1N1) in the United States (April 2009 April 2010)." Clinical Infectious Diseases 52(suppl 1):S75- S82. StataCorp, LP. 2013a. "Stata 13." StataCorp LP., College Station, Texas, United States. stata. com b. "Stata 13 Base Reference Manual: Linktest." Suro, R., and J.S. Passel The rise of the second generation: Changing patterns in Hispanic population growth: Pew Hispanic Center Washington, DC. Telles, E.E., and V. Ortiz Generations of exclusion: Mexican Americans, assimilation, and race: Russell Sage Foundation Publications. Thompson, Deborah L, Emily Hancock, Chad Smelser, PA-C David Selvage MHS, MD Joan Baumbach, and MS Mack Sewell DrPH "Risk factors for 2009 pandemic influenza A (H1N1)-related hospitalization and death among racial/ethnic groups in New Mexico." American journal of public health 101(9):1776. Tognotti, Eugenia "Influenza pandemics: a historical retrospect." The Journal of Infection in Developing Countries 3(5): Umaña-Taylor, A.J., and M.A. Fine "Methodological implications of grouping Latino adolescents into one collective ethnic group." Hispanic Journal of Behavioral Sciences 23(4):

21 Uscher-Pines, Lori, Jurgen Maurer, and Katherine Harris "Racial and ethnic disparities in uptake and location of vaccination for 2009-H1N1 and seasonal influenza." American Journal of Public Health:AJPH v1. Wisnivesky, Juan P, Katherine Krauskopf, Michael S Wolf, Elizabeth AH Wilson, Anastasia Sofianou, Melissa Martynenko, Ethan A Halm, Howard Leventhal, and Alex D Federman "The association between language proficiency and outcomes of elderly patients with asthma." Annals of Allergy, Asthma & Immunology 109(3):