Content uploaded by Peter Grindrod
Author content
All content in this area was uploaded by Peter Grindrod on Sep 10, 2024
Content may be subject to copyright.
The National Risk Register 2023: Some Reasoned Reflections
Clive E. Bowman1, Peter Grindrod CBE1, Graham M. Smith2,3, and Argyris Stasinakis4
1Mathematical Institute, University of Oxford, Oxford, UK
2GMS Abingdon Ltd, Abingdon, UK
3Clemson University, South Carolina, US
4Oxford, UK
September 9, 2024
Abstract
We provide a constructive response to the recent National Risk Register 2023, highlighting a number of generic
issues. These primarily include reflections on the deployed methodology, and its possible biases.
We suggest that the National Risk Register should be extended to adopt a whole systems perspective, reasoning
that it is unlikely that the acute risks will each occur in an isolated way, and additionally there may be (non-risk)
precursor contexts and events which might amplify many acute risks and/or nullify planned responses.
We discuss some potential causes of a lack of national resilience; and some possible generic failures in acute risk
response.
We make a number of recommendations for consideration by HM Government.
Key words: Acute Risks, Resilience, Cascades, Systems Approach, Dynamics
1 Introduction
The UK’s National Risk Register (NRR), 2023 edition, [1] outlines the most serious individual risks facing the United
Kingdom. These include a very wide range of risks arising from terrorism; chemical, biological, radiological and nuclear
(CBRN) threats; cyber activities; state-on-state threats; geographic and diplomatic risks; accidents and systems
failures; natural and environmental hazards; human, animal, and plant health; societal fissures and public disorder;
and conflict and state-based instability. These potential risks are set out separately and are partially evaluated with
respect to their likelihood and potential impact.
In this reflection we shall consider a number generic issues and perspectives. Our assumption about the general
UK administrative risk landscape is summarised in Fig 1.
Our aim is to provide some points that might be addressed within future, similar, efforts, and to suggest how the
existing approach might be extended to avoid some identified conceptual and methodological omissions and challenges.
These include intrinsic biases, inherent within the present risk review and evaluation procedure, some conceptual
prisms, and reliance on some generic assumptions regarding the background situation (context) within which any
regulatory controls and responses may become active.
Here we will only discuss specific risks as illustrations of the potential impacts of the generic issues raised. It is not
our aim to provide a critique of the individual risk outlines or their evaluations within the NRR.
We realise that some of these matters may be already dealt with by the internal ‘National Security Risk Assessment’
that we are not privy too. Our prime objective is to provide constructive and cheerful challenge from a perspective
that is independent of the Cabinet Office and its network of advisory, regulatory, and responsive agencies, individuals,
and institutions.
We make a number of recommendations arising directly out of our reflections. We hope that this will catalyse a
constructive engagement with Government Science and the Cabinet Office.
2 Familiarity
Hindsight is mostly perfect; foresight clearly is not. Inevitably the description, evaluation, and prioritisation of any
individual risk is governed by any available relevant experience.
1
Figure 1: Assumed broad administrative framework for handling UK crises (links to Devolved Governments if relevant
are omitted). ‘Lead Government Department’ (LGD) model to handle Emergency and Security National risks (left hand
side) with their own Ministerial and Permanent Secretary accountability plus associated governance. Underpinning
‘Resilience Framework’ (right hand side) including Local Resilience Fora (LRFs).
For example prior to the 2019/20 Covid-19 pandemic, the 2017 NRR [2] highlighted the risk from human diseases,
including those with a respiratory mode of human-to-human transmission, saying “Over the past 25 years more than
30 new (or newly recognised) emerging infectious diseases have been identified around the world, such as Ebola, Zika
and Middle East Respiratory Syndrome. The latter emerged recently in 2012 and poses a global health threat.” The
stated responses to this possibility may be summarised as reliant on detection; antivirals; vaccines (“taking at least
four to six months after a pandemic begins”); and personal protective equipment.
In contrast the NRR 2023 [1] is much more assertive about non-pharmaceutical interventions, including border
measures: “Local and national plans for managing excess deaths should be present, and arrangements for effective UK
and global coordination. Plans for social, educational, and economic impacts of the pandemic and expert scientific and
clinical advice should also be in place. Our response capability would need to be able to channel significant research and
development resource to genomics and development of tests, vaccines and therapeutics. A national communications
plan would also be needed to increase awareness and encourage good hygiene. Every sector, including but not limited
to health and social care, will be affected by the pandemic and will require capabilities to respond.”
It is no surprise that the national experience of Covid-19, and its consequences (which included lock-downs, fur-
loughs, school closures, and so on, which were never alluded to in the NRR 2017), should have considerably sharpened
the discussion. The nation’s preparedness for a similar future risk is still learning some of those lessons [3], which
include considerations of early intelligence and response; the media and public’s possible compliant response; the (anti-
vax) fake news, various conspiracy theories (concerning the scale, origin, prevention, diagnosis, and treatment of the
pandemic, which were easily spread through social media, text messaging,and mass media), as well as the government’s
reliance upon, and any public discussion of, scientific modelling and advice.
This is properly the business of the ongoing UK Covid-19 Inquiry, chaired by Baroness Heather Hallett. That
enquiry has twin aims [4]:
•To examine the COVID-19 response and the impact of the pandemic in England, Wales, Scotland and Northern
Ireland, and produce a factual narrative account (including the responses of UK public health, the health and
care sector, the economic response to the pandemic and its impact, including governmental interventions);
•To identify the lessons to be learned from the above, to inform preparations for future pandemics in the UK.
2
These outputs are for very distinct audiences and follow-on purposes, and the details of the former objective may
well dominate those of the latter, which are our principal focus herein. This internal tension and the resulting
marginalization of the “lessons to be learned” objective has been strongly criticised recently by Lord Sumption [5].
Similar observations could be made comparing approaches before and after the Fukushima Nuclear Power Plant
accident. A whole range of “stress tests” were mandated after the event, involving the analysis of processes which had
not been recognised previously or were dismissed. “Following the events at Fukushima, Japan, on 11 March 2011, the
nuclear industry in the UK responded quickly to review UK nuclear installations against seismic and flooding hazards.”
[6]. However, the additional plans for addressing social education impacts referred to above for a pandemic, were
not instituted for nuclear accidents though they could be huge. Nor is there presently any effective communication
strategy. STUK, the Finnish nuclear regulator, has updated its guidance on making decisions after a major nuclear
release; at least identifying holistically the large range of the issues that need to be balanced alongside mitigation of
direct radiation effects [7] .
Optimising the risk responses across all such facets , not just the headline, is presently under consideration in NEA-
OECD. See for example, the activities of the NEA Expert Group on development of a Holistic Process for Decision
Making on Decommissioning and Management of Complex Sites [8]. Similar challenges in addressing complexity and
coordination across the conventional management silos were noted recently at a conference on the execution of the
United Nations Sustainable Development Goals, organised by the International Atomic Energy Agency [9].
Similarly, one might compare the acute risk narratives before and after the Icelandic volcanic ash cloud in April
2010 (and its affect on air travel); or before and after the 2007–2008 financial crisis (which begat a whole new range
of stress tests and required some massive national interventions).
It seems clear that familiarity with, and previous experience of, the risk mechanism must inform analysis within
the risk register. However, this leads to a natural tendency only to discuss in depth those risks that we are already
prepared for or have previous experience of.
This is classic confirmation bias, of course, and it needs to be addressed, perhaps with (i) an outside-in approach,
similar to that discussed in [10], and/or (ii) with an entrepreneurial approach deliberately designed to be trophy-
driven and disruptive in highly novel ways (by definition). As an example, it took a senior external physicist, Richard
Feynman, not a NASA space engineer, to get to the root cause of the Challenger failure [11].
The latter issue has been observed previously with respect to government horizon scanning. In [12] it is pointed
out that horizon scanning is common for many types of institution, but for the public sector they usually involve a
designated expert group of civil servants and particular domain experts within the present relevant paradigms (not the
next generation). Domain expertise seemingly is used to bias the membership of horizon scanning groups to an inner
complicit cabal (with group think). The group produces its report and submits it to another similar committee of civil
servants and particular domain experts for challenge or approval. It may thus be reliant upon rather un-entrepreneurial
types of challenge by people in both producing and marking their own homework.
3 Schr¨odinger’s Ostrich and Black Swans
What is the real value and mission of seeking out, and making observations concerning, possible novel or emergent
risks that might arise suddenly from novel disruptive technologies, novel human actions, or from instabilities (tipping
points or bifurcations)? What of such risks that ‘balloon up’ unexpectedly? How should the NRR evolve?
For example, consider the way that generative AI has emerged recently in 2023 (via large language models such as
ChatGPT, launched November 30, 2022) and has since had the government’s agencies and commentators alike on the
back foot. This begets novel risks that are already upon us, identified in part by the recent AI Safety Summit [13, 14].
Arguably the decadal data science focus on ethics and privacy should have been far less important than the societal risk
of fake truths and the widespread inability to discriminate between human and AI generated commentaries, opinions,
and content; and the introduction of bias. These risks now require an in-depth consideration, as we have to uncover
possibilities.
Yet, of course, if you find something unanticipated is actually occurring now, or else is deemed more likely to occur
soon as a result of investigation, then the government or independent regulator would have to intervene directly at
some cost, and not simply add it to the NRR. This possible challenge to the status quo (and/or implicit identification
of earlier failure to act) induces a behaviour that might be called “Schr¨odinger’s Ostrich”. That is the process
of deliberately not observing, or playing down, theoretically possible (yet uninvestigated) risks and instead being
comfortable in carrying the hidden risk as ongoing uncertainty; as opposed to turning over some stones and finding
that some direct action is required (one way or another). The phrase “We are focusing limited resources on current
priorities” comes to mind here.
Thus some special effort should be invested in generating new and emerging risk scenarios, particularly within fast-
moving areas such as technology and social behaviour. The Schr¨odinger’s Ostrich type of behaviour invites exposure
3
to novel risks, by underestimating their prior likelihood, so that they become Black Swans. A Black Swan [15] is
an outlier (it lies outside the realm of regular expectations) since nothing in the past can convincingly point to its
possibility. Yet it carries an extreme impact. Taleb [15] argues that the consequences of Black Swans play a dominant
role in history.
Of course all of the risks set out within the NRR are anticipated, by definition, and so they will not become Black
Swans; though their combination may do so (see Section 5.2). Hence the NRR process requires some assurance and
active mechanism that ensures that outlier events have been sought out and will become uncovered at the earliest
possible stage.
4 The Prevailing Context and Over-Prescriptive Responses
Any risk scenario and its consequent mitigation and prescribed response ought to consider the prevailing context at the
time of the foreseen risk. This sounds obvious yet there may be strong constraints, or amplifying factors. It requires
a wider range of expertise than is usually available within specific regulators.
A positive example arose recently during an International Atomic Energy Agency (IAEA) discussion of risks asso-
ciated with operating and decommissioning nuclear facilities. At the November 2022 the Radiation Safety Standards
Committee (RASSC) meeting it was stated that elements to be considered within future Safety Standards should
include: (i) pandemics, which may reduce the availability of competent personnel; (ii) international crises such as a fi-
nancial crisis, energy crisis, electricity supply failure, and national and international travel restrictions; (iii) belligerent
actions, including potential military impacts on facilities and terrorist acts, which may damage the integrity of safety
systems; (iv) cyber risks that might impact on those responsible for facilities and regulators and have consequences for
occupational, public and medical exposures; and (v) emerging new technologies which might have impact on imple-
mentation of safety standards. They propose an “All Hazards” approach, to examine any particular scenario through
multiple expert prisms, and allow for constraining overlaps. RASSC suggested that the scope of the term “all-hazards”
should consider all human-induced hazards, including consequence of social unrest and lost political stability, in which
the operation of nuclear facilities, waste management, and disposal facilities, or the use of high-activity sources should
be performed. While acknowledging that the scope of the activities of the IAEA is limited to radiation risks, it is
essential that decisions should not be taken based around radiation risks alone. Where any regulatory body has no
formal competency outside of radiation risks there should be a requirement for that regulator to establish formal links
with other relevant regulatory bodies so that an all-hazards approach can be effectively facilitated. The scope to learn
from such an approach has been acknowledged in the work of the Nuclear Energy Agency of the OECD [16].
The typical response to a recognised hazard, once it materializes is to introduce further prescriptive measures to
reduce the chance of it occurring again or to mitigate the consequences, based on the past experience. However, no two
accidents or events are the same, and even similar accidents and events can occur in highly different circumstances, e.g.
peacetime or war, as we see today in Ukraine. This can lead to over-prescriptive responses to emerging events that may
well be impossible to implement in the present context. Conditions that constrain available expertise, safe access to
facilities, or security of monitoring and operational staff may nullify any planned, intended, mitigation. There is thus
a need for some agility and command options. Not doing so invites a non-response or an ineffective one, amplifying
the risk from the foreseen business as usual context.
Hence the prescribed regulatory action may need to be over-ruled when even more dire events are happening at
the same time, to the extent that you have to break the (over-prescriptive) law to do the right thing.
For example, the scale of damage to the monitoring and radiation protection systems within the Chornobyl Exclu-
sion Zone and considering the loss of physical control Ukraine, the radiological situation in the Kyiv region became a
matter of both public and governmental concern. However, given the level of Russian aggression, the national services
focused upon the potential for massive numbers of immediate civilian deaths. The early investigation of possible
radiation harms to the people were not practically possible [17].
In such situations it is difficult to strike the right balance between prescription that adds clarity and supports fast
decision in emergency, and an agility to act within the specific prevailing circumstances.
5 A Systems Focus
The NRR sets out a set of risks, each considered in isolation. This ensures that the various government departments,
agencies and institutions have a common view of risks in the round and the appropriate stakeholders, with either
monitoring, response, or regulatory authority, can plan for each eventuality. Of course the UK itself is an open
systems with many interactions, non-linearities, and dynamical effects (including stability and instability behaviours).
The UK’s NRR covers a wide range of eventualities. However all of those risk threats occur within a common domain:
4
the geographical situation (the UK) and the relevant time frames. From an expertise point of view each risk is
considered in isolation. In fact, worse, some of the risks are no more than the consequences of some of the others.
Yet the experience of COBRA, the Civil Contingencies Committee that is convened to handle matters of national
emergency or major disruption in order to coordinate different departments and agencies in response, is that taking
the UK as a shared, common, domain and therefore an open coupled system means that some distinct risks may be
coupled, overlapping, or even cascading. This occurs as pressures and disruptions are activated and passed around
within the common system.
The general problem is that addressing those wider acute risk requires the whole of government (pretty much all
government departments) to be involved: they cannot be addressed piecemeal with any suitable levels of priority and
proportionality.
In this section we will discuss a few practical aspects of systems thinking. This includes possible co-occurrences
and cascades of different risks; and the consequent difficulties in implementing mitigation strategies devised a-priori
in silo, which may overlap, interfere, or even become impossible. We strongly agree with Spiegelhalter and Freeman,
who point out the problems of co-occurrences and cascades [18]. We provide the following discussion.
5.1 Limitations of a List-Based Approach
The NRR is presented as a longish and wide-ranging list, classified by source/cause. It is intended to catalyse
an appropriate response to each risk from those government departments, institutions and agencies with relevant
jurisdictions and operating responsibilities.
The problem with this is that it places each risk into its own silo along with its possible mitigation and response.
Yet the UK is an open system, and there is some high chance that risks treated as distinct in the NRR list may occur
close together. This may be due to the following mechanisms.
•Risks may be co-dependent upon some hidden or known precursor conditions or event, which may increase both
of their chances of occurrences.
•Risks may actually be conditionally dependent on one another, with one occurrence increasing the chance of
another. Such cascades are common in risk networks [19].
•Risks may be entirely independent, but could co-occur by non-trivial random chance.
Moreover, although for each risk the probability of occurrence is low, the probability that two (of the set of) risks
may coincide is quite high. Consider the birthday paradox which states that in a random group of 23 people, there is
about a 50 percent chance that two people have the same birthday (despite your chance of having your birthday on a
specific day being low ≈1/365). The more risks that are enumerated (109 in this version of the NRR) the worse this
effect gets. Of course this type of calculation conservatively assumes that the risks are all independent, but in fact this
may be far from true (as observed above, and see Section 5.2), and hence co-occurrences could be even more likely.
These issues are raised in other national efforts. For example, the Finnish National risk assessment 2023 [20] states,
“More than one of the national threat scenarios and disruptions analysed in the risk assessment may materialise simul-
taneously, due to networked society, interdependencies between different functions or other simultaneous unintentional
and intentional actions, for example. The prolongation of large-scale disruptions may also, for its part, increase so-
ciety’s susceptibility to other simultaneous disruptions. With the concatenation of disruptions and the cascading of
impacts, a single disruption may have extensive impacts on different functions in the interdependent operating envi-
ronment. To support risk management, the assessment of threat scenarios and disruptions conducted in the national
risk assessment aims to identify the potential concatenation and cascading of impacts.”
5.2 Compound Risks and Cascades
Analysis of disaster interdependencies has been previously considered within compound, interacting and interconnected
risks. Compound risk refers to the interaction of multiple hazards or events that combine to produce extreme disasters
capable of generating widespread losses [21, 22, 23].
Cascades [24, 25], are just “one thing after another” (see Figure 2 for a climate based example), and often seem
obvious in retrospect, yet the list-based, siloed, consideration of risks seems to count these out, in seeking to invite
stakeholder/institutional response plans for individual risks.
In [25] distinct types of cascade dynamics are usefully identified, as follows.
•Cascades spreading by a snowball dynamic. Examples include environmental pollution, epidemics or other social,
technological, economic or political phenomena that spread and result in cumulative damage over time and space.
5
Figure 2: Simplified cascade model from 2
•Cascading disasters spreading due to the development of mass hysteria or panic, or people failing to exert effort
to effectively reduce the damage.
•Information cascades (and any subsequent learning via common knowledge) that underpin the mobilization,
mediation, or the failure, of the response efforts.
In fact the word “cascade” does not appear at all in a risk cascade context within the NRR23 [1]: it occurs just
once in the context of a cascade (or shower) of space debris (on page 89).
The problem with the co-occurrence (or the sequential triggering) of two or more risks is that it results in their
possible amplification. It also reduces the effectiveness or even the actionability of an individual risk’s designated
response (see Section 4), which had been planned, in silo making implicit assumption that the risk event would trigger
on its own. Indeed there may be co-occurrences that make the planned responses impossible or impotent (see Section
4, for example).
Thus, while a list is useful, the NRR should not encourage planning that consider risks in isolation. A medium
worst case might examine all possible risks pairwise and discuses their possible interaction or connivance.
5.3 An Audit of Second Order Interactions
Following on from the previous section, we suggest that the NRR should examine all second order interactions (at least),
as pairs of risks, set out in a matrix rather than a list. 1Then the (i, j )th term should consider the combined risk
from Risk ioccurring, given that Risk jhas already recently occurred or is presently occurring. Such a second order,
pairwise, consideration would be valuable in pulling the various agencies of state together to war-game those critical
combined scenarios. The order in which such pairs of risks occur may be important for their composite consequences.
In summary, it is a category error to assume that risks will occur independently, in isolation, in-silo.
1Note that although the NRR [1] refers to a “Risk Matrix” (on pages 14 and 15), that should be more properly referred to as a scatter
plot based on the list of risks, each having two attributes: likelihood and impact. It does not consider interactions between pairs of separate
risks.
6
5.4 Amplification via Direct Action and Societal Disorder
Mass protests and civil disobedience in the UK include: “Stop the War”-style protest coalitions, and nonviolent, direct
action and civil resistance, initiatives such as the “Extinction Rebellion” (XR) and “Just Stop Oil” movements; the
mass non-payment of council tax or of TV licence fees; possible national protests against energy prices, the cost of living,
the cost of mortgage/rent payments (as successive interest rate rises may directly cause poverty and homelessness), new
ULEZ costs; or any much more general mistrust in government, democracy, and globalisation. The modern prevalence
of social media has made ground-up self-organisation both efficient and effective.
The public disorder (labelled as “highly unpredictable” and any large scale industrial action (national strikes within
key sectors such as the NHS) risks, set out in the NRR 23, Chapter 4 p.178-180, should surely address this exposure
to ‘mass citizen effort’. Though how any resilience might be assured is not discussed.
Consider President Trump’s American support base predicated on anti-establishment and anti-globalisation inter-
ests (e.g., MAGA and “America first”) and fueled by fake news, conspiracy theories and a mistrust in the political
elite. This ultimately resulted in the Capital Insurrection January 6, 2021 (and may likely return through the next
presidential election). Yet for several weeks before that event there were over one million mentions of storming the
capital on social media (ignoring that was possibly a case of Schr¨odinger’s Ostrich).
Mass scale mistrust in the prevailing establishment’s narratives have also played a role in other political movements,
including a single-issue dominated December 2019 UK general election, which cut across the normal party allegiances.
We may speculate what would have occurred had parliamentarians blocked or ignored the 2016 ‘BREXIT’ referendum
result entirely.
Clearly any discontinuities between the popular public mind and prevailing government policy or direction may
result in direct action and a fracturing of society, based on age, economic, or social divides. For example, in the
Trans-rights versus feminism debate. All such chasms might result in mass disobedience, civil disturbances, or worse.
Besides a much better coverage of such risks, we suggest that there should be a direct consideration of the way
in which public participation may amplify and propagate other (physical, social or economic) risks in the NRR, and
possibly make some existing mitigation strategies (foreseen with a complaint,or at least a passive, public) unworkable.
The threat here is clearly one of second order: the potential “amplification” effect of consequent (knock-on) direct
action and societal disorder should be widely considered with regard to many risks in the NRR, in a way that is
distinct from its own (independent) spontaneous occurrence. This is a system phenomenon.
5.5 General communication of risks
Threats are routinely exacerbated by failures of official communication systems resulting in some lack of clarity [26].
Clearly there is a threat that giving notice of an event that might need intervention can cause panic (hording, extremist
groups stepping in, socio-economic or business losses), but failure to give any warning may result in there being
ultimately too little time to prepare.
More challenging is the failure to distinguish between a significant threat at the individual level, involving real
risks that some specific group of, say ten, identifiable individuals will die, and a large-scale threat that can affect a
huge population, e.g. the population of London, but where the individual risks are miniscule. A one in a million risk
to 10 million people is still ten dead people. Examples of this failure include how the government miss-communicated
on risk of Creutzfeldt-Jakob Disease, how the health department has confusingly reported risks from processed meat
consumption (bacon, etc.), and how the UK’s Health and Safety Executive (HSE) presents the risks from radon and
the actions to mitigate them.
The wider understanding of risk and risk-response concepts and quantitative issues within the general public and the
media is rather poor. This is evident in the level of understanding and engagement with relevant statistics provided by
the UK’s Office of National Statistics and other institutions, and the poor level of questioning addressed to government
spokespersons at press conferences and like.
5.6 On the Interference Between Separate Mitigations
In other fields where up-front documented mitigations for documented risks is standard practice, like for instance the
medical treatment of diseases with drugs, interactions between mitigations are well known (e.g. undesirable or fatal
drug-drug interactions). Ignoring the possibility of interactions is often endearingly typified as “...we all did the right
thing but the patient is still dead”.
Moreover, the existence of one risk may even completely preclude in advance sensible mitigations for another. In
the medical field these are called contraindications, as well-meaning and well justified mitigations when a single risk
(disease) is considered, should not ever be used when another risk (disease) is present. A classic example is pain relief
with thalidomide, it is an effective therapeutic but should not be given to women of child-bearing potential.
7
The NRR does not appear to consider any pairwise contraindications (even obvious ones), nor how mitigations in
general may influence mitigations being used for other risks. Furthermore, it seems to only consider mitigations when
the overall picture is one of ‘business as usual‘. To give a medical example again, the treatment one might give to
prevent a further disease or an exacerbation of a current risk may be different depending upon whether the patient
is chronically ill (i.e., simply somewhat perturbed normal physiological state ) or acutely seriously ill as in emergency
rooms or intensive care. Mitigations interacting with each will be considered differently if the patient is just about to
die without a prompt intervention.
6 Systemic Precursor Risks Not Included in the NRR
The NRR focuses on acute risks rather than chronic risks. However, within the overall system there are some chronic,
generic, embedded behaviours that are precursors of acute risks or are amplifying mechanisms. Any discussion of the
identified acute risks, and potential responses, is incomplete without their consideration. Worse, since they are generic
they should really be owned centrally rather than within individual response silos.
6.1 Common Failures in Emergency Response Command
The actual response to any particular risk, emergency, or disaster may itself amplify or widen the extant risk. Strictly
speaking though, the risk response is not the business of the NRR, which is a register of causes or sources designed to
elicit response planing by relevant departments and institutions, so such internally generated, knock-on, failures (and
additional risks) remain unaddressed by the NRR.
However, in this section we shall argue that there is a systematic issue that is often observable within many emer-
gency responses. And such a common denominator indeed represents a chronic national risk that requires articulating
and addressing. It is caused by institutional internal processes and by a systemic behaviour across a wide number of
the responsible agencies of the state. It results in a chronic national risk to the UK due to the systemic failures in
command response, which should be addressed.
In any fast moving emergencies, human “denial” [27] and disbelief (is this really happening?) can result in a
slow response from those with command responsibility. “Hesitancy” is how one might describe the delay or lack of
actions of emergency services commanders.
In addition, within quite general situations the commander is often reasoning, consciously or subconsciously, that
taking any responsive course of action offered may actually incur further risks and make themselves culpable.
The potential consequence of their own decision-making are not lost on the commanders themselves, and in their
protocols and mindset. For example, the Chief Fire Officers Association’s “Future of Incident Command” [28] states
(in only its second paragraph), “There is considerable public, peer and legal scrutiny of the way in which incidents
are dealt with, more so today than ever before. Fire Incident Commanders are expected to work in challenging and
highly pressured situations.”
It is obvious that a combination of denial and hesitancy means that there may be a slow response, but these may
also lead to a systemic lack of agility on many occasions. The pre-prepared playbook may be followed too rigidly and
situational opportunities and advantages may be lost resulting in increased harms.
Response commanders need to be selected based on their ability to respond appropriately and assertively under
pressure. Not doing so invites a risk.
Here are three well documented yet highly distinct examples where the common denominator, the common cause,
is the failure of the response command leadership. We aver there is a pattern here.
•The Grenfell Tower Public Enquiry [29] stated that there was a leadership failure at the scene in reacting to
unfolding events and in not changing the rigid “stay put” advice to residents inside their flats. The inquiry
concluded that the London Fire Brigade’s delay in evacuating the burning building cost lives. The London Fire
Commissioner Dany Cotton told the inquiry she would not have done anything differently on the night of the
fire in June 2017 [30]. She subsequently stood down over that matter and changed her stance.
•At the 2016 second coroner’s inquests into the deaths of 96 football fans at the FA Cup semi-final at Hillsborough
in 1989, it was stated that 2,000 fans arrived late through the turn-styles and, as Gate C was opened, most of
them headed straight down a tunnel towards the already full central pens, creating the fatal crush. If the
tunnel had been closed, fans would have been diverted towards the relatively emptier side pens. The police
match commander, Ch. Supt. David Duckenfield agreed his failure to close the tunnel was the direct cause of
the deaths of 96 people. Mr Duckenfield admitted that he “froze” [34] because of the pressure he was under
[35]. After two criminal trials, on 28 November 2019, Mr Duckenfield was found not guilty of gross negligence
manslaughter [36].
8
•Prior to the Manchester Arena bombing at 10.31pm on 22 May 2017, the Greater Manchester Fire and Rescue
Service had trained for a multi-centred marauding gun attack rather than a lone-wolf suicide bombing. This
resulted in the senior officer in charge on the night, Andy Berry, holding firefighters back for two hours, instead
of them being able to carry victims to safety [38]. Berry assumed that “other components to the attack were
likely” and this “contributed to a failure to challenge and interrogate information that tended to confirm that
scenario” [39]. Clearly, there was some confirmation bias at work. Berry vetoed a fire service-police rendezvous
near Victoria Station, saying it was too close to the site of the possible ongoing attack, and instead he chose
a fire station two miles away. Berry also said he would speak to the police force duty officer, but he never got
through. Peter O’Reilly, the chief fire officer, arrived at the service headquarters at 11.50pm, and discovered
that no fire crew had been sent to the arena. Yet no fire engines were deployed for another 40 minutes. In fact,
in 2018 [40], following the Kerslake Report, Mr O’Reilly said that he will ‘always regret’ that fire crews were not
there within minutes..
Physiologically humans also have to deal with the fight-flight-freeze response [37] , the body’s natural stress reaction
to danger. It causes hormonal changes. Freezing, called reactive immobility or attentive immobility, is similar to fight-
or-flight except that you stay completely still and redtry to get ready for the next move.
Leach [41] found that ‘freezing’ behaviour was a frequently cited response by witnesses to a disaster, causing
evacuation delays (increasing the danger). This could be accounted for by temporal constraints on cognitive information
processing within a rapidly unfolding environment. He states that cognitive limitations help to explain why survival
training works.
One aspect of freezing is an apparent inability to think beyond the agreed playbook for events. Or to default to a
fixed playbook without question. Whilst the planning may ensure a minimal level of performance, rigorous obedience
may well result in lost opportunities for better outcomes in the particular evolving context. Of course the Chief Fire
Officers Association guidance [28] gives advice about the public and legal scrutiny of any leadership decision that might
override that default “safe” playbook.
Leadership and command within emergency situations may need to take on some new risks in acting in an agile
way to diminish existing risks: sometimes the greater ill is to continue with the default approach in the playbook and
the best-worst scenario might require the command to be agile and proactive in the face of evolving events.
The central issue here is not the particular details of the various incident failures, and the consequent delays, but
the question of how any person with inappropriate abilities and even a subconscious tendency to freeze might be ever
appointed to a position of operational leadership. In many emergency services senior appointments appear to be a
consequence of longish service and having the right set of contacts and influences. But these career assets actually say
nothing about a person’s ability to act quickly, or in an agile way, in the face of unannounced perilous emergencies.
Not for nothing do the special forces have a keen selection process designed to weed out people who are otherwise
gifted yet who may not perform when a critical moment comes.
There is some research claiming to show that mindfulness training [31] might be an effective cognitive tool in US
special forces commanders. It is not clear though whether this should ever become standard training/practice within
the UK emergency services [32, 33].
However it is clear that an ability to take command of fast moving, dissonant, extraordinary situations is not
necessarily something that comes with senior or long service: it is an innate human ability and it cannot be treated
as one of many required responsibilities of response force leadership.
One potential problem is with the training exercises, which are rarely unseen and may be designed to enable
subjects to learn and to pass. They are often planned by ex-staff within the existing culture and range of expectations,
rather than with a radical or opportunistic mind-set. By contrast the special services have exercises that are designed
to find out the psychological operational limits of individuals, and to determine if and how they may jeopardise their
missions and/or become a risk to colleagues. The Response to many of the NRR risks surely must suffer from this
problem. Indeed, not cogently addressing this is, in itself, a major risk for the UK.
6.2 Proliferation of Component Infrastructure With Then Unknown Issues
National risks can arise through the proliferation of structural materials and components with similar buildings and
physical infrastructure. Where cost is the driver such a phenomenon is a classic example of CATNAP (see Section
6.3 below), though the proliferation may arise for some other reasons, such as the easy availability, the ease of use,
the dominance of common supply chains, the existence of large-scale horizontal suppliers, and so on. It is usually the
case that such components are thought to be safe (for some foreseen lifetime) at the time of their deployment, and
consequently that any individual failures, as they arise, result in the need for a national emergency response.
The fact that these types of emergent event require an “emergency response from government” (see the NRR 2023
[1], page 17)) means that they cannot be classed as “chronic” (and thus cannot left out of the present NRR). They
9
result in acute events and national response emergencies.
We give three examples, followed by a generic explanation based on the theory of irreversible (hysteretic), acute,
catastrophes.
•Aluminium Composite Material (ACM) panels were commonly used for building cladding. They consists of two
skins of aluminium bonded to either side of a lightweight core of materials such as polyethylene , polyurethane,
profiled metal, or a mineral core. ACM cladding became notorious following the Grenfell Tower fire on 14 June
2017, when ACM cladding with a polyethylene core was thought to have contributed to the rapid spread of the
fire up the outside of the tower.
After the Grenfell Tower fire in June 2017, the Government established a Building Safety Programme with the
aim of “ensuring that residents of high-rise residential buildings are safe, and feel safe from the risk of fire, now
and in the future.” The full cost of national remediation work on all affected blocks is estimated at around
£15 billion. The Government announced a succession of funding commitments for the removal of cladding from
all residential buildings. In 2019 the Secretary of State at the Ministry of Housing, Communities and Local
Government (now the Department for Levelling Up, Housing and Communities, DLUHC), said the Government
would fully fund replacements of “unsafe aluminium composite material cladding on high-rise private residential
properties where building owners have failed to do so.” In 2022, the Secretary of State at DLUHC, said the
next phase of the Building Safety Fund would drive forward taking dangerous cladding off high-rise buildings,
prioritising the government’s £5.1 billion funding on the highest risk. Despite the announcement of additional
funding and measures in the Building Safety Act 2022, there are still many ongoing concerns.
•Over the late 19th century and early 20th century, asbestos was considered an ideal material for use in the
construction industry. It was an excellent fire retardant, inexpensive, and easy to use. The dangers related to
asbestos arise mainly when the fibers become airborne and are inhaled, causing illnesses, such as asbestosis and
mesothelioma, and subsequent death. Worldwide, 67 countries and territories (including those in the EU and
the UK) have banned the use of asbestos. One of the major issues relating to asbestos in civil proceedings is the
latency of asbestos-related diseases.
The UK’s Control of Asbestos Regulations, November 2006, were an amalgamation of previous sets of legislation
(Asbestos Prohibition, Asbestos Licensing and the Control of Asbestos at Work Regulations) aimed at minimising
the use and disturbance of asbestos containing materials within British workplaces.
Meanwhile, insurance companies alleged that asbestos litigation had taken too heavy a toll on insurance and
industry. Equitas, the reinsurance vehicle, which assumed Lloyd’s of London’s liabilities, argued that asbestos
claims were the “greatest single threat” to Lloyd’s of London’s existence [42].
•Recently, the long term concerns of the use of reinforced autoclaved aerated concrete (RAAC), a lightweight and
weak cementitious material used in public buildings, have come home to roost after the collapse of some building
structures within some UK schools in Summer 2023. More that 150 schools are known to contain some RAAC
(at the present time of writing) and were all closed or partially closed at extremely short term notice. RAAC
was used in schools and other public buildings from the 1950s to the mid-1990s, meaning that more or less all
instances of it are now officially deemed not fully safe. The government says expert advice has changed and that
in the 1990s it was believed that even time-expired RAAC was not a risk.
A senior building inspector said, “The risk has been known about for decades” [43]. The September 2023
emergency decision to close (or partially close) schools was triggered by incidents in which RAAC panels in
public buildings (having been thought to be safe) failed without any warning, with the final straw believed to
have been the collapse of a beam at a UK school in late August 2023 [43].
Catastrophe theory [44], a consequence of nonlinear dynamical systems theory considering the interplay between
fast and slow timescales, considers the situation where small, slow, changes in certain underlying parameters of a
nonlinear system can cause equilibria to disappear, leading to large and sudden changes of the system behaviour.
Typically these catastrophes are irreversible in the short term since there is a hysteresis effect.
Accordingly, we suggest that the occurrence of sudden hidden thresholds (in the sense of catastrophe theory), and
sudden emergent large-scale events, arising from slowly varying, chronic, situations as some underlying factors (such
as age, physical integrity, or temperature) change, should be considered more clearly in the NRR. They may not result
in steadily increasing (chronic) harms, but can result in concurrent, large scale, catastrophic failures.
6.3 CATNAP: Cheapest Available Technology Narrowly Avoiding Prosecution
Principles such as CATNAP [45, 46] result in low-cost low-quality components being deployed across a wide range of
individual facilities and situations. This “cookie cutter” tendency produces a national lack of resilience. High resilience
10
is achieved by deploying a range of alternative suppliers and solutions, where available, rather than the proliferation
of a possible common points of failure. This issue applies to physical components (within buildings, utility networks,
transport networks, and so on); to technological, IT, and communications components; and even to conceptual lines
of argument and/or justification.
The attractiveness of having ubiquitous technologies and common cost savings result in many co-dependencies and
correlations. The result is that when just once instance fails all instances must all be closed down/turned off. So, this
often creates an “All or Nothing” situation: short-term economic gains resulting in a lack of diversity of design/supply
resulting in sector-wide vulnerabilities (for example, nuclear power life-extension for a whole set of similar graphite
reactors). These are all non-independent “bets”, perhaps even one single massive aggregated bet: whole sectors may
have all of their eggs in just one basket.
A far better approach would be to ensure that there is some in-built diversity in solutions across operations.
Although this might cost a little more in the short term (they cannot all be the cheapest) it avoids the single point
of failure occurring as a result of propagating CATNAP. Diversity of solutions, methods, operational processes and
equipment is an essential response to this problem.
There could be a policy of spending slightly more to increase the diversity of underpinning technologies and
processes. This is particularly essential with fast-moving fields such as AI and data science, or in the face of explosive
medical emergencies. The national benefit is to encourage a plethora of diverse and competing solutions and to reduce
co-dependencies. Then any failures become more individual risks, rather than aggregating into larger losses, with high
costs and high national impacts. Of course, interoperability of such diverse solutions by appropriate standards has to
be a must.
Even the vocabulary of procurement and resourcing is not helpful: we talk of “supply chains” rather than “supply
webs” or “supply systems or networks”. This subconscious preconditioning implicitly ignores how real resilience
comes about though achieving multiple independent pathways (multiple supply paths) though a directed network.
Such sustainability comes from having no single point of failure or possibility of cascade failures. Furthermore, there
appears to be no mechanism for transparent debate on how much resilience we can afford, to allow benefits and dis-
benefits of alternative resilience strategies to be formulated. And the arguments used to support such spending should
be clear about what risk is being mitigated.
Whilst encompassing some risks of the type discussed in Section 6.2, the risk also subsumes potential issues arising
from single-source national procurement programmes.
7 Risk Resilience
We have seen how common cost savings (efficiencies) can be causes of fragility and a lack of resilience. Yet it is
obvious that any system resilience, avoiding single points of failure, comes at a price. Separate procurement decisions,
distributed widely will all come out the same if the solution is optimal from a costing point of view, and that is the
driver. But who bears the cost of any lack of reliance? How much resilience can we as a country afford? And how
can a portfolio approach to risk spreading be ensured? Only with a pan-sector, pan-risk, perspective can we factor in
any requirement to ensure a plurality of technologies, suppliers, and dependencies are adopted. This issue should be
owned centrally by the Cabinet Office.
In fact the reverse forces are at work. Many of the risks in the NRR pertain to facilities or processes that have
been centralised in a bid to improve efficiency of delivery of that function for the UK. Per force such systems have
lower resilience to disruption than de-centralised systems as each overall forms a ‘single-point-of-failure’. Aircraft and
space craft have multiple independent apparently wasteful duplicated sub-systems to avoid catastrophic risks. This is
not discussed in the NRR, either generically, or for any specific systems. (Perhaps because the list-based register is
designed to distribute risk narratives to lead-owners, rather than than as a tool for cross-cutting, coordinated multi-
risk management.) Even for decentralised systems and processes, the flow through the network of sub-systems and
processes encompassing that function can have ‘bottlenecks‘ and ‘critical paths‘ that significantly enhance the risk of
overall system failure. No discussions about how to avoid such phenomena is included in the document. This is not
just about assessing the particular system’s risks, but also the mitigations that may be suggested. Poorly designed
mitigations may fail themselves overall in that they are not designed to be resilient too.
Furthermore, the NRR structure, in taking a ‘one-at-a-time’ approach, does not appear to consider how various
mitigations themselves may interact with each other in amplifying or mollifying other risks.
One of the great strengths in not seeking overly automated solutions is that there are empowered human agents at
each node within systems or active within risk mitigations. The NRR does not appear to discuss the lack of human
embedding as a significant risk to the UK whether generically or for specific systems.
11
8 Recommendations for future NRRs and the Cabinet Office
1. Systems Perspective. The need for a systems perspective should be explicitly recognised. In particular, the
NRR should develop a process and designate an owner for considering likely co-occurring risks, whether these
risks are directly causal (cascades) or indirectly linked by plausible mechanisms or mere coincidence.
2. Pairwise Acute Risk Assessment. The NRR itself should be complemented by, or extended to include, a
consideration of all possible pairwise co-occurrences of the identified acute risks, highlighting those pairings that
would amplify one or the other, or would undermine the mitigation of one or the other.
3. Precursor Events. The Cabinet Office should ensure that adequate consideration is given to combinations of
acute risks and of precursor contexts (‘non-risks’), including the potential impacts of socio-economic events and
changes in legislation, which may have the unintended consequence of amplifying risks or retarding responses.
4. Level of Influence Across Government. The Cabinet Office should ensure that national risks and resilience
issue is led at a sufficiently senior level to direct and influence the risk-specific lead departments to ensure that
they liaise with other departments with responsibilities of related risks and subsidiary consequences.
5. Resilience Audit. The NRR should document where a lack of national resilience may contribute to the impacts
of certain individual acute risks.
6. Cost of Resilience Building. The Cabinet Office and HM Treasury should ensure that the cost of resilience-
building measures is properly acknowledged and owned by all public procurement processes considering common
solutions (common sources, technologies, and operations) in parallel or series. A portfolio view, across silos,
needs to be present.
7. Public Understanding. There should be a national programme to improve greatly the understanding of risk
and risk-response issues within the general public and the media.
8. Communications. There should be up-to-date acute risk communication strategies to share transparent intel-
ligence, quantitative assessments, and risk comparisons; and to inhibit disinformation and fake news, including
all social media channels
9. Risk Response Leadership. The human ability to command the response to acute rolling risks or to extreme
emergency situations, requires a specific type of individual and it is not a given for those with wide experience or in
high office (including political, organisational, and civil service leaders; and expert advisors), who may be highly
effective during business-as-usual scenarios. A national programme should manage training exercises designed
to uncover the psychological capabilities and limitations of those undergoing response command selection and
act accordingly to remediate this.
10. Emergence of Black Swans Besides the regular NRR review cycle of acute risks and possibly co-occurring
risks (cascades), there should be an active scanning and alert system for unforeseen issues ‘boiling out’ quickly
from political, technological, economic or societal changes.
Declarations
All opinions expressed here are personal to the co-authors and do not in any way represent the official position of their
affiliations. All authors contributed to the document. Grindrod is the lead contact for follow-up.
Data Availability Statement
There was no new data generated or deployed within this research. The authors confirm that the data supporting the
findings of this study are available within the article and the references cited.
12
References
[1] Cabinet Office, National Risk Register 2023, Published 3 August 2023, https://assets.publishing.service.gov.
uk/government/uploads/system/uploads/attachment data/file/1175834/2023 NATIONAL RISK REGISTER
NRR.pdf.
[2] Cabinet Office, National Risk Register Of Civil Emergencies 2017, September 2017, https:
//assets.publishing.service.gov.uk/government/uploads/system/uploads/attachment data/file/644968/
UK National Risk Register 2017.pdf.
[3] Goriely, A., and Grindrod, P. (2021). Lessons to be Learned from the Covid-19 Experience in the UK
February 2021, https://www.researchgate.net/publication/349645530 Lessons to be Learned from the Covid-19
Experience in the UK.
[4] The UK Covid-19 Inquiry, Covid-19 Inquiry Terms of Reference https://covid19.public-inquiry.uk/wp-content/
uploads/2023/05/Covid-19-Inquiry-Terms-of-Reference-Final-2.pdf.
[5] Sumption, J. (22023). Why the Covid inquiry is a farce, Comment, The Sunday Times, Sunday November 5th
2023 https://www.thetimes.co.uk/article/covid-19-inquiry-farce-jonathan-sumption-7vgtxp037.
[6] Office for Nuclear Regulation, An agency of HSE (2012), “Stress Tests” for UK non-Power Generating Nuclear
Facilities, Final Report May 2012 https://www.onr.org.uk/fukushima/ngpf-report.pdf.
[7] Regulatory guide on preparedness (VAL), STUK https://stuk.fi/en/val-guides.
[8] NEA Expert Group on development of a Holistic Process for Decision Making on Decommissioning and Manage-
ment of Complex Sites https://tinyurl.com/2b4juub4.
[9] IAEA (2023). Relationship Between Safety and Sustainability Explored at IAEA Conference https://tinyurl.com/
3beuh2hf.
[10] Tetlock, P.E., and Gardner, D. (2016). Superforecasting: The Art and Science of Prediction, Random House
Books; 1st edition (7 April 2016).
[11] Committee On Science And Technology, House Of Representatives 99th Congress, 2nd Session, House
Report99-1016 INVESTIGATION OF THE CHALLENGER ACCIDENT, https://www.govinfo.gov/content/
pkg/GPO-CRPT-99hrpt1016/pdf/GPO-CRPT-99hrpt1016.pdf
[12] Grindrod, P. (2022), Economic and Social Resilience and Vulnerability, Discussion paper for workshop at the
Changing Character of War Centre, Pembroke College, Oxford, July 2022, Researchgate https://tinyurl.com/
ydjzn4bu.
[13] UK Government (2023), Chair’s Summary of the AI Safety Summit 2023, Bletchley
Parkhttps://www.gov.uk/government/publications/ai-safety-summit-2023-chairs-statement-2-november/
chairs-summary-of-the-ai-safety-summit-2023-bletchley-park.
[14] UK Government (2023), Capabilities and risks from frontier AI: A discussion paper on the need
for further research into AI risk, https://assets.publishing.service.gov.uk/media/65395abae6c968000daa9b25/
frontier-ai-capabilities-risks-report.pdf .
[15] Taleb, N.N. (22 April 2007). The Black Swan: Chapter 1: The Impact of the Highly Improbable. The New York
Times. Retrieved 20 January 2016.
[16] Towards an All-Hazards Approach to Emergency Preparedness and Response: Lessons
Learnt from Non-Nuclear Events. NEA No. 7308. Nuclear Energy Agency Organisa-
tion For Economic Co-operation and Development, https://www.oecd-nea.org/jcms/pl 15010/
towards-an-all-hazards-approach-to-emergency-preparedness-and-response?details=true.
[17] Balashevska, Y., Chala, M., Ivanov, Z., Myshkovska, A., Shevchenko, I, Pecherytsia, O., Yesipenko, Y., Siegen, K.,
Jova Sed, L., Smith, G.M. (2023) Preliminary assessment of the radiological consequences of the hostile military
occupation of the Chornobyl Exclusion Zone, Society for Radiological Protection, IOP, Journal of Radiological
Protection, Volume 43, No. 3, https://iopscience.iop.org/article/10.1088/1361-6498/acf8d0.
13
[18] Spiegelhalter, D. and Freeman, A. (2021). Written evidence (RSK0057), 28 January 2021, Winton Centre for Risk
and Evidence Communication, University of Cambridge, https://committees.parliament.uk/writtenevidence/
21909/pdf/.
[19] Zehra Zaidi, R. (2018). Beyond the Sendai indicators: Application of a cascading risk lens for the improvement
of loss data indicators for slow-onset hazards and small-scale disasters, International Journal of Disaster Risk
Reduction, Volume 30, Part B, pp306-314, ISSN 2212-4209, https://doi.org/10.1016/j.ijdrr.2018.03.022.
[20] Finnland Ministry of the Interior, Helsinki (2023). National risk assessment 2023, https://intermin.fi/en/
rescue-services/preparedness/national-risk-assessment.
[21] Hewitt, K. (1997), Regions of Risk: A Geographical Introduction to Disasters. Routledge.
[22] IPCC (2012) Managing the Risks of Extreme Events and Disasters to Advance Climate Change Adaptation. A
Special Report of Working Groups I and II of the Intergovernmental Panel on Climate Change (IPCC) [Field, C.
B., V. Barros, T. F. Stocker, D. Qin, D. J. Dokken, K. Cambridge, United Kingdom and New York, NY, USA:
Cambridge University Press.
[23] Wisner, P., Blaikie, P., Cannon, T., David, I. (2004). At Risk: Natural Hazards (2004), People’s Vulnerability
and Disasters, Routledge.
[24] Nichols W., and Hewston, R. (2022), World ‘unprepared’ for magnitude of cascading climate
risks, Environmental Risk Outlook 2022, Verisk Maplecroft https://www.maplecroft.com/insights/analysis/
world-unprepared-for-magnitude-of-cascading-climate-risks/.
[25] Mizrahi, S. (202). Cascading disasters, information cascades and continuous time models of domino effects. Int J
Disaster Risk Reduct. 2020 Oct;49:101672. doi: 10.1016/j.ijdrr.2020.101672. Epub 2020 May 21. PMID: 32501417;
PMCID: PMC7240265.
[26] Spiegelhalter, D., Risk and Uncertainty Communication (March 2017). Annual Review of Statistics and Its
Application, Vol. 4, Issue 1, pp. 31-60, 2017, https://ssrn.com/abstract=2937730orhttp://dx.doi.org/10.1146/
annurev-statistics-010814-020148.
[27] This is Your Brain on Emergencies, November 7, 2016 by Jerzell Black, Operation Coordinator,
CDC Office of Safety, Security, and Asset Management https://blogs.cdc.gov/publichealthmatters/2016/11/
this-is-your-brain-on-emergencies/.
[28] Chief Fire Officers Association, Future of Incident Command https://www.nationalfirechiefs.org.uk/write/
MediaUploads/NFCC%20Guidance%20publications/Operations/CFOA Incident Command future.pdf
[29] Booth, R. (2019). London fire chief Dany Cotton resigns after Grenfell criti-
cism, The Guardian, 6th Dec. 2019, https://www.theguardian.com/uk-news/2019/dec/06/
london-fire-chief-dany-cotton-resigns-after-grenfell-criticism.
[30] Bulman, M. (2019). Grenfell fire chief U-turns on comments she ‘wouldn’t have done anything
differently’, The Independent, 5 November 2019, https://www.independent.co.uk/news/uk/home-news/
grenfell-fire-chief-dany-cotton-london-assembly-a9185866.html.
[31] Zanesco, A.P., Denkova, E., Rogers, S.L., MacNulty, W.K. and Jha, A.P. (2018), Mindfulness training as cognitive
training in high-demand cohorts: An initial study in elite military service members; Progress in Brain Research,
244:323-354, Epub 2018 Nov 27https://doi.org/10.1016/bs.pbr.2018.10.001.
[32] Fitzhugh, H., Michaelides, G., Connolly, S., and Daniels, K. (2019), Mindfulness in policing: A randomized
controlled trial of two online mindfulness resources across five forces in England and Wales. College of Policing,
London.
[33] Fitzhugh, H., Michaelides, G., Daniels, K., Connolly, S., and Nasamu, E. (2023). Mindfulness for Performance
and Wellbeing in the Police: Linking Individual and Organizational Outcomes. Review of Public Personnel Ad-
ministration, 0(0), https://doi.org/10.1177/0734371X231155794
[34] Conn, D. (2019). Hillsborough: chief officer said he ’froze’ after opening gate, The Guardian, Thu 28 Feb 2019,
https://www.theguardian.com/uk-news/2019/feb/28/hillsborough-chief-officer-said-he-froze-after-opening-gate.
14
[35] Turner, R. (2016). Hillsborough disaster: Five key mistakes, BBC News Online, 26 April 2016, https://www.bbc.
co.uk/news/uk-england-merseyside-35462767.
[36] BBC News, “Hillsborough police chief cleared of manslaughter”, 28 November 2019, https://www.bbc.co.uk/
news/uk-england-50592077.
[37] Nunez, K. (2023). Fight, Flight, Freeze: What This Response Means, Updated on February 10, 2023, https:
//www.healthline.com/health/mental-health/fight-flight-freeze#takeaway.
[38] Gardham,D. (2020). Manchester Arena bombing: fire service ‘trained for wrong kind of
terrorist attack, The Times, October 05 2020, https://www.thetimes.co.uk/edition/news/
manchester-bombing-fire-service-trained-for-wrong-kind-of-terrorist-attack-stn3v0q0z.
[39] Britton, P. and Carter, H. (2020). Manchester Arena bombing inquiry: Day two recap, Manch-
ester Evening News, 9 SEP 2020, https://www.manchestereveningnews.co.uk/news/greater-manchester-news/
manchester-arena-bombing-inquiry-live-18895613
[40] Hurst, P. and Slater, C. (2018). Former Greater Manchester fire chief blames police for brigade’s delayed response
to the Arena attack, Manchester Evening News, 20 AUG 2018, https://www.manchestereveningnews.co.uk/news/
greater-manchester-news/former-greater-manchester-fire-chief-15053120
[41] Leach, J. (2004). People ‘Freeze’ in an Emergency: Temporal and Cognitive Constraints on Survival Re-
sponses Aviation, Space, and Environmental Medicine, Vol. 75, No. 6, June 2004 https://www.hptinstitute.
com/wp-content/uploads/2014/01/Why-People-Freeze.pdf.
[42] Dunne, H. (2002). Equitas warns on asbestosis. The Daily Telegraph, 22 June 2002, Retrieved 18 November 2017.
[43] Walker, P. and Booth, R. (2023), How long has Raac in schools been a concern and what hap-
pens now?, The Guradian, 1st Sept 2023, https://www.theguardian.com/education/2023/sep/01/
how-long-has-raac-in-schools-been-a-concern-and-what-happens-now#:∼:text=It%20was%20put%20into%
20schools,Raac%20was%20not%20a%20risk..
[44] Poston, T. and Stewart, I. (1998). Catastrophe: Theory and Its Applications, New York: Dover ISBN 0-486-
69271-X
[45] Highley, S., Health and Safety Communication – Abbreviations, Acronyms and
Other Confusions, Hastam, gen eral Health and Safety, hhttps://www.hastam.co.uk/
health-and-safety-communication-abbreviations-acronyms-and-other-confusions/.
[46] MOD, Acronyms and Abbreviations, https://assets.publishing.service.gov.uk/government/uploads/system/
uploads/attachment data/file/227048/acronyms and abbreviations dec08.pdf.
[47] Infected Blood Enquiry (2022), Expert Report to the Infected Blood Inquiry: Statistics, September
2022, https://www.infectedbloodinquiry.org.uk/sites/default/files/documents/Expert%20Report%20to%20the%
20Infected%20Blood%20Inquiry%20-%20Statistics.pdf .
[48] Technology Readiness Levels (2014-2015), HORIZON 2020 – WORK PROGRAMME 2014-2015 Gen-
eral Annexes, G, https://ec.europa.eu/research/participants/data/ref/h2020/wp/2014 2015/annexes/
h2020-wp1415-annex-g-trl en.pdf
15
