The Anatomy of Institutional Failure: What Leadership Science Tells Us

Digital Health and Care Wales has spent over £100 million across a portfolio of programmes, every one of which is now subject to Level 3 intervention for delivery failures. Senior employees have allegedly been dismissed after raising concerns. The CEO cannot demonstrate return on investment for any programme. The Chair admits to being informed "late in the day" about critical issues.

These are not random misfortunes. They are the predictable consequences of well-documented organisational pathologies. Six established frameworks from leadership science describe, with uncomfortable precision, exactly what has gone wrong at DHCW — and why it cannot self-correct.

1. Psychological Safety: The Foundation That Was Never Built

In 1999, Harvard Business School professor Amy Edmondson published research defining psychological safety as "a shared belief held by members of a team that the team is safe for interpersonal risk-taking." Her work demonstrated that organisations where people feel safe to speak up, admit mistakes, and challenge decisions consistently outperform those where they do not.

Google's Project Aristotle — a multi-year study of 180 teams — confirmed this finding at scale. Psychological safety was the single strongest predictor of high-performing teams. Not talent, not resources, not strategy. Safety.

Now consider DHCW's predecessor, the NHS Wales Informatics Service (NWIS). In 2018, the Public Accounts Committee (PAC) of the Senedd investigated NWIS and found what it described as the "antithesis of open." Staff were "reluctant to be critical on the record." The committee's findings painted a picture of an organisation where candour was not merely discouraged — it was dangerous.

DHCW was created in April 2021, ostensibly as a fresh start. But the same senior leaders remained. The same management structures persisted. And the outcomes tell us the culture persisted too.

DHCW publishes zero whistleblowing disclosure data. It publishes no record of how many staff have raised concerns, how those concerns were handled, or what outcomes resulted. At least two senior employees have allegedly been dismissed after raising concerns, with Employment Tribunal claims filed alleging unfair dismissal connected to whistleblowing.

In an organisation with genuine psychological safety, concerns flow upward freely. At DHCW, the evidence suggests they are met with consequences. When this happens, the entire organisation loses its ability to detect and correct problems. Bad decisions go unchallenged. Risks go unescalated. Programmes drift toward failure in silence — which is precisely what happened across all nine programmes now under intervention.

2. The Toxic Triangle: Leaders, Followers, and Environments

Jean Lipman-Blumen's 2005 work on toxic leadership described leaders who engage in destructive behaviours that inflict serious harm on their organisations. But toxicity at the top is only part of the story.

In 2007, Art Padilla, Robert Hogan, and Robert Kaiser proposed the toxic triangle — a model showing that destructive leadership requires three interlocking elements:

• Destructive leaders who prioritise self-preservation over mission
• Susceptible followers — a management layer that stays silent, either through conformity or fear
• Conducive environments — governance structures that provide insufficient external scrutiny

Each element reinforces the others. Remove any one, and the system corrects. Leave all three in place, and deterioration accelerates.

At DHCW, the triangle is visible in the public record:

Destructive leaders: The CEO has been in post since DHCW's creation in 2021. In January 2025, she appeared before the Senedd's Health and Social Care Committee and was unable to provide programme delivery timelines. She has never publicly demonstrated return on investment for any programme. Yet she remains in post. When leaders prioritise position security over accountability, the organisation's mission becomes secondary to its leadership's survival.

Susceptible followers: The management layer beneath the CEO includes individuals described in tribunal documents as proteges — people whose careers are tied to the current leadership. In this dynamic, loyalty to the leader substitutes for loyalty to the mission. Concerns are not escalated because escalating them threatens the network of relationships that sustains careers.

Conducive environments: DHCW operates within NHS Wales governance structures that provide limited external scrutiny. The organisation is sponsored by the Welsh Government, but oversight has been characterised by what the Cabinet Secretary described as a "pattern of late notification." When governance is distant, slow, and reliant on the organisation's own reporting, the conditions for the toxic triangle are ideal.

3. Organisational Silence: When Information Stops Flowing Upward

In 2000, Elizabeth Morrison and Frances Milliken published their landmark study of organisational silence — the systematic barriers that prevent information from flowing upward in institutions. They found that silence is not simply the absence of voice. It is an active, self-reinforcing dynamic that organisations develop over time.

Linn Van Dyne, Soon Ang, and Isabel Botero (2003) identified three distinct types of employee silence:

• Acquiescent silence: Staff have given up. They believe speaking up will make no difference. They have resigned themselves to the status quo.
• Defensive silence: Staff stay quiet to protect themselves. They have seen what happens to those who speak up, and they choose self-preservation.
• Prosocial silence: Staff stay quiet to protect colleagues or the organisation itself. They fear that raising concerns will damage people they care about.

All three types produce the same outcome: critical information never reaches decision-makers.

At DHCW, the evidence of organisational silence is striking. The Chair of the Board has admitted to being informed "late in the day" about significant issues. The Cabinet Secretary has confirmed a "pattern of late notification" from DHCW to the Welsh Government. If information is not reaching the Chair or the sponsoring government department, it is certainly not reaching the public.

This is not a communication problem. It is a silence problem. When at least two senior employees have allegedly faced dismissal after raising concerns, the rational calculation for any remaining employee is clear: say nothing. The cost of this silence is measured in programmes that fail without warning, budgets that overrun without scrutiny, and patients who receive care supported by systems that no one dared to criticise.

4. Level 5 Leadership — and Its Precise Inversion

In Good to Great (2001), Jim Collins studied companies that made the transition from sustained mediocrity to sustained excellence. He found that every single one was led by what he called a Level 5 Leader — someone who combined deep personal humility with fierce, unwavering commitment to the organisation's mission.

Level 5 leaders channel ambition into the institution, not themselves. They take responsibility for failures and give credit to others for successes. They measure their own performance by the outcomes they deliver for the people they serve.

DHCW's leadership represents the precise inversion of this model.

Rather than channelling ambition into patient outcomes, the CEO accumulated a series of credentials and honours during the 18-month period surrounding her appointment: a BCS fellowship, FedIP registration, an honorary "Professor of Practice" title from the University of Wales Trinity Saint David, and a "Digital CEO of the Year" award. These are markers of personal brand-building, not institutional delivery.

Rather than taking responsibility for failures, DHCW's leadership has consistently attributed problems to external factors — complexity, legacy systems, the challenges of working across NHS Wales. When the CEO appeared before the Senedd in January 2025 and could not provide programme timelines, the response was not accountability but deflection.

Rather than measuring success by patient outcomes, the organisation publishes no return-on-investment data for any programme. Success is measured, if it is measured at all, by activity — meetings held, strategies published, conferences attended — rather than by results delivered.

Collins found that Level 5 leadership was not merely desirable but essential. Without it, organisations could achieve competence but never greatness. At DHCW, the inversion of Level 5 leadership has produced something worse than mediocrity — it has produced systematic failure while maintaining the appearance of institutional respectability.

5. Self-Determination Theory: Destroying Motivation From Within

Edward Deci and Richard Ryan's self-determination theory (2000) identifies three fundamental psychological needs that must be met for people to be motivated, effective, and well:

• Autonomy: The need to feel ownership over one's work and decisions
• Competence: The need to feel effective and capable in one's role
• Relatedness: The need for trust, connection, and mutual respect with colleagues

When these needs are met, people are intrinsically motivated. When they are denied, people disengage, burn out, or leave.

At DHCW, the public record suggests all three needs are systematically undermined.

Autonomy denied: A former senior employee — the Chief DevOps Officer, recruited at Band 9 (£101,000-£117,000) to lead DevOps transformation — was building a team and recruiting subordinates when they were allegedly dismissed. The role was subsequently replaced at a significantly lower grade. When senior technical leaders are removed and their roles downgraded, the message to remaining staff is unambiguous: your professional judgement does not matter here.

Competence denied: DHCW has repeatedly overridden expert technical recommendations. Programmes have been procured, deployed, and then abandoned when they failed to deliver — only for the cycle to repeat. When an organisation ignores the expertise of the people it hires, it tells those people that their competence is irrelevant. The predictable result is that competent people leave, and those who remain stop offering their expertise.

Relatedness poisoned: Trust is the foundation of relatedness. When colleagues are allegedly dismissed for raising concerns, when information is withheld from the Board, when the Chair is informed "late in the day," trust is not merely damaged — it is destroyed. In the absence of trust, collaboration becomes performative. People attend meetings but withhold their real assessments. Teams function on paper but not in practice.

The cumulative effect is an organisation that has systematically dismantled the conditions under which its own staff can do good work. This is not a recruitment problem or a skills shortage. It is a leadership problem.

6. The Learning-Disabled Organisation

Peter Senge's The Fifth Discipline (1990) described organisations that are structurally incapable of learning from their own experience. He identified several characteristic disabilities, three of which map directly onto DHCW:

"I am my position" — When people define themselves by their role rather than their purpose, they protect their territory rather than serve the mission. At DHCW, the accumulation of titles, fellowships, and honorary positions by senior leaders suggests an organisation where identity is tied to status, not outcomes. Turf protection takes precedence over collaboration.

"The enemy is out there" — When organisations fail, learning-disabled institutions attribute the failure to external forces rather than examining their own decisions. DHCW has consistently pointed to the complexity of NHS Wales, the challenges of legacy systems, and the difficulties of multi-organisation working. These factors are real, but they apply equally to NHS Digital in England and NHS Scotland — organisations that have delivered comparable programmes successfully. The enemy is not out there. The enemy is in the leadership.

"The boiled frog" — Senge's most vivid metaphor describes organisations that fail to recognise gradual deterioration because each individual step is small enough to normalise. DHCW's programmes did not fail overnight. They deteriorated over years. OpenEyes: seven years and two missed deadlines. LIMS: eight years and a terminated supplier. WCCIS: eleven years and £42 million with organisations trying to leave. Each delay was explained, each overrun justified, each missed deadline rescheduled. The deterioration was normalised until the Welsh Government imposed Level 3 intervention across the entire portfolio.

The most damaging characteristic of the learning-disabled organisation is that it repeats identical patterns. DHCW's programme history shows a consistent cycle: ignore expert advice, procure the wrong system, discover it fails, pay again. This is not bad luck. It is an organisation that cannot learn because the conditions for learning — psychological safety, honest feedback, accountability for outcomes — do not exist.

The Only Mystery

These six frameworks — psychological safety, the toxic triangle, organisational silence, Level 5 leadership, self-determination theory, and the learning-disabled organisation — are not obscure academic theories. They are established, peer-reviewed, widely taught models of institutional behaviour. They have been validated across thousands of organisations over decades of research.

Applied to DHCW, they do not merely explain the organisation's failures. They predict them. Every pathology documented in the public record — the silencing of dissent, the accumulation of personal credentials over institutional outcomes, the systematic failure to learn from repeated mistakes, the destruction of the conditions under which good work is possible — is textbook.

DHCW's failures are not mysterious. They are not the product of unique circumstances or exceptional complexity. They are the entirely predictable result of leadership behaviours that have been studied, named, and understood for decades.

The only mystery is why those with the power to intervene — the Welsh Government, the Senedd, the NHS Wales governance structures — have waited so long. The frameworks exist. The evidence is in the public record. The diagnosis is clear. What remains is the question of whether anyone will act on it.

References

• Collins, J. (2001). Good to Great: Why Some Companies Make the Leap... and Others Don't. HarperBusiness.
• Deci, E. L. & Ryan, R. M. (2000). "The 'what' and 'why' of goal pursuits: Human needs and the self-determination of behavior." Psychological Inquiry, 11(4), 227-268.
• Edmondson, A. (1999). "Psychological safety and learning behavior in work teams." Administrative Science Quarterly, 44(2), 350-383.
• 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.
• 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.
• 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.
• Google re:Work. (2015). "Guide: Understand team effectiveness." Project Aristotle findings.