Study design

We conducted a three-round modified Delphi study between April and October 2021. The Delphi process is a structured and systematised method, which uses a series of sequential questionnaires interspersed with controlled feedback to collect and distil knowledge from a dispersed panel of anonymous experts to build reliable group consensus on a specific judgment issue. Delphi exercises have been used to develop consensus on priorities, methodological approaches, and definitions in immunisation research [12]. Here, we report only on data collected in Round I of the Delphi study (April-May 2021), which was designed to elicit broad and general concepts from the expert panellists using unstructured, open-ended questions. We piloted the survey with a convenience sample of researchers with global health and/or vaccine experience to gather feedback on survey length and clarity, which was incorporated into the final survey.

Participants & recruitment

In accordance with the Delphi study design, we identified participants who were recognised experts in the field of global vaccine delivery and implementation. Individuals were eligible if they were at least 18 years of age, had access to the Internet, could read and write in English, and had been identified as experts in vaccine program implementation. We deliberately sought representativeness with respect to gender, geographic distribution, and specialty areas.

Participants were recruited using two complementary purposive and criterion sampling methods. The initial list of experts was identified through professional contacts of the study team, as well as both grey and peer-reviewed published literature. We identified expert members of publicly listed immunisation committees and panels from various global health organisations, including the Strategic Advisory Group of Experts on Immunisation (SAGE) at the World Health Organization (WHO) and the Scientific Advisory Committee of the Coalition for Epidemic Preparedness Innovations (CEPI). Participants were recruited through standardised emails that were sent to their publicly reported email addresses. Second, we utilised snowball sampling to identify additional experts who may not have been captured in the literature. Upon invitation into the study, participants were offered the opportunity to nominate up to three other experts within their network. Each new nominated participant was subsequently invited into the study, and similarly offered the opportunity to nominate additional experts. Given that some experts may have had multiple affiliations, and therefore could have had multiple email addresses, the study team manually reviewed the list of nominated participants to remove duplicates.

Data collection

A minimal set of socio-demographic data were collected using a structured questionnaire, including participants’ self-reported gender, geographic and content areas of expertise (in which participants could select multiple options), organisational affiliation, and years of experience. Participants were then asked to identify three to five of the most important challenges in i) distribution, ii) prioritisation, and iii) administration of COVID-19 vaccines in LMICS, along with three to five potential solutions to address these challenges. To ensure consistency, we provided standardised definitions for vaccine distribution, prioritisation, and administration at the start of the survey: distribution referred to acquisition, storage, and deployment of vaccines; prioritisation referred to the order in which COVID-19 vaccines should be distributed to populations and geographical areas, and for special considerations for program implementation; administration referred to actions and mechanisms that ensure vaccines are provided to the population safely and effectively. Answers to these open-ended questions were recorded in free text boxes following each question. Data were collected using the Research Electronic Data Capture (REDCap) survey platform hosted at the University of Toronto [13]. Round I of the survey was open from April 20, 2021 to May 15, 2021; participants with incomplete responses were sent reminders on a weekly basis until survey close to increase response rate.

This study received ethical approval from the University of Toronto Research Ethics Board (Protocol Number: 40797). All participants provided online written informed consent before being permitted to access survey questions.

Patients and the public were not involved in the design, recruitment, conduct or dissemination of the study.

Analysis

Using our guiding framework, this analysis adopted a qualitative descriptive approach, which is descriptive rather than interpretive in focus, is well suited to cross-cultural data, and has been used in other qualitative analysis of free text survey responses [14–16]. Given that our qualitative data were limited to brief statements in response to open-ended questions, the qualitative descriptive approach enabled us to organise the results without attempts to interpret our participants’ meaning or “move beyond” the text as it was presented [17]. To enhance the rigour of our analysis we adopted a committee approach whereby four research team members (VH, IB, AA, ML) analysed the data in an iterative process [18]. This involved an initial meeting after each team member had read the consolidated data to discuss initial impressions and define the framework application. Then, each member analysed a subset of the data individually and assigned codes to the text using our framework. Any disagreements or differences in interpretation were resolved through discussion amongst the group.

Data were organised in QSR NVivo 12, coding was conducted first in QSR NVivo 12 and then through team discussion using Microsoft Word. Data was coded deductively using framework analysis as described by Ritchie and Lewis, while allowing for elements of thematic analysis as described by Braun and Clarke, namely inductive identification of themes [19, 20]. We then engaged in multiple rounds of discussion to iterate on the framework (e.g., adding and consolidating domains) and to refine the assignment of codes to the framework. This work was also reviewed by our two senior authors (FSB, RU) and discussed as a team to further ensure congruence in our analytic interpretation of the data.

Governance, decision-making & financing

Participants from across regions described how global vaccine governance structures challenge vaccine acquisition in LMICs. Similarly, participants across donor agencies, governments, and multilateral organisations alike emphasised the impact of global shortages and limited purchasing options for LMICs. One participant with expertise in South Asia emphasised that this was particularly challenging for small countries and countries with financial constraints. Another participant from a multilateral organisation elaborated on this, noting how smaller LMICs may not be in strong positions to negotiate vaccine acquisition with global suppliers. Others described how dependency on vaccine donations or initiatives such as COVAX disadvantaged LMICs. For instance, participants with expertise in East Asia and the Pacific reported that delays in receiving vaccines through COVAX were a source of failure. Another participant from a multilateral organisation reported on a lack of transparency in COVAX’s management. Others emphasised how vaccine nationalism and vaccine hoarding by HICs amplified inequitable access. A participant from a multilateral organisation underscored that a lack of coordinated distribution and procurement mechanisms across countries skewed procurement towards high income and vaccine-producing countries.

At a national level, participants from all regions reported the negative influence of political interests on vaccine programs. Participants with expertise in Africa in particular, described a lack of transparent decision-making. For instance, a participant with expertise in the Middle East and North Africa identified abuse and favouritism in distribution as key challenges. Similarly, participants across multiple regions emphasised that vaccine prioritisation was influenced by corruption, with some participants including reports of ‘queue jumping’ by influential persons. A participant with expertise in Latin America and the Caribbean reflected on how purchasing of vaccines at the national level posed a challenge to sub-national responses, with monopolies preventing local governments from making direct purchases from manufacturers. Another participant with expertise in South Asia described challenges relating to who makes health decisions and considered that this is often done by ‘administrators’ instead of health experts. As one participant from a multilateral organisation summed-up, vaccines are being used as a political tool rather than as a public health intervention.

Participants across all regions described challenges related to inadequate financial resources to procure the quantity of required vaccines, as well as inadequate and unsustainable funding for logistics and cold chain requirements. Some participants highlighted challenges related to a lack of operational funding in general, and the resultant difficulties for delivering vaccines to remote and hard-to-reach areas. A participant with expertise in Latin America also reported how a lack of domestic funding for vaccine procurement limits a country’s vaccine options to only those offered through COVAX, and how this could pose challenges should there be vaccine hesitancy towards the procured vaccine.

Regulatory structures, planning & coordination

Participants described how a lack of strong national regulatory agencies limited LMIC capacity to evaluate and approve new vaccine technologies and manufacturing processes. In addition, one participant with expertise in Latin America and South Asia noted that even once vaccines are approved, many Ministries of Health do not have the capacity and capabilities to manage programs at the required scale. Similarly, a participant with expertise in Sub-Saharan Africa reported on poor and uncoordinated implementation of vaccine programs due to overlaps in functions among tiers of government. Another participant, with expertise in East Asia and the Pacific, pointed out that different regional capacities within countries, including infrastructure, logistics, human resources, and systems, challenge planning and coordination.

Others reported inadequate implementation processes, with one participant affiliated with a donor agency describing how lack of transparent, accountable, and unbiased processes challenges vaccine distribution. Another participant, with a multilateral organisation affiliation, emphasised the challenges of national vaccine deployment capacity and preparedness in constrained health service delivery systems, with limitations in service delivery capacity slowing down vaccine uptake. A participant with expertise in Sub-Saharan Africa described how plans need not be made from scratch, underscoring challenges that stem from ‘reinventing the wheel’ and ignoring existing mechanisms, particularly those used for polio vaccination.

Other participants described challenges due to a lack of microplanning—a specific process in vaccine delivery used to identify priority communities, address barriers, and develop workplans with solutions [21]. For instance, one participant with expertise in Latin America emphasised challenges caused by insufficient microplanning and the need to consider all possible deficiencies and required resources. The same participant emphasised that coordinating across stakeholders can be a major challenge if the stakeholders are not included in initial planning and decision making. However, two participants reported divergent views, instead describing how recent investments in vaccine programs (such as those for Human Papilloma Virus) and already established vaccination campaigns would make distribution less of a challenge.

Prioritisation, demand generation & communication

Participants reported numerous challenges related to prioritising the order in which different population sub-groups are permitted to access vaccines. Participants with expertise across regions reported that vaccine prioritisation should be decided by a National Immunisation Technical Advisory Group (NITAG), with some highlighting how WHO guidance can inform prioritisation decisions. One participant with expertise in South Asia explained that while WHO guidance aims to achieve public health goals (namely to prevent overwhelming the health system and to prevent severe disease and death amongst at-risk groups), decision makers in LMICs make political prioritisation decisions based on economic objectives and political commitments. Other participants reported challenges relating to the role of bribery or favours to ensure certain groups, or certain people, are prioritised. One participant with expertise in South Asia explained how in some LMICs frequent changing of the prioritisation criteria sends mixed messages and creates mistrust amongst the public. Another participant with expertise in Latin America underscored that a major challenge is communicating to populations how priority groups were determined and why, as this is crucial to priority populations understanding when and how to access vaccines, and to ensure trust is built and/or maintained with the general population while they wait their turn. This participant went on to emphasise the importance of understanding and addressing concerns that priority populations may have regarding being the first to be selected to get vaccinated, especially in the case of health care workers.

Participants described how even with prioritisation, there may be additional challenges to demand generation and communication in LMICs, particularly related to vaccine hesitancy and misinformation. Two participants with expertise across regions highlighted challenges related to hesitancy from health workers as a priority group, and how this could have knock-on effects should health workers decline vaccination or not encourage their patients to get vaccinated. Participants across regions reported how misinformation, mistrust, social media, and inadequate communication strategies regarding vaccines can lead to hesitancy, which results in inadequate vaccine coverage. Other experts highlighted the challenges in addressing hesitancy related to the risk profiles of vaccines being used in LMICs, particularly, as one participant described, when it is perceived that HICs are using “premium vaccines” compared to LMICs. Others expanded on this and emphasised the challenges in addressing misinformation related to adverse events following immunisation (AEFIs) and how many in LMICs are taking a “wait and see” approach. One participant with expertise in Latin America and the Caribbean specifically highlighted that communication and demand generation was an area where LMICs were underinvesting and that “anything that has the potential to shake the population’s trust in the vaccine, the vaccinator, or the immunisation program, can negatively impact uptake.”

Vaccine, cold chain, logistics & infrastructure

Participants across regions reported challenges related to the vaccine itself. One participant with expertise across Asia Pacific, Europe and Central Asia noted that vaccines offered through COVAX have a short expiry date, which posed logistical challenges when compounded with delays in vaccine shipment due to lengthy acquisition processes. Others from across regions reflected how unstable vaccine supplies made it challenging to ensure sufficiency and consistency, which created programmatic challenges for making decisions on timing of administration for vaccine types requiring second doses. Participants from across regions reported challenges in LMICs regarding vaccines requiring cold chain. For example, one participant with expertise in North America highlighted that mRNA vaccines, in particular the Pfizer-BioNTech vaccine, may not be appropriate candidates for remote communities due to inadequate cold chain capacities. Others from across regions reported how many LMICs not only have limited cold chain capacity but fragile supply chains in general, and widely differing capacities within countries.

Some participants related these supply chain issues to larger interconnected logistics challenges to vaccine distribution in LMICs. Participants from across regions described logistical challenges including complexities of distributing multiple types of COVID-19 vaccines at the same time, managing expiry of products with short delivery timelines and/or narrow temperature storage ranges, issues to mitigate theft or falsification of COVID-19 vaccines in the primary supply chain or adjacent to it, as well as transport issues such as fuel shortages, difficulties in obtaining transport permits, and road conditions and safety. One participant with expertise in Sub-Saharan Africa reflected on inadequate infrastructure including unreliable power grids to maintain cold chain integrity, while a participant with expertise in Latin America highlighted inadequate facilities to serve as vaccination centres, or centres not placed in locations acceptable or accessible to the population. Others from across regions emphasised challenges brought about by inadequate mobile and outreach activities, and the significant costs associated with creating accessible, acceptable, and adequate infrastructure to safely delivery large-scale COVID-19 vaccination programs.

Service delivery and human resources & supplies

Many participants from across regions reported challenges in service delivery related to reaching remote populations in LMICs. One participant with expertise in North America highlighted a lack of safe roads or airports for delivery of supplies that enable service provision. Others from across regions reported that geographical limitations will ultimately bias prioritisation with remote areas being excluded. One participant affiliated with a multilateral organisation reflected on previous experience with polio vaccination campaigns in Afghanistan and Pakistan in which door-to-door vaccine campaigns were successful to overcome challenges with populations reaching vaccination centres. Participants across regions reported that areas with armed conflicts would also be unreachable for vaccine campaigns. One participant with expertise in the Middle East and North Africa underscored the challenges in vaccinating populations along borders and how ensuring service delivery in these regions requires negotiating agreements between governments.

When considering vaccine service delivery, some participants called into question the capacity and preparedness of LMICs given their already constrained health systems. Others highlighted that few LMICs have adult immunisation programs across the life course in place, thus programmatic readiness for mass adult immunisation was raised as a significant challenge. One participant with expertise in South Asia noted that vaccine programs in LMICs would be further challenged should they need to administer multiple COVID-19 vaccine types at the same time.

Human resource challenges were reported by many participants across regions as a key barrier to vaccination efforts. Participants described an insufficient number and inadequate capacity of vaccinators and a need to train health workers not only for administering vaccines but also for screening and post-vaccine observation. Participants with expertise across regions described how shortages of workers were related to routine immunisation workers being deployed to provide care as part of the COVID-19 response. Participants’ responses across regions also emphasised that a lack of personal protective equipment (PPE), needles, syringes, cold packs, and other vaccine delivery materials posed an ongoing challenge to ensure sustainable and safe vaccine service delivery. As one participant with expertise in South Asia summarised: ensuring enough space, ventilation, volunteers, security staff, and health workers was a challenge for infection prevention and control in the context of overcrowded service delivery sites.

Surveillance and monitoring & evaluation

Participants across regions reported challenges related to poor pharmacovigilance systems, which are of great importance given the administration of new vaccines. Participants also reported the importance of having adequate surveillance capacity to identify and provide timely management of AEFIs. One participant from the Middle East and North Africa region reported how inadequate surveillance systems in general made it difficult to monitor the progress and effectiveness of immunisation activities. Another participant with experience across regions highlighted how limited systems for tracking and follow-up were a challenge to ensuring that two-dose vaccines were appropriately scheduled and provided. Others described challenges to monitoring and evaluation related to inadequate systems infrastructure. For example, monitoring systems were also limited for tracking and monitoring the distribution of vaccines and other key supplies, with disaggregated information systems leading to incomplete or inaccurate data about vaccine target populations.

(In)equity

Our analysis underscores the inequities in global vaccine efforts and the way these, along with the pervasive mistrust that characterises the COVID-19 infodemic, shape vaccine programs. Participants described global inequities in distribution, access, and availability of vaccines, with most emphasising those between HICs and LMICs. Two participants with expertise in Latin America and the Caribbean described how “elite capture” is the biggest challenge. One participant with experience across regions explained how the limited supply to LMICs created a sense of being “left out.” Another participant emphasised that economic inequities underpinning vaccine acquisition are a result of limited progress on decolonisation. Others highlighted inequities within countries where unequal geographic access to health services and distribution infrastructure deepen the urban-rural health divide. One participant affiliated with a multilateral organisation highlighted how migrants are particularly vulnerable to exclusion due to their lack of access to health services. Another participant with expertise in Latin America linked these inequities to public policy, explaining that inequities will be deepened if decision makers don’t take into consideration the fragility and vulnerability of some communities, in particular Indigenous communities. Another described the consequences of inequity as increased corruption and the formation of an ‘underground market’ for vaccines. Participants also reported a lack of trust challenging all levels of the COVID-19 vaccination effort.

Strengths and limitations

Our study used an existing framework to situate the findings and drew on a wide breadth of expert perspectives to inform our analysis. This enabled us to highlight the constellation of challenges and potential solutions to LMIC vaccine implementation. A further strength is the immediate relevance of the analysis to current discourse driving future pandemic preparedness efforts including recommendations from the IPPPR and other groups. On the other hand, the fact that our respondents were able to select multiple regions of expertise could be a limitation, as we were unable to robustly report on regional differences in implementation challenges. However, given that our participants were informed that our Delphi was a global consensus building exercise, their responses were overall more generalised.