The current SARS-CoV-2 (COVID-19) pandemic has brought into sharp focus the readiness and capacities of health and wider systems in the ability to respond and protect the public [1,2]. Real-time situational analyses  are essential as the pandemic evolves, but this learning must build on what is already known from (albeit smaller scale) pandemics, and the role of important wider environmental factors which contributed to control or conversely were found to have delayed an adequate response. Assessment of the environment or situational analyses in health planning and emergency responses are fundamental for effective design and revision of national level policies and implementation of plans based on these. The scope and content of such analyses, of course, must include basic underlying demographic, epidemiological and health metrics of the population, but also factors on the ‘supply-side’ which should account for the wider infrastructure, including technological capabilities. In the case of infectious diseases, analyses must also include the prevailing social norms and cultural context, which may pose additional risks to spread, with an understanding informing which interventions are most appropriate for breaking the chain of transmission . During infectious disease outbreaks, advancements in surveillance, monitoring and modelling have enabled early warning systems and communications via the World Health Organisation (WHO), the Africa Centres for Disease Control and Prevention (Africa CDC), the European Centre for Disease Prevention and Control (ECDC), the US Centers for Disease Control and Prevention (CDC), and others. Together, they form the mechanisms for alerting the global community as outbreaks evolve to an epidemic or pandemic [5–7]. But in addition to these ‘situation reports’ (i.e. what is happening in terms of the disease transmission and its impact), and ideally before the emergence of a pandemic, what do we know about the capacity of a given country to respond? And how do we assess the wider contextual influences which are particularly relevant in a pandemic scenario where advanced health systems and national economies are not enough to ensure successful containment [8,9]?
Our recent work on what can be described as the ever-present pandemic threat of antimicrobial resistance, has suggested the PESTELI framework , which draws attention to the following environmental domains: Political factors, Economic influences, Sociological trends, Technological innovations, Ecological factors, Legislative requirements and Industry analysis . These are more fully defined in Table 1.
Table 1. Definition of PESTELI domains
PESTELI – Political, Economic, Sociological, Technological, Ecological, Legislative, Industry
We conducted a literature review to identify 1) situation analyses in pandemic management, and 2) studies which examined contextual factors influencing pandemic management. In this study, we defined ‘pandemic’ as an infectious disease outbreak that has spread across multiple continents or worldwide, affecting a large population [12,13].
Any study published in English from 01 January 2000 to 01 June 2020 that has 1) performed a situation analysis to assess the environment for pandemic management, or 2) examined macro-level contextual determinants influencing pandemic management of one or more of the following pandemics: Severe Acute Respiratory Syndrome (SARS), Middle East Respiratory Syndrome (MERS), COVID-19, Influenza A (H1N1), Ebola virus disease, and Zika virus disease, were considered in this review, in any country(ies) setting(s). The PICO (Population, Intervention, Comparison and Outcomes)  and SPIDER (Sample, Phenomenon of interest, Design, Evaluation, Research type) inclusion and exclusion criteria were applied at the review stages . Studies focussing solely on other infectious diseases (eg, tuberculosis, malaria, HIV/AIDS, cholera, dengue), non-communicable conditions (eg, obesity, diabetes, Alzheimer disease, substance misuse), or local outbreaks (eg, a Methicillin-resistant Staphylococcus aureus outbreak in one hospital) were excluded.
Search strategy and information sources
The methods used in this review are in line with the PRISMA extension for scoping reviews (PRISMA-ScR) guidelines . The protocol is available from the authors upon request. The PRISMA-ScR checklist was completed to guide study selection and data extraction. We restricted the search period from January 2000 onwards to capture major pandemics. We limited the language to English. We searched PubMed, Ovid MEDLINE, Ovid EMBASE, Global Health, Health Management, and the Cochrane Library databases. Searches included both controlled vocabulary (pre-defined subheadings) (eg, Pandemics) and text words (eg, strategic analysis). The search strings used are provided in Appendix S1 of the Online Supplementary Document.
The title and abstract of the studies yielded from the database and reference list search were randomly assigned into two groups. Three researchers (NZ, RA, HK) participated in the title and abstract screening and in each group, by rotation, one pair independently reviewed each title and abstract and the third researcher resolved the disagreements in decisions (Group 1 – RA, Group 2 – HK). Two researchers (NZ, HK) independently reviewed the full-text articles which passed the title and abstract screening. All discrepancies were discussed and re-examined by the third reviewer (RA) until agreement was reached.
Assessment of study quality and risks of bias
We excluded those studies where a full article was not available (eg, conference proceedings, meeting minutes). We excluded studies that did not include the sections in the preferred reporting items set out in the PRISMA-ScR checklist .
Formal quality appraisal of the included individual studies was not performed, as this would be beyond the aim of this scoping review, which was to map key concepts, types of evidence, and gaps in research [17,18]. Evaluation of intervention and policy effectiveness is not the aim of the current review [19,20].
Data extraction and analysis
Three researchers (NZ, HK, RA) carried out data extraction, with cross-validation for 50% of the studies using a standardised data extraction table (Microsoft Excel, Microsoft Inc, Seattle, WA, USA). We anticipated descriptive results given the qualitative nature of the studies. Key study characteristics, methods of data collection, situational analyses frameworks employed, and which of the PESTELI domains had been examined (E), findings reported on (F) and recommendations made (R) were extracted (Table 2). Factors influencing pandemic management into facilitators and inhibitors against the 7 domains were synthesised (Table 3-6Table 4Table 5Table 6).
Table 2. Study design and PESTELI domains covered in individual studies
PESTILE – Political, Economic, Sociological, Technological, Ecological, Legislative, Industry, E – examined, F – findings reported, R – recommendation proposed
*Indicates types of data included in the study.
Table 3. Facilitators and inhibitors in pandemic management identified: COVID-19
Table 4. Facilitators and inhibitors in pandemic management identified: Ebola
Table 5. Facilitators and inhibitors in pandemic management identified: Influenza A (H1N1)
Table 6. Facilitators and inhibitors in pandemic management identified: multiple pandemics
This study did not require ethics approval as is a literature review.
A total of 176 studies were identified from the primary electronic databases. Two further studies were identified through a search of reference lists. After removal of duplicates and studies in diseases not of interest, a total of 144 records remained for screening. 45 studies were eligible for full text review and 26 studies were excluded with reasons, yielding 19 studies that met the inclusion criteria. Figure 1 summarizes the flow of literature search and screening.
Figure 1. Study flowchart.
Of the included studies, 6 were single country analyses [21–23,25,26,31], and 13 were regional level multi-country studies [24,27–30,32–39]. Fourteen studies had a single disease focus, with 3 studies on COVID-19 [21–23], 4 studies on Ebola virus disease [24–27], and 7 studies on Influenza A [28–34]. Five studies evaluated responses to one or more of COVID-19, SARS, MERS, Influenza A (H1N1), Ebola virus disease, and Zika virus disease pandemics [35–39].
No study included in this review explicitly set out to employ the PESTELI framework, but 3 studies employed alternative frameworks, including the SWOT (Strengths, Weaknesses, Opportunities, and Threat) framework , the PIP (Pandemic Influenza Preparedness) framework , and the SYSRA (Systemic Rapid Assessment) framework . The other 16 studies examined macro-level determinants affecting the response and ability to manage the pandemic, including workforce mobilisation and deployment; adherence of vaccination and antiviral therapy; public knowledge, awareness, and perception; and compliance of non-pharmaceutical interventions. All studies were published after pandemic emergence. The timeline of the pandemics against the publication of the included studies (Figure 2), shows a notable gap for SARS and Zika.
Figure 2. Pandemic and study publication timeline.
Four studies [21,25,28,30] used primary data via interviews or panel discussion with experts and stakeholders as well as secondary data collected through review of literature or other textual sources.
Ten studies were results of work by researchers from a single country [21–23,26,29,32–34,36,39]. Nine studies were outcomes of international collaborations, where all corresponding authors of these international study groups were from high income countries bar 1 [24,25,27,28,30,31,35,37,38]. Two studies involved co-operation between research institutes and international agencies (ie, WHO and UN) [25,28]. Two studies had co-authors from national and local health authorities [21,25]. One study bridged research institutes, national and local health authorities, and the private sector .
Analysis using the PESTELI framework
Though the PESTELI framework was not utilised, one study reported findings in each of the domains . Most studies (16) included analysis of the sociological domain. Notable gaps are evident in the legislative (14 studies), ecological (12 studies) and economic (11 studies) domain. While the political domain features in 11 studies [21,24–28,30,35–37,39] only five of these make recommendations in this domain.
Political facilitators influencing the response included demonstration of political commitment [21,24,26], and strength in governance and stewardship . Inhibitors within the political domain emanated from lack of coordination between central and local governments and inadequate preparedness plans (21,36); discord about which experts and institutes should lead  and the extent of inclusivity of stakeholders .
Under economic factors, international aid and external funds were a facilitator (28,29) but over reliance on external funding was also reported as a barrier . Level of health system financing was an inhibitor [24,30] and facilitator depending on country context, particularly in regards to sustained community health worker investment and enhanced support during pandemics in the case of Ebola in Uganda and Sierra Leone .
Sociological facilitators were high media coverage and maintaining public attention ; professional training of staff in health care and social care organisations ; and social support to citizens in isolation . Conversely, the most frequently reported sociological inhibitors include lack of public knowledge and public health education in infectious disease prevention [23,26,30,31]; stigma and discrimination against infected patients and health care professionals involved in direct patient care [25,26,33,39]; cultural, traditional, and/or religious practices that may over-ride guidance and health protection messages [24,28,39]. Perceived low risk of infection threat and the low value of infection preventive measures [32,34], and, diametrically opposite, anxiety and fear [26,29], also hindered progress. Lack of trust and confidence in authorities and abilities of the health care system to cope affected health-seeking behaviours [26,33]. Recommendations were proposed in 9 studies to address these sociological inhibitors, and some repeated from the first of these studies in 2014 to the latest in 2020. Recommendations include transparent communication between government and citizens to share information that is up-to-date, easy to interpret, and relevant to contexts (eg, tailored information for vulnerable groups) [23,25,26,29,31,33,35–37].
Among the 7 studies, which included ecological analysis, 6 also analysed sociological factors [24,28,30,35,37,39]. The findings suggested that the drastic change in human lifestyle exerted an impact on ecological and environmental profiles, which then influenced human behaviour further. For instance, globalisation (S) and deforestation and climate change (E); dietary habits (S) and livestock production (E); population age distribution (S) and epidemiology profile (E); and international travel (S) and infection transmission (E). High vaccination coverage was the only ecological facilitator reported in 3 studies [28,30,35]. Ecological inhibitors were centred around human behaviour; contact/proximity with wild animals; transmission of zoonotic diseases through livestock production, and high levels of international travel [24,28,37,38].
Among the 11 studies which assessed factors in the technological domain [21,22,25,26,28,30,35–39], existing information technologies did facilitate progress [22,30], but delayed deployment and limited utilisation of such technologies remained an inhibitor resulting in weak surveillance capacity [21,22,25,28,35,36,39]. In terms of the wider industry, internet coverage was cited as a facilitator  and inhibitor when coverage was low . Industry inhibitors were an inadequate supply of personal protective equipment (PPE) and other medical resources [24,25,27,35,36]; and medical staff shortages [24,25,28,30]. As expected, the interdependence between the technological and industry domains is highlighted. Technologies reliant on uninterrupted power and network coverage are obvious examples, but also more basic equipment and supply-and-distribution chains rely on the existing wider industry or the ability to quickly scale up and deploy emergency provisions. Recommendations, including, for example, accelerated mobilisation of research and development (R&D) through incentives, were proposed to mitigate inhibitors in both technological and industry domains to enhance preparedness for future pandemics , but the timescales for this varied.
Overall, as noted above, the legislative domain was a gap in analyses and also was not explicitly assessed in the otherwise comprehensive assessment using the SYSRA framework of the Influenza A pandemic . Five studies reported legislative facilitators [21,24,26,28,37] including travel bans and border closures [21,24,26]. The absence of legal frameworks for declaring an emergency and taking actions was cited as an inhibitor in the Eastern Mediterranean region .
Our findings appear to show missed opportunities for capture and synthesis of learning, based on a comprehensive analysis within and across pandemics. Wider and more timely dissemination of learning is needed. Large time delays between pandemic event and analysis are evident (see Figure 2). There are recommendations that had been made, from the relatively sparse set of studies, but which now appear again in the current pandemic as inhibitors across the 7 domains. This slow knowledge mobilisation has contributed to the apparent lack of preparedness in many countries for the current COVID-19 pandemic [41,42]. The vast range of outputs chosen for situational analyses could be interpreted as a signal that the endeavour is somehow seen as less scientific, or that the application of strategic management analyses in health has yet to mature. Public health journals have provided rapid turnaround on numerous opinion pieces which may have contributed to a disparate body of work lacking a common framework for synthesis. Additionally, this vacuum has left social media platforms as a fertile ground for debate on these macro-level influences . We encourage a more robust and comparable approach. Additionally, data sources used for analyses are largely confined to secondary sources with only 6 studies employing primary and secondary or mixed methods approaches, which means that findings do not benefit from multi-disciplinary inquiry and the necessary data triangulation. While the PESTELI framework is designed to help draw out the influences specific to each domain, the approach also highlights the interconnections and complexity between the domains. The idea of interconnectivity is certainly not a new one when looking at health systems strengthening [2,44]. For example, inclusion of wider industry experts including project managers, data analysts, engineers, and experts in health systems and applied system methodologies must be coupled with the advocacy work and mobilisation of ‘thought leaders’ . We have recently been urged to use this crisis as an opportunity to equip and strengthen the system. The role of social care in this wider definition of health systems needs to be made more explicit. This review unveiled the missed opportunity in integrating community-based care and collaborating with social care organisations in the previous Ebola pandemic and in high income countries in particular, in the current COVID-19 pandemic. The sector was not only underprepared but also inadequately supported, a concern raised well before the COVID-19 pandemic .
We acknowledge that limiting the study language has missed some national/local level studies but made this decision as the aim here was to look at potential for international learning. We encourage future analysis to include studies published in different languages and assess how the facilitators and inhibitors across the PESTELI domains might influence pandemic responses differently in world regions.
While this review was confined to the lessons from emergent pandemics since 2000, previous pandemics, notably HIV, provide us with key lessons about the importance of protecting the most vulnerable groups and the impressive economic gains when a global health coordinated perspective is taken. We need to capture the lessons which enabled that novel threat to be not only contained but also integrated in the planning of robust, holistic health and social care provision, with the political, sociological and technological domains working over time. Further within- and cross-domain analysis may be strengthened using established assessment tools, for example, the governance TAPIC (Transparency, Accountability, Participation, Integrity, Capacity) framework , building on previous work and enhancing comparability. The traditional use of such analysis in management sciences is then to guide a force-field analysis where strategies are formulated to either weaken the inhibitors or strengthen the facilitators whilst also explicitly acknowledging which factors are immutable for the short or medium term. Where political or economic barriers are unlikely to change (as evident by the lack of recommendations in these domains), these constraints are still useful when projecting potential impacts of the programmes with a sociological or technological focus, for example. As we learn and adjust to this novel pandemic we need to prepare for the short, medium and long-term and the framework suggested here can help with the required 360-degree view.
Ex-post analysis using the seven-domain strategic management framework provides further opportunities for a planned systematic response to pandemics which remains critical as the current COVID-19 pandemic evolves.
Contribution of all members of the COMPASS (COntrol and Management of PAndemicS through Strategic analysis) study group is acknowledged: Raheelah Ahmad (City, University of London, London, UK); Rifat A Atun (Harvard T. H. Chan School of Public Health, Boston, Massachusetts, USA); Gabriel Birgand (Nantes University Hospital, Nantes, France); Enrique Castro-Sánchez (City, University of London, London, UK); Esmita Charani (Imperial College London, London, UK); Ewan B Ferlie (King’s College London, London, UK); Alison H Holmes (Imperial College London, London, UK); Izhar Hussain (Dow University of Health Sciences, Karachi, Pakistan); Andrew Kambugu (Makerere University, Kampala, Uganda); Jaime Labarca (Pontificia Universidad Católica de Chile, Santiago, Chile); Gabriel Levy Hara (Hospital Carlos G Durand, Buenos Aires, Argentina); Martin McKee (London School of Hygiene & Tropical Medicine, London, UK); Marc Mendelson (London School of Hygiene & Tropical Medicine, London, UK); Sanjeev Singh (Amrita Institute of Medical Sciences, Kerala, India); Jay Varma (Africa Centres for Disease Control and Prevention, Addis Ababa, Ethiopia); Nina J Zhu (Imperial College London, London, UK); Walter Zingg (University Hospital of Zurich, Zurich, Switzerland). Authors AH, EC, EF, ECS, GB, MM, NZ, RA, and SS gratefully acknowledge the support of ESRC as part of the Antimicrobial Cross Council initiative supported by the seven UK research councils, and also the support of the Global Challenges Research Fund. No funding bodies had any role in study design, data collection and analysis, decision to publish, or preparation of the manuscript.