|Home | About | Journals | Submit | Contact Us | Français|
State and local governments employed various distribution strategies to minimize racial and ethnic disparities in influenza vaccination during the 2009-H1N1 pandemic. Using nationally representative survey data of U.S. adults, we found disparities in 2009-H1N1 vaccine uptake between Blacks and Whites (13.8% vs. 20.4%), while Whites and Hispanics had similar 2009-H1N1 vaccination rates. Physician offices were the dominant location for 2009-H1N1 vaccinations, especially among minorities. Our results highlight the need for a better understanding of how various methods of communication and vaccine distribution strategies affect vaccine uptake within minority communities.
Epidemiologic data collected over the past century suggests that racial and ethnic minorities are at greater risk of contracting seasonal and pandemic influenza and experiencing more negative consequences compared to Whites 1–3. Despite this heightened risk, minorities in the U.S. have historically been vaccinated for influenza at rates as much as 15%–18% percentage points lower than Whites, reflecting access barriers, negative attitudes towards vaccination, distrust of the medical system, and perceived risk of side effects.3–7 In order to minimize disparities in vaccine uptake during the 2009-H1N1 pandemic, local public health authorities adopted specifically targeted outreach efforts to encourage 2009-H1N1 vaccination among minorities. These outreach efforts included the use of alternative vaccination sites, such as retail clinics and school-located clinics, engagement of faith-based organizations, and communication in multiple languages and through ethnic media.8–10 Furthermore, the federal government made 2009-H1N1 vaccine available free-of-charge to remove cost-related barriers to uptake. However, local public health officials reported disparities in uptake of 2009-H1N1 vaccine.11 To our knowledge, the only published national data on this topic measures uptake through the first few weeks of the vaccination campaign.12 In order to assess whether targeted outreach to minority populations during the 2009-H1N1 pandemic succeeded in narrowing historic disparities in influenza vaccination, we used national, cross-sectional survey data measuring influenza vaccination of adults to estimate uptake of seasonal and 2009-H1N1 influenza vaccination, vaccination location, and attitudes toward influenza vaccination by race and ethnicity.
We fielded a nationally representative survey of U.S. adults age 18 and older (n= 4,040) between March 5–24, 2010 using an online research panel operated by Knowledge Networks. Knowledge Networks recruits panelists through a probability-based sampling methods that include both online and offline households.13–14 To ensure diversity, we oversampled older panelists, Blacks, and Hispanics (Table 1). The completion rate among sampled panelists was 73%. Respondents self-identified their race(s) as White (Non-Hispanic), Black (Non-Hispanic), Hispanic, Other, or Multiracial. Respondents (n=503) who self-identified as “Other” or “Multiracial” were excluded from the analysis for a final sample of 3,537. We used questions about age, chronic health conditions, work as a healthcare professional, and personal contact with high-risk individuals to determine whether a respondent was recommended for seasonal and/or 2009-H1N1 influenza vaccination by the Centers for Disease Control and Prevention’s Advisory Committee on Immunization Practices.15 To determine vaccination status, survey respondents were asked: “Have you received a H1N1/Swine [seasonal] flu vaccine this flu season?” We report estimates of 2009-H1N1 and seasonal influenza vaccination, location of vaccination, and attitudes toward influenza vaccination. All data were weighted to produce nationally representative estimates adjusting for known selection probabilities, sample stratification, non-response, and internet use prior to recruitment into the panel.16 Pearson’s chi squared tests were used to calculate p-values. We assessed the robustness of our bivariate results by estimating logistic regression models of influenza vaccination and vaccination location controlling for household income, insurance status, age, gender, employment status and recommendation status. All statistics were generated using STATA 11.
Whites were significantly more likely than Blacks to receive a 2009-H1N1 vaccination (20.4% vs. 13.8%, p=.020) and a seasonal influenza vaccination (42.6% vs. 32.2%, p=.004) during the 2009–2010 vaccination season (Table 2). While Whites were also more likely to receive a seasonal influenza vaccination than Hispanics (42.6% vs. 29.5%, p=.002), there were no significant differences with respect to 2009-H1N1 vaccination between Whites and Hispanics (20.4% vs. 18.6%, p=.62). The statistical significance of the uptake of 2009-H1N1 vaccination between Blacks and Whites persisted when controlling for the aforementioned covariates (OR: 0.67, p=.05).
For vaccinated adults of all races, physician offices were the dominant site of 2009-H1N1 and seasonal influenza vaccination. Blacks were less likely than Whites to be vaccinated in retail clinics for both 2009-H1N1 (2.9% vs. 12.4%, p=.001) and seasonal influenza (3.7% vs. 17.6%, p<.001). The statistical significance of these differences remained when controlling for covariates in multivariate models (results available from authors upon request). Hispanics were less likely than Whites to be vaccinated for 2009-H1N1 (6.2% vs. 18.8%, p=.01) and seasonal influenza in the workplace (10.5% vs. 21.4%, p=.03), though these differences were not statistically significant in multivariate models.
Attitudes toward influenza vaccination differed by race and ethnicity. Both Blacks and Hispanics were less likely than Whites to agree that vaccines are “safe in general” and more likely to agree that influenza vaccines can cause people to get the flu (p<.001–.02).
Our analysis showed that historic racial and ethnic disparities in influenza vaccination persisted during the 2009-H1N1 pandemic, though their magnitude varied across groups and type of vaccination. Most notably, we found that Hispanics and Whites had similar vaccination rates for 2009-H1N1. Since the virus originated in Mexico this finding may in part be attributable to heightened awareness of 2009-H1N1 within the Hispanic community. This finding, however, is also consistent with previous research suggesting that Hispanics face cost-related barriers to vaccination, which may have been alleviated by offering 2009-H1N1 vaccine free-of-charge in community settings.4 The lower vaccination rates for both seasonal and 2009-H1N1 vaccine among Blacks, on the other hand, may suggest that attitudinal barriers such as historic distrust (often mentioned by unvaccinated Black adults)4 were not as easily overcome though targeted outreach and making vaccinations available free of charge. This conjecture is supported by literature showing that Blacks are less likely to have positive attitudes towards vaccination (and get vaccinated) even when vaccination is specifically recommended.5 Furthermore, our results indicated that physician offices were the dominant vaccination site for seasonal and 2009-H1N1 vaccine among all races, but Whites were more likely to also use complementary vaccination locations, such as health department clinics and retail settings. Previous research demonstrates that retail clinics serve communities with fewer Black residents and thus may not be effective at alleviating vaccination-related disparities.17–18 Furthermore, research has suggested that offering mass vaccination clinics, of the type used in the 2009-H1N1 response in which several thousand people received vaccine at a school or health department on a given date, may exacerbate disparities because they require individuals to actively seek vaccine.2
Although there is no comparable public data on influenza vaccination uptake by race for the 2009–2010 season, our results for seasonal and 2009-H1N1 vaccination uptake for all adults tracks closely (between 1–3 percentage points) with results from the Behavioral Risk Factor Surveillance System and National 2009 H1N1 Flu Survey.19 A validation study conducted using data from 2004–2008 suggested that in the past our approach has yielded estimates of influenza vaccination among racial and ethnic minorities that were moderately higher than those derived from the National Health Interview Survey.20 Thus, our results should be interpreted as a conservative measure of disparities in influenza vaccination. As such, the substantial disparities in influenza vaccination rates reported here clearly highlight the need for a better understanding of how various methods of communication and vaccine distribution affect vaccine uptake within minority communities.
The survey data used in this study was collected under contract with GlaxoSmithKline (GSK). GSK had no role in the design, management, analysis, or interpretation of these data, and had no role in the preparation, review, or approval of the manuscript. The authors also acknowledge financial support from the National Institute of Allergy and Infectious Diseases (NIAID) under grant R03AI095084. The opinions expressed here are solely those of the authors and do not represent those of RAND, GSK or NIAID.
The authors have no conflicts of interest to declare and have no financial disclosures.
Contributor Statement: LUP contributed to the design of the study, analyzed and interpreted the results, and wrote the manuscript. JM and KH procured the funding for the study, contributed to the design of the study, assisted with the interpretation of the results, and contributed to paper revisions.
Human Participant Protection: IRB approval was received.
Lori Uscher-Pines, RAND Corporation, Arlington VA.
Jurgen Maurer, RAND Corporation, Arlington VA.
Katherine M. Harris, RAND Corporation, Arlington VA.