The first few cases of 2009 H1N1 influenza were detected in Mexico and California in April 2009. Subsequently, the virus rapidly spread around the world and a pandemic was declared by the World Health Organization (WHO) on 11^th June 2009 [1]. With the advent of the pandemic, significant attention turned to the timely production of vaccine to prevent morbidity and mortality and the first vaccines became available in the United States (US) in October 2009 [2]. Attention also turned to ensuring equity in the global distribution of vaccine to countries unable to afford or finance vaccines themselves. Specifically, the WHO director-general called for “international solidarity to provide fair and equitable access to pandemic influenza vaccines for all countries” [1], [3]. The Bill & Melinda Gates Foundation proposed a set of ethical principles to “guide the global allocation of pandemic vaccines,” including “availability of vaccine in developing nations in the same timeframe as developed nations” [4]. In practice, equitable access to vaccine would mean that countries concurrently (or in the same timeframe) receive a proportionate share of vaccines based on the percentage of their population at risk of disease, and irrespective of their ability to place advance orders with vaccine manufacturers, or indeed, their ability to pay. In other words, equitable global access to vaccines in a pandemic is dependent on not only the number of doses accessible, but also the timing of access to vaccines.

Based on these principles, early access to vaccines during the 2009 pandemic, which was largely dependent on nation states' ability to pay for advance purchase agreements, was inequitable. Many nations bought vaccine to cover a large proportion of their population. For example, Canada ordered 50 million doses for 33 million citizens and estimated that 45% of its population was immunized by January 2010 [5] (See Figure 1 for a timeline of vaccine availability and donations during the 2009 H1N1 pandemic). Though research in Canada showed that public support for increased global assistance—even at the expense of resources for Canadians—existed before the pandemic [6], it was only one day after the 2^nd wave of the flu pandemic was declared “over” in Canada that the country announced that it would donate five million excess vaccine doses to the WHO [7]. The US ordered ∼195 million doses at first, adding another 55 million to make a total of 250 million doses for a population of 300 million [8]. The country announced on September 17^th 2009 that it would donate 10% of its supply to poorer countries through the WHO [9]. However, when only limited vaccine became available in October 2009, vaccine was recommended for target groups in the US totaling about 160 million people [2] and vaccine donation was delayed until at-risk people in the US got the vaccine [10], [11]. By January 11^th 2010, vaccine shortages had eased considerably—61 million people in the US had received the vaccine—and US Secretary of Health and Human Services, Kathleen Sebelius, announced that 25 million doses were ready to ship to the WHO [12]. Only an estimated five million people accepted the vaccine in France, which bought 94 million doses; the country tried to sell excess vaccine to countries with no vaccine in January [13]. The irony associated with reduced uptake in developed countries of a vaccine that was completely unavailable to most of the world population has been commented on by others [14]. Manufacturers of the vaccine also donated doses. For example, GSK's vaccine was approved in September 2009 and the company pledged 50 million doses to the WHO in November 2009. It shipped these doses beginning in January 2010, approximately three months after vaccine was first shipped to paying customers [15].

Donations by manufacturers and nations to poorer countries were coordinated through the WHO. The WHO aimed to provide vaccine to 95 low- and middle-income countries, home to two billion people [16], to cover 10% of their population [3]. Two hundred million doses of vaccine were pledged to the WHO, including 120 million doses from US, Australia, Brazil, France, Italy, New Zealand, Norway, Switzerland, and Britain [17]. The first shipments of donated vaccine to arrive in a developing nation reached Mongolia in January 2010 [18] compared to the arrival and availability of vaccine in the US and France, for example, in October 2009 [2], [19]. As of August 2^nd 2010 (close to the official end of the pandemic), 65 countries had received 60.5 million donated vaccine doses [18].

In sum, sufficient vaccine was donated to cover 10% of the population of poor countries whereas much larger proportions of countries receiving the earliest doses were covered; the principle of equitable access, as defined by the Gates Foundation, was further violated because the timing of donations was much after those at risk in the donating nations had access to vaccine (See Figure 1 for a timeline of the pandemic).

Given a similar risk for specific populations across select nations—based on the hemisphere, season, number of detected cases, and demographic features of the population—we contend that it becomes a moral issue or an issue of justice for access to be dependent on a nation's ability to pay for doses through advance purchase agreements [20]. The nature of the 2009 H1N1 pandemic resulted in surplus doses in many countries easing the decision to donate purchased vaccine doses [21] and reduced demand from poorer nations; however, in the event of a more severe pandemic with higher mortality rates, demand could outstrip supply of vaccines, leaving nations unwilling to donate vaccine at all.

Facilitated by the WHO, the “Open-Ended Working Group of Member States” negotiated a “Pandemic Influenza Preparedness Framework for the Sharing of Influenza Viruses and Access to Vaccines and Other Benefits” [22], [23], [24]. The framework, presented to the World Health Assembly in 2011, authorizes the WHO to establish and maintain a stockpile of 150 million doses of H5N1 vaccine, of which 100 million doses are meant for distribution to developing countries in need once a pandemic begins [24]. Such a stockpile would enable timely access to vaccine in developing countries in the event of an H5N1 pandemic, though the adequacy of the size of the stockpile will depend on the severity of the pandemic. The framework also encourages WHO member states to “consider making donations and in-kind contributions” to the WHO to improve global pandemic preparedness. Efforts to build self-sufficiency in developing nations' vaccine production capabilities have been limited in the past and focused on diseases other than influenza [25]. The US recently announced grants to the WHO to strengthen developing countries' ability to produce flu vaccine [26], but voluntary donations by rich countries are likely to remain an important means of access to vaccines across the globe in an influenza pandemic in the foreseeable future in spite of the Pandemic Influenza Preparedness Framework [24], especially in the event of a severe pandemic, or a non-H5N1 pandemic.

Accurate knowledge of the potential public support for such a decision could be important to policy makers. We explored what the US public believes about vaccine donation by their country to poorer countries. We hypothesize that views on this issue will vary by demographic characteristics, at-risk status, party affiliation and political ideology, as well as whether the respondent has access to healthcare resources. We describe our findings and provide broad recommendations for researchers, advocacy groups, policy makers, and the public. We contend that on the one hand, governments of developed nations may be able to call on principles of fairness and equity to donate vaccine to poorer nations in a timely manner, and on the other, public and advocacy groups could, in turn, call upon those principles in advocating to policy makers for decisions to donate vaccine such that nations would have concurrent access to vaccines in a pandemic.

This study was approved by the University of Pittsburgh Institutional Review Board. Knowledge Network provided an anonymized data file for analysis and all data were analyzed anonymously.

A representative sample of the US population was randomly drawn from the Knowledge Networks (KN) online research panel. To recruit people to the panel, KN uses random-digit dial and address-based probability sampling methods and provides panelists with access to the internet and hardware, if necessary. For this study, a national sample of 3689 adults aged ≥18, including oversamples of African American and Hispanics, was contacted by email. Between January 22^nd and February 1^st 2010, 2079 respondents completed the survey for a 56% completion rate. KN provided a data file with weighting variables, which incorporate design-based weights to account for the recruitment of panelists and both panel-based and study-specific post-stratification weights benchmarked against the most recent Current Population Survey (CPS) with respect to demographic and geographic distributions of the population aged ≥18. Information on the KN panel is available from their website [27],[28].

The demographic characteristics of the entire sample, including access to health insurance as well as subjective and objective priority, are shown in Table 1.

Ensuring timely access to influenza vaccine throughout the world is a public health goal of the WHO. We have shown that there exists considerable support for timely donations of at least 10% of the US share of vaccines during a pandemic. We acknowledge, however, that our study was conducted in the context of a pandemic in which illness was mild for most and the overall fatality rate was lower than the severe 1918 Spanish influenza pandemic with which it was often compared in media characterizations of the pandemic. The perception that the disease was mild may have played a role in recipients' support for timely vaccine donations. Our results encourage further research—including experimental studies employing hypothetical scenarios such as pandemics caused by a virulent viral strain—to understand the public's potential attitudes toward vaccine donations in a severe pandemic.

Vaccine donations are a political and fiscal decision within donor countries. Developing political consensus for timely donation of influenza vaccines during a pandemic is a challenge. For example, though research showed that public support for increased global assistance during a pandemic existed in Canada [6], the absence of public and political discourse regarding this issue presumably made it impossible for the government to donate excess doses until after the 2^nd wave of the flu pandemic was declared “over” [7]. It may also be challenging to garner the fiscal resources to support vaccine availability internally as well as to ensure adequate supply for donation to other countries. Therefore, to address these policy decisions requires discussion and planning before as opposed to during the next pandemic. This may be increasingly important in the current fiscal environment for many countries.

Our results suggest that there is more public support for vaccine donations in the US than policy makers may have expected. Considerable bi-partisan—though short of majority—support exists for the timely donation of 10% of vaccine. If global health advocacy groups wanted to boost support within the US for vaccine donations to poorer countries, they could potentially frame either a moral argument that is based on justice or a utilitarian argument that we live in a global economy, which makes it critically important for economic reasons that vaccine must be available throughout the world in order to maintain the ‘just-in-time’ distribution of products and goods across the globe. Arguments can also be centered on maintaining basic capabilities and functionings of people in poorer countries [30]: in countries that lack adequate access to high quality curative healthcare, increased access to vaccines (at a level depending on people's vulnerability) could be imperative if people are to maintain their livelihood and other capabilities during a pandemic. Arguments could be received differently, however, depending on the political ideology of the recipient. Recent multi-stage research conducted recently—employing qualitative focus groups and quantitative survey methods—furthered understanding of not only how the public and opinion elites understand messages dealing with another public health issue with moral underpinnings—health disparities and the social determinants of health—but also of wording of messages that would most resonate with the public and build support for evidence-based policy change [31]. We suggest the need for similar research on message wording that would resonate in a bi-partisan manner to build further support for vaccine donations in future pandemics.

We found support for donation among lowest income groups who were, in fact, the most likely to support donation of more than 10%. That nativity was related to views on the amount and timing of vaccine donations is a potentially interesting finding. In the absence, however, of knowledge regarding the country of origin for those not born in the US, we hesitate to draw a generalizable conclusion from this finding.

Although we found substantial support for donation, the 2009 H1N1 pandemic was perceived by many to be mild and it is unclear what the impact on attitudes would be if mortality rates were higher in a pandemic. Future research should explore the extent to which severity may affect public attitudes and political calculations of policy makers. While one might speculate that severity may decrease public willingness to support donation, an economic and pragmatic argument, which focuses on the critical importance of vaccine worldwide to maintain the global just in time economy, may yet be persuasive. Policy makers may also imagine that in a more severe pandemic, support for donation would decrease if it requires prioritization of vaccine distribution in the US. However, we did find bipartisan support for prioritization of vaccine recipients; prioritization would, in reality, have to occur, at least initially in a pandemic, regardless of vaccine donation decisions.

Ensuring that influenza vaccine is available in adequate supplies and distributed in a timely manner during an influenza pandemic is a global public health goal. As part of the ‘Open-Ended Working Group of Member States’ of the WHO [22], countries negotiated a potential material transfer agreement that will govern the transfer and use of biological materials and resulting products during a pandemic. Neither the agreement nor our study addresses the issue of delays in vaccine delivery after receipt of donated shipments by the WHO or the ethical implications of requiring poorer nations to meet stringent conditions before receiving donation shipments; these are subjects for future research.

The new ‘Open-Ended Working Group of Member States’ framework [24] stipulates that a recipient (such as a vaccine manufacturer) of viral material commit to donating “at least 10% of real-time pandemic vaccine production to the WHO,” and reserve an equal amount “at affordable prices to WHO.” Faced with a severe pandemic, however, demand for scarce vaccine in poorer nations could outstrip the 20% of production available to the WHO under this framework. In such a scenario, it may become morally imperative for countries with large advance purchase agreements with vaccine manufacturers to consider donating vaccine to poorer countries. We argue that in light of our results, it is important for US policy makers as well as international partners to have a clear understanding of existing public support for timely vaccine donations. In order to frame arguments that would resonate in a bi-partisan manner, further research on communication strategies is required. The results of such research could inform methods to bolster existing support among the public and to advocate with policy makers on this issue prior to the next pandemic.