Leadership Analysis — Campaign for Responsible Leadership in NHS Wales

DHCW's failures follow patterns so well-documented in leadership science that researchers have given them precise names. This is not interpretation — it is diagnosis.

What follows draws on six established frameworks from organisational psychology and management science. Each one describes a specific dysfunction. Each one maps directly onto the documented evidence at DHCW. Taken together, they explain why nine programmes are failing simultaneously under the same leadership — and why no amount of additional funding, restructuring, or ministerial intervention will fix the problem until the leadership itself changes.

These are not obscure theories. They are taught in every serious business school in the world, referenced in tens of thousands of peer-reviewed papers, and used by organisations from Google to the US military to diagnose precisely the kind of failure we see at DHCW. The leaders of a national digital health organisation should know them. The fact that every pattern described below is present — simultaneously, in the same organisation — tells you everything you need to know.

1. The Destruction of Psychological Safety

The theory

In 1999, Professor Amy Edmondson of Harvard Business School published a landmark study examining why some hospital nursing teams reported more medication errors than others. The surprising finding was that the best-performing teams reported more errors, not fewer. The reason: they worked in an environment where it was safe to admit mistakes. Teams that reported fewer errors were not making fewer — they were hiding them (Edmondson, 1999).

From this insight, Edmondson developed the concept of psychological safety — the shared belief within a team that it is safe to take interpersonal risks, speak up with questions, raise concerns, admit errors, and challenge the status quo without fear of punishment or humiliation. Her subsequent two decades of research, culminating in The Fearless Organization (Edmondson, 2018), demonstrated that psychological safety is not a "nice to have" — it is the foundational condition for learning, innovation, and the prevention of catastrophic failure.

The evidence is overwhelming. In 2015, Google conducted "Project Aristotle," a two-year study of 180 teams across the company to identify what made some teams dramatically more effective than others. They examined over 250 attributes. The single most important factor — by a significant margin — was psychological safety (Duhigg, 2016; Rozovsky, 2015). Teams where members felt safe to speak up outperformed teams of individually brilliant people who did not. Intelligence, experience, and technical skill were all secondary to whether team members believed they could tell the truth without being punished for it.

This finding has been replicated across healthcare settings specifically. Edmondson and colleagues have shown that psychological safety in clinical teams directly affects whether staff report safety incidents, raise concerns about protocols, and challenge senior colleagues when they believe a patient is at risk (Edmondson, 2004; Nembhard & Edmondson, 2006). In healthcare, the absence of psychological safety does not merely reduce performance — it kills people.

The evidence at DHCW

In November 2018, the Senedd's Public Accounts Committee (PAC) published its investigation into DHCW's predecessor organisation, NWIS. The committee's findings on organisational culture were devastating:

The culture was "the antithesis of open." Staff were "reluctant to be critical on the record." The committee felt it was receiving "pre-prepared lines" rather than honest testimony.

The committee explicitly warned that this culture of silence "may be masking wider and deeper problems." They were right. Every problem they feared has since materialised — and been confirmed by the Welsh Government's own Level 3 escalation.

DHCW was created in April 2021 explicitly to provide a fresh start. But the same leaders remained in post. The CEO of the new organisation, Helen Thomas, had been the Deputy CEO of the old one. The culture did not change — it was perfected.

We are aware of at least two senior employees — both holding significant technical leadership roles — who were allegedly dismissed after raising concerns about programme failures and organisational dysfunction. In one case, the employee held the role of Chief DevOps Officer and was responsible for critical national infrastructure. In both cases, the employees' concerns were subsequently validated by the very government intervention that their dismissals were, it is alleged, designed to prevent. In both cases, their roles were downgraded or eliminated after their departure.

DHCW adopted the Welsh Government's "Speaking Up Safely" framework, which guarantees protection for staff who raise concerns. In practice, the organisation publishes zero data on how many whistleblowing disclosures it has received, what they concerned, or what happened to the people who made them. Zero disciplinary data. Zero leavers analysis. The policy exists on paper. Its violation is systematic.

Edmondson's framework predicts exactly this pattern. When leaders punish those who speak up, the message radiates through the entire organisation: truth-telling is dangerous. The result is not that problems disappear — it is that problems become invisible to the people who need to know about them. Information stops flowing upward. Leaders make decisions based on the optimistic picture their subordinates have learned to present. Programmes fail in slow motion while everyone pretends they are on track. By the time the truth becomes undeniable, the damage is catastrophic — and the leaders who created the silence blame the people below them for not speaking up sooner.

This is precisely what happened at DHCW. Nine programmes were failing simultaneously while the CEO presented optimistic timelines to the Senedd. The staff who knew the truth had learned what happens to people who say it out loud.

2. The Toxic Triangle

The theory

Professor Jean Lipman-Blumen of Claremont Graduate University spent a career studying a question that most leadership research ignores: why do organisations tolerate — and even embrace — leaders who damage them? Her book The Allure of Toxic Leaders (Lipman-Blumen, 2005) argued that destructive leadership persists not because organisations fail to notice it, but because multiple forces conspire to sustain it. Toxic leaders offer certainty in uncertain times, project confidence that others find reassuring, and build networks of personal loyalty that make challenging them dangerous.

Art Padilla and colleagues at North Carolina State University extended this analysis with the concept of the Toxic Triangle (Padilla, Hogan & Kaiser, 2007). Their framework identifies three elements that must be present simultaneously for destructive leadership to take hold and persist:

Destructive leaders — individuals who prioritise self-interest over organisational mission, who demand personal loyalty, who control information, and who retaliate against perceived threats to their position.
Susceptible followers — two types. "Conformers" comply out of fear, unmet needs for security, or identification with the leader. "Colluders" actively participate because they benefit personally from the leader's patronage — through promotions, protection, or access to resources they would not otherwise receive.
Conducive environments — organisational and institutional contexts characterised by instability, lack of oversight, perceived threats, and cultural norms that discourage dissent. Bureaucratic complexity and opaque governance structures are particularly conducive because they make it difficult for external observers to see what is happening.

The Toxic Triangle model is important because it explains why removing a destructive leader is necessary but not sufficient. If the follower dynamics and environmental conditions remain unchanged, a new leader will face the same pressures — or the same patterns will reassert themselves under different names.

The evidence at DHCW

Each vertex of the Toxic Triangle is clearly identifiable at DHCW.

Destructive leaders. The CEO and her inner circle have presided over the escalation of all nine major programmes to the highest tier of government intervention. When asked at the January 2026 public accountability meeting what return the public had received on hundreds of millions in expenditure, the CEO admitted: "We don't have an ROI on all of our investments." She then compared measuring digital returns to measuring the value of electricity — arguing that accountability is inherently impossible. Meanwhile, the CEO accumulated personal credentials — a BCS fellowship, an honorary professorship, a "Digital CEO of the Year" award — within an eighteen-month window timed precisely to her appointment. Resources flow toward the CEO's flagship programmes while critical national infrastructure starves for investment. Ifan Evans, who wrote the Welsh Government's national digital health strategy, was subsequently appointed to the DHCW role responsible for implementing it — marking his own homework. These are the behaviours of leaders who prioritise self-preservation and self-advancement over the mission they were appointed to serve.

Susceptible followers. A layer of middle and senior managers at DHCW knows what is happening. They attend the programme boards where missed deadlines are discussed. They see the gap between what is reported upward and what is true. They remain silent — not because they are bad people, but because they have witnessed what happens to those who speak up. The alleged dismissal of senior technologists for raising concerns sends a message that is impossible to misinterpret. Those who remain have made a rational calculation: silence is safer than truth. Some go further — actively participating in the narrative of success because their own promotions and continued employment depend on the current leadership remaining in place. Padilla's framework calls these individuals "colluders," and their presence explains why dysfunction persists even when individual leaders are challenged.

Conducive environments. DHCW operates within the NHS Wales governance structure — a system of extraordinary bureaucratic complexity and limited external visibility. Senior directors maintain no public profile whatsoever. Off-payroll workers making operational decisions are unnamed in any public document. Contract values are unpublished. Whistleblowing data is suppressed. Even the independent digital expert appointed by the Welsh Government to assess DHCW under Level 3 escalation has never been publicly identified. The opacity is not accidental — it is structural, and it provides precisely the "conducive environment" that the Toxic Triangle requires. No external observer can see enough to intervene effectively.

Lipman-Blumen observed that toxic leaders survive because the cost of challenging them is always borne by the challenger, while the cost of tolerating them is diffused across the entire organisation. At DHCW, the people who bear the cost of leadership failure are the patients of Wales — people who will never know that their delayed treatment, lost records, or missed screening invitation was the downstream consequence of a specific leadership decision made by a specific person in a specific meeting.

3. Organisational Silence

The theory

In 2000, Professors Elizabeth Morrison and Frances Milliken published a seminal paper examining why organisations systematically fail to hear what their own members know. They called the phenomenon organisational silence — a collective-level condition where employees withhold information about problems, concerns, and opportunities for improvement because they believe speaking up is pointless or dangerous (Morrison & Milliken, 2000).

Morrison and Milliken identified several forces that create organisational silence. Managers fear negative feedback and surround themselves with people who agree with them. Top management teams develop shared beliefs that employees are self-interested and cannot be trusted. Organisational structures centralise decision-making and create few formal channels for upward communication. The result is a self-reinforcing system: silence begets ignorance, ignorance begets poor decisions, poor decisions create more problems, and the problems are met with more silence.

Van Dyne, Ang and Botero (2003) refined this analysis by distinguishing three forms of employee silence:

Acquiescent silence — withholding information because one feels resigned and believes nothing will change regardless. "There's no point. They won't listen."
Defensive silence — withholding information out of fear of personal consequences. "If I speak up, they'll come after me."
Prosocial silence — withholding information to protect others or to avoid causing problems for colleagues. "If I report this, my team will suffer."

All three forms have the same effect: critical information fails to reach decision-makers. But they require different interventions. Acquiescent silence requires demonstrating that speaking up leads to change. Defensive silence requires guaranteeing that speaking up is safe. Prosocial silence requires building systems where raising concerns helps rather than harms colleagues.

The evidence at DHCW

All three forms of silence are evident at DHCW.

Defensive silence is the most visible. The alleged dismissals of senior technologists for raising concerns about programme failures represent the ultimate demonstration that speaking up carries career-ending consequences. When an organisation removes the people who were proven right — and does so in a manner that is visible to remaining staff — it creates a chilling effect that no amount of "Speaking Up Safely" posters can counteract. The message is not written in any policy document. It is written in what happened to the people who used those policies.

Acquiescent silence is evident in the institutional response to repeated failure. DHCW has cycled through the same patterns across multiple programmes — WCCIS, LINC, OpenEyes, the NHS Wales App — each time repeating the same mistakes. Initial optimism, missed deadlines, programme resets, rebrands, and eventual government intervention. Staff who have witnessed this cycle multiple times learn that the organisation does not change regardless of what they say. The rational response to institutional indifference is withdrawal.

Prosocial silence is perhaps the most insidious form at DHCW. Many staff members are deeply committed to the NHS. They stay not because of the leadership but despite it — because they believe in the mission and care about their colleagues. These individuals may withhold concerns about leadership failures because they fear that public exposure will harm the organisation and, by extension, harm the patients they serve. They protect the institution that is harming them out of loyalty to its purpose. This is exactly the dynamic that Morrison and Milliken described: silence born of good intentions producing catastrophic outcomes.

The data gaps tell their own story. DHCW publishes zero whistleblowing data — no figures on disclosures received, topics raised, outcomes, or consequences. Zero disciplinary data. Zero leavers analysis. Zero exit interview summaries. An organisation that genuinely wanted to hear from its staff would measure whether they were speaking up. An organisation that wanted silence would measure nothing — because measurement creates accountability, and accountability creates pressure to act.

Morrison and Milliken's central warning is this: by the time organisational silence becomes visible to external observers, the damage is usually far advanced. The information that leaders needed to prevent failure was available within the organisation. It was simply never allowed to reach them — or, worse, it reached them and was punished. At DHCW, the Welsh Government's Level 3 escalation confirmed problems that staff had been raising internally for years. The information existed. The organisation chose not to hear it.

4. Level 5 Leadership — What Good Looks Like

The theory

In 2001, Jim Collins published Good to Great, one of the most influential management studies of the past fifty years. Collins and his research team spent five years analysing companies that made the transition from sustained mediocrity to sustained excellence, comparing them with carefully matched companies that attempted the same transition and failed.

Their most surprising finding concerned leadership. The executives who led the most successful transformations were not the charismatic, self-promoting, empire-building leaders that business culture celebrates. They were what Collins called Level 5 Leaders — individuals characterised by a paradoxical combination of personal humility and fierce professional will (Collins, 2001).

Level 5 Leaders channel their ambition into the organisation and its mission, not into themselves. They give credit to their teams when things go well and take personal responsibility when things go wrong. They make decisions based on evidence rather than ego. They surround themselves with the best people they can find — including people who disagree with them — because they care more about getting the right answer than about being right. They build organisations that succeed because of their systems and culture, not because of the leader's personal brilliance.

Collins contrasted Level 5 Leaders with what he called Level 4 Leaders — talented, driven individuals who build high performance but orient it around themselves. Level 4 Leaders take credit for success, blame others for failure, surround themselves with loyal supporters rather than talented challengers, and leave organisations that cannot function without them. They may achieve impressive short-term results, but they leave behind institutions weaker than they found them.

The distinction matters because it is observable. You can tell the difference between Level 5 and Level 4 leadership not from what leaders say about themselves, but from what happens to the organisation under their stewardship. Level 5 Leaders leave behind thriving institutions. Level 4 Leaders — and below — leave behind dependency, decline, and dysfunction.

The evidence at DHCW

DHCW's leadership exhibits the precise inversion of every Level 5 characteristic Collins identified.

Where Level 5 Leaders channel ambition into the mission, DHCW's leaders channel resources into themselves. The CEO accumulated personal credentials — fellowship, honorary professorship, industry award — timed to her appointment. Critical leadership positions were filled with allies and proteges rather than through genuinely competitive processes. The workforce grew by 25% while programme delivery deteriorated across every major area. When asked what return the public had received on hundreds of millions in expenditure, the CEO could cite only £0.5 million in "equivalent savings" — not cash — from a single system. The organisation grew larger. Its leaders grew more decorated. Its patients received less.

Where Level 5 Leaders take responsibility for failure, DHCW's leaders blame others. When the CEO was summoned to the Senedd Health Committee in May 2025 over the OpenEyes programme — seven years behind its original timeline — she acknowledged that engagement with health boards "could have worked a lot better." The passive construction is telling. It could have worked better — as though the failure happened to the CEO rather than under her direction. At the January 2026 accountability meeting, DHCW leadership attributed programme delays to supplier changes, health board readiness, and funding constraints — never to their own decisions, judgement, or competence.

Where Level 5 Leaders surround themselves with the best people, DHCW's leaders remove them. The alleged dismissals of senior technologists — people who held significant technical leadership roles and whose concerns were subsequently validated by government intervention — represent the opposite of Level 5 behaviour. Collins specifically found that great leaders "first get the right people on the bus" and create environments where truth is heard. DHCW's leaders, it is alleged, removed the right people from the bus for telling the truth.

Where Level 5 Leaders build institutions that outlast them, DHCW's leaders build personal dependencies. Senior directors maintain no public profile. Institutional knowledge concentrates in a small inner circle. Decision-making authority is centralised rather than distributed. The organisation is structured so that it cannot function without the current leadership — not because the current leadership is irreplaceable, but because they have made themselves so by concentrating power, removing challengers, and ensuring that no alternative leadership base exists within the organisation.

The contrast with what good digital health leadership looks like is available in the same sector. NHS England's Transformation Directorate recruited leaders from organisations including Deloitte Digital, Jaguar Land Rover, and the Government Digital Service — people with demonstrated delivery track records at scale, independently verifiable credentials, and no prior relationship with the existing leadership. They were hired to challenge the status quo, not to reinforce it. Collins would recognise the difference immediately. DHCW's leaders would not.

5. The Theft of Purpose and Pride

The theory

In the 1970s and 1980s, Professors Edward Deci and Richard Ryan of the University of Rochester developed Self-Determination Theory (SDT), now one of the most extensively validated frameworks in motivational psychology. SDT proposes that human beings have three fundamental psychological needs that must be satisfied for wellbeing, engagement, and sustained high performance (Deci & Ryan, 1985; Ryan & Deci, 2000):

Autonomy — the need to feel that one's actions are self-directed and meaningful, that one has genuine choice in how work is performed.
Competence — the need to feel effective and capable, to master challenges, and to see the results of one's skills applied to meaningful problems.
Relatedness — the need to feel connected to others, to belong to a community, and to trust the people one works with and for.

When these three needs are met, people are intrinsically motivated — they work hard because the work itself matters to them. When these needs are frustrated, motivation collapses. People disengage, burn out, or leave. Decades of research across healthcare, education, military, and corporate settings have confirmed that the satisfaction of autonomy, competence, and relatedness predicts job satisfaction, performance, wellbeing, and retention more reliably than pay, status, or any other external factor (Gagné & Deci, 2005).

This matters enormously in the public sector, where pay is typically below market rates and people choose to work because they believe in the mission. Public sector motivation is disproportionately intrinsic. When leaders destroy the conditions for intrinsic motivation, they destroy the very thing that makes public service work.

The evidence at DHCW

DHCW's leadership has systematically undermined all three of Deci and Ryan's fundamental needs.

Autonomy has been destroyed. Technical decisions are overridden by leaders without technical expertise. Staff report through Glassdoor reviews — DHCW's own public-facing reputation — that decision-making is centralised, that professional judgement is disregarded, and that directions change without explanation or consultation. When the organisation's most senior technologists raised concerns about programme direction, they were not engaged in dialogue — they were, it is alleged, removed. The message to remaining staff is unambiguous: your professional judgement is not wanted. Do as you are told.

Competence is being wasted. DHCW employs over 1,100 people, many of whom are skilled technologists, data professionals, and healthcare informaticians who joined because they wanted to build something that mattered. These individuals watch their skills wasted on programmes that cycle through failure, reset, and failure again — WCCIS to Connecting Care, LINC to its third iteration, OpenEyes through seven years of missed deadlines. Their expertise is not deployed against the hardest problems; it is consumed by rework, political manoeuvring, and the administrative overhead of programmes that were poorly conceived and poorly governed. The experience of competence — of applying one's skills to solve real problems and seeing the results — is denied to them. They are not allowed to succeed.

Relatedness has been poisoned. Trust is the foundation of relatedness, and trust at DHCW has been systematically destroyed. When colleagues are dismissed for speaking up, every remaining relationship is contaminated by the knowledge that honesty is dangerous. When promotions go to allies rather than to the most capable, the social contract that binds a team together — the belief that effort and excellence are recognised and rewarded — is broken. When leaders take credit for what little goes right and blame staff for what goes wrong, the bond between leader and team — the most important relationship in any organisation — becomes adversarial rather than supportive.

The consequence, predicted precisely by Self-Determination Theory, is what Deci and Ryan call "amotivation" — a state beyond mere disengagement where people no longer see any connection between their effort and meaningful outcomes. They go through the motions. They stop caring — not because they are lazy, but because the organisation has taught them that caring is futile and dangerous. The people who cannot tolerate this state leave. The people who remain adapt to it. Neither outcome serves the patients of Wales.

This is what makes DHCW's leadership failures so damaging beyond the immediate programme costs. They are not merely wasting money. They are destroying the motivation of the people who would need to deliver any future improvement. Even if new leaders arrived tomorrow, they would inherit a workforce whose intrinsic motivation — the single most valuable resource in any public service organisation — has been systematically damaged by years of autonomy denial, competence waste, and trust destruction. Rebuilding it will take years.

6. The Learning-Disabled Organisation

The theory

In 1990, Peter Senge of the MIT Sloan School of Management published The Fifth Discipline, introducing the concept of the learning organisation — an institution capable of continuously adapting, improving, and transforming itself through the collective learning of its members. More importantly for our purposes, Senge identified seven "learning disabilities" that prevent organisations from learning, even when the information they need is available (Senge, 1990).

Several of these disabilities are directly relevant:

"I am my position." People define themselves by their role and title rather than by the purpose they serve. They focus on their own tasks and lose sight of how their work connects to — and affects — the broader system.
"The enemy is out there." When things go wrong, the blame is always externalised. It is the supplier's fault. The health boards were not ready. The funding was insufficient. The previous organisation left a legacy of problems. There is always an external explanation that protects internal leadership from accountability.
"The illusion of taking charge." Leaders respond to problems with visible activity — restructures, rebrands, new programme names, strategy documents — that creates the appearance of action without addressing root causes. Proactiveness is confused with reactiveness dressed in the language of initiative.
"The fixation on events." Attention focuses on individual incidents — a missed deadline, a system outage, a critical news story — rather than on the underlying patterns that produce them. Each event is treated as isolated rather than symptomatic.
"The parable of the boiled frog." Gradual deterioration does not trigger a response because each increment of decline seems small enough to tolerate. By the time the cumulative damage is undeniable, it is too late for incremental correction.
"The delusion of learning from experience." Organisations believe they learn from their mistakes, but in reality they repeat them because the consequences of today's decisions are separated in time and space from the decisions themselves. The leaders who made the decisions are never present when the consequences arrive — and if they are, they attribute the consequences to other causes.

Senge argued that these disabilities are systemic, not individual. They are properties of the organisation's structure, incentives, and culture — and they can only be addressed through fundamental changes to how the organisation thinks, measures, and governs itself.

Chris Argyris, the Harvard organisational theorist whose work influenced Senge, distinguished between single-loop learning (correcting errors within existing assumptions) and double-loop learning (questioning and revising the assumptions themselves). Organisations stuck in single-loop learning repeat the same mistakes in different forms because they never examine the beliefs and structures that produce those mistakes (Argyris, 1977; Argyris & Schön, 1978). They do more of the same, harder and faster, and wonder why the results do not change.

The evidence at DHCW

DHCW exhibits every learning disability Senge described.

"The enemy is out there." At the January 2026 accountability meeting, DHCW leadership attributed programme failures to supplier implementation plan changes, health board readiness issues, workforce constraints, and funding gaps. At the Senedd Health Committee appearance in May 2025, the CEO acknowledged that engagement with health boards "could have worked a lot better." In every public forum, responsibility is located outside the leadership team. The system failed them. They did not fail the system.

"The illusion of taking charge." WCCIS was rebranded to "Connecting Care" after a programme reset. NWIS was rebranded to DHCW. Programme structures are reorganised. Strategy documents are commissioned — nine months and two reports from Channel 3 Consulting for the National Target Architecture. The activity is visible. The outcomes do not change.

"The parable of the boiled frog." The slide from NWIS to DHCW to Level 3 escalation took eight years. At no point was there a single catastrophic event. Instead, there was a steady accumulation of missed deadlines, rising costs, deteriorating systems, and departing talent — each individual step tolerable, the cumulative trajectory devastating. The Welsh Government's escalation in March 2025 was not triggered by a sudden crisis. It was triggered by the belated recognition that chronic underperformance had become the permanent state.

"The delusion of learning from experience." DHCW has presided over programme failures that repeat identical patterns. WCCIS (launched 2015): initial ambition, missed targets, organisations refusing to participate, independent review, programme reset. OpenEyes (funded 2020): initial ambition, two missed national deadlines, CEO summoned to Senedd. LINC (started 2017): initial ambition, original supplier contract terminated, revised approach, eight years for one lab partially live. NHS Wales App (conceived 2021-2022): initial ambition, "mired in delay," Cabinet Secretary unable to identify a critical path. The pattern is identical. The lessons are never learned. No post-implementation reviews are published. No root cause analyses are made available. No institutional memory is built.

This is not single-loop learning failing to reach double-loop learning. This is the absence of learning altogether. The organisation does not ask why its programmes fail because asking that question would produce answers that threaten the leadership. So the question is never asked, the answers are never produced, and the next programme repeats the mistakes of the last one.

Senge warned that learning-disabled organisations can survive for long periods in stable environments because their failures accumulate slowly and external pressures build gradually. But they cannot survive a crisis, because a crisis demands rapid adaptation — and rapid adaptation requires the ability to learn. Thomas herself admitted in January 2025 that NHS Wales systems were not ready for another pandemic. This is a learning-disabled organisation's confession: we cannot adapt because we have never learned how.

Diagnosis

The six frameworks examined above — Edmondson's psychological safety, Lipman-Blumen and Padilla's toxic leadership and toxic triangle, Morrison and Milliken's organisational silence, Collins's Level 5 leadership, Deci and Ryan's self-determination theory, and Senge's learning disabilities — were developed independently by researchers working in different fields, different countries, and different decades. They agree on a single point: there is a specific, identifiable pattern of leadership behaviour that produces organisational failure, and it is different from ordinary incompetence.

Ordinary incompetence produces random failures — some programmes succeed, some fail, and the pattern is inconsistent. What we see at DHCW is not random. It is systematic. Every major programme is failing simultaneously. The most capable staff are leaving or being removed. The organisation cannot learn from its mistakes because it punishes the people who identify them. Information flows are blocked. Accountability is structurally impossible. Resources are channelled into leadership self-preservation rather than mission delivery. And the people who bear the cost — the patients and taxpayers of Wales — have no visibility into any of it.

This pattern has a name. Lipman-Blumen called it toxic leadership. Padilla called it the toxic triangle. Collins called it the absence of Level 5. Edmondson called it the destruction of psychological safety. Morrison and Milliken called it organisational silence. Deci and Ryan would call it the systematic frustration of basic human needs. Senge would call it a learning-disabled organisation.

They are all describing the same thing from different angles: an organisation whose leaders have arranged it to serve themselves rather than its mission, and who have built structural defences against the accountability that would expose them.

This diagnosis matters because it determines the prescription. If DHCW's failures were caused by insufficient funding, the answer would be more money. If they were caused by technical complexity, the answer would be better technology. If they were caused by external factors — supplier failures, health board resistance, pandemic disruption — the answer would be patience and support.

But the evidence does not support any of those explanations. The evidence supports the explanation that leadership science has documented extensively: DHCW is failing because it is led by people whose primary objective is not the delivery of the mission, but the preservation of their own positions. And the organisation has been structured — through silence enforcement, information suppression, ally promotion, and accountability avoidance — to make this pattern invisible from the outside and unchallengeable from the inside.

Until the leadership changes, the outcomes will not change. This is not opinion. It is the central, replicated finding of fifty years of organisational research.

References

Argyris, C. (1977). Double loop learning in organizations. Harvard Business Review, 55(5), 115-125.
Argyris, C. & Schön, D. A. (1978). Organizational Learning: A Theory of Action Perspective. Addison-Wesley.
Collins, J. (2001). Good to Great: Why Some Companies Make the Leap... and Others Don't. HarperBusiness.
Deci, E. L. & Ryan, R. M. (1985). Intrinsic Motivation and Self-Determination in Human Behavior. Plenum Press.
Duhigg, C. (2016). What Google learned from its quest to build the perfect team. The New York Times Magazine, 25 February.
Edmondson, A. C. (1999). Psychological safety and learning behavior in work teams. Administrative Science Quarterly, 44(2), 350-383.
Edmondson, A. C. (2004). Learning from failure in health care: Frequent opportunities, pervasive barriers. Quality and Safety in Health Care, 13(suppl 2), ii3-ii9.
Edmondson, A. C. (2018). The Fearless Organization: Creating Psychological Safety in the Workplace for Learning, Innovation, and Growth. Wiley.
Gagné, M. & Deci, E. L. (2005). Self-determination theory and work motivation. Journal of Organizational Behavior, 26(4), 331-362.
Lipman-Blumen, J. (2005). The Allure of Toxic Leaders: Why We Follow Destructive Bosses and Corrupt Politicians — and How We Can Survive Them. Oxford University Press.
Morrison, E. W. & Milliken, F. J. (2000). Organizational silence: A barrier to change and development in a pluralistic world. Academy of Management Review, 25(4), 706-725.
Nembhard, I. M. & Edmondson, A. C. (2006). Making it safe: The effects of leader inclusiveness and professional status on psychological safety and improvement efforts in health care teams. Journal of Organizational Behavior, 27(7), 941-966.
Padilla, A., Hogan, R. & Kaiser, R. B. (2007). The toxic triangle: Destructive leaders, susceptible followers, and conducive environments. The Leadership Quarterly, 18(3), 176-194.
Rozovsky, J. (2015). The five keys to a successful Google team. re:Work, 17 November. Available at: rework.withgoogle.com.
Ryan, R. M. & Deci, E. L. (2000). Self-determination theory and the facilitation of intrinsic motivation, social development, and well-being. American Psychologist, 55(1), 68-78.
Senge, P. M. (1990). The Fifth Discipline: The Art and Practice of the Learning Organization. Doubleday.
Van Dyne, L., Ang, S. & Botero, I. C. (2003). Conceptualizing employee silence and employee voice as multidimensional constructs. Journal of Management Studies, 40(6), 1359-1392.
Audit Wales (2022). Welsh Community Care Information System. Report to the Senedd.
National Assembly for Wales, Public Accounts Committee (2018). Informatics Systems in NHS Wales. Report of the Public Accounts Committee.
Welsh Government (2025). Written Statement: Digital Health and Care Wales — Escalation to Enhanced Monitoring (Level 3). March 2025.