|Home | About | Journals | Submit | Contact Us | Français|
Since many unvaccinated individuals do not regularly contact the traditional health care system, we sought to determine the role that alternative vaccination locations (e.g., workplaces and retail clinics) could play in increasing influenza vaccination coverage. Between February 14, 2008 and March 10, 2008, a 25-question influenza vaccine questionnaire was administered to a nationally representative, stratified sample of panelists. Our result found that while alternative locations are covering some segments not captured by the traditional health care system (e.g., younger, working, white individuals in metropolitan areas), they are not serving many other segments (e.g., minority, rural, or lower income patients).
Despite the demonstrated benefits of influenza vaccine, [1–13] each year a large percentage of adults go unvaccinated.[14, 15] Efforts aimed at increasing influenza vaccine coverage have focused primarily on increasing vaccination in physicians’ offices (e.g., computer reminders, standing order sets, advertising campaigns)[16–24] but these efforts may have diminishing returns. Working individuals or those without any other reason to see a physician may be less willing to endure the inconvenience of making an appointment in advance and the associated work absence, travel and wait time with an appointment. Alternative locations (e.g., retail stores and workplaces, could be increasingly viable because of the increasing number of retail clinics (health clinics located in retail stores that are staffed primarily by nurse practitioners) and prevalence of influenza immunization at worksites.[26–30]
Our study used data from a nationally representative survey of US adults from the 2007–2008 influenza season to better understand the potential for these alternative locations (retail clinics or worksites) to improve influenza vaccination rates. Our goals were to (1) describe the characteristics of patients who did not receive the influenza vaccine and determine the fraction of those in this group who had little contact with the medical system and therefore could benefit from an alternative location, (2) determine where vaccinated patients received the influenza vaccine (i.e., traditional health care location vs. alternative locations) and the characteristics of patients who were vaccinated at different locations, and (3) compare the characteristics of unvaccinated patients with those using alternative locations. Such comparisons can help identify whether alternative locations are serving populations that are not being vaccinated at traditional health care sites.
We analyzed responses drawn from a nationally representative, stratified sample of panelists recruited by Knowledge Networks (KN) fielded between February 14, 2008 and March 10, 2008. This panel has been used for many previous research studies.[31–34] KN recruited panelists using random digit dialing (RDD) methods and administers surveys on a variety of topics to panelists via the internet in exchange for small financial incentives. KN provides Web-TV hardware, software, and training to households without computers. Telephone exchanges with higher fractions of African-Americans and Hispanics and elderly were oversampled. Sample weighting adjusted for potential biases such as survey non-coverage and non-response.
The 25-question questionnaire consisted of 5 major sections: receipt of influenza vaccine during the past influenza season, experiences talking to health care providers about vaccines, information and views about vaccines, demographic information, and information about health care utilization and health status.
Our analysis separated patients into patients high-risk versus low-risk for influenza and patients with and without a regular health care provider. High-risk was defined using the following Centers for Disease Control and Prevention (CDC) criteria: age greater than 50 years old, significant co-morbidities, or living or working with high-risk patients. We defined patients who did not have a regular health care provider as those who indicated that they did not have a regular health care provider or did not see a physician in the previous 12 months.
Statistical analyses were conducted using STATA 10 (Statacorp, College Station, Texas) and SAS 9.1 (SAS Institute, Cary, North Carolina). Differences were considered statistically significant at a p-value less than 0.05.
The survey was sent to 4,458 panelists aged between 18 and over. A total of 3,043 (68 percent) panelists responded. Table 1 shows the distribution of completed surveys by ethnic and age groups. All analyses were sample-weighted in order to draw nationally representative inferences.
Respondents who were over 50 years old, college-educated, employed, white, chronically ill, and considered CDC high risk and has a personal physician, a longer relationship with the physician, and more physician visits were significantly more likely (p<0.05) to have been vaccinated (Table 2).
Of patients who received the influenza vaccine, 68.4 percent received the vaccine at a traditional health care location (i.e., physician’s office, hospital, or public health clinic), 22.5 percent received the vaccine at work, and 9.1 percent received it at a retail store such as a pharmacy or grocery store. Multivariate analyses (Table 3) showed that respondents who used traditional health care locations were significantly (p≤0.05) more likely to be unemployed, not white and not African American, living in an urban (vs. rural) area, located in the Northeast, Midwest, or South, and to indicate that convenience was an important factor in deciding whether to get vaccinated. Compared to those vaccinated at traditional health care locations, respondents vaccinated at work were significantly (p≤0.05) more likely to live in the South, have an employer who requires influenza vaccination, and to indicate that convenience was an important factor in deciding whether to get vaccinated. Compared to those vaccinated at traditional health care locations, respondents vaccinated at a retail store were significantly (p≤0.05) more likely to be over 50 years of age, non-white, located in the Northeast or Midwest United States, have a designated physician, have had more than one physician visit over the prior year, and considered high risk for influenza. The p-value for living in an urban area was 0.0502. Those vaccinated at a retail store were more likely to be unsure about how many physician visits they had over the prior year.
Our survey was primarily designed to look at national influenza vaccine coverage rates with oversampling of racial minorities and subpopulations covered by the CDC/Advisory Committee on Immunization Practices influenza vaccine recommendations. As described previously, our survey appears comparable to the National Health Interview Survey (NHIS) with some added advantages of being influenza vaccination-focused and Internet-based: more rapid data collection, analysis, and modification to accommodate questions about emerging topics (e.g., vaccination in non-traditional settings) or key subgroups (e.g., employed versus non-employed). Our study has limitations inherent in single year survey-based studies: e.g., recall and response bias, extrapolating sample results to the entire population, potential misunderstanding of questions, and a single year may not represent other years. Additionally, RDD recruitment will miss individuals without telephones, and an Internet-based survey may still deter those not comfortable with such technology, despite active instruction.
Consistent with other sources, our results demonstrate that many individuals, including a large number of those considered high-risk, go unvaccinated each year.[36–38] Thirty-one percent of unvaccinated respondents did not have regular contact with a health care provider (i.e., no physician visits and no personal doctor). Can alternative locations, such as workplaces and retail stores, be an effective means to improve vaccination rates? We believe yes, with some important caveats.
Alternative locations address some population segments not captured by the traditional health care system. Retail stores capture patients who are older, non-white, and high-risk for influenza. Workplaces appeal to those who value convenience in the vaccination decision. Nonetheless alternative locations are not serving many other population segments (e.g., rural or lower income patients) that do not regularly interact with the traditional health care system. In fact, retail stores are serving many individuals who have had multiple physician visits over the prior year.
It is possible that alternative locations can better serve these missed segments. Future studies may explore whether workplaces with lower income or rural employees have influenza vaccination programs. It also may be helpful to look at whether retail stores in rural or non-white communities are offering the influenza vaccine.[37, 39–42] Subsequent studies could take a closer look at the location, advertising, marketing, and staff composition of alternative locations and how these characteristics affect the types of patients who access these locations.
Acknowledgements and Disclaimers: The survey data was collected under a contract with GlaxoSmithKline. The sponsor had no role in the research design or in the preparation, review, or approval of the document. The study design, data management, analysis and opinions expressed here are solely those of the authors and do not represent those of RAND or GlaxoSmithKline.
Publisher's Disclaimer: This is a PDF file of an unedited manuscript that has been accepted for publication. As a service to our customers we are providing this early version of the manuscript. The manuscript will undergo copyediting, typesetting, and review of the resulting proof before it is published in its final citable form. Please note that during the production process errors may be discovered which could affect the content, and all legal disclaimers that apply to the journal pertain.