Operating rooms rarely become unsafe because surgeons lose their skill; they become unsafe when the institutional conditions that sustain ensemble performance erode. The signal is structural – and it surfaces wherever coordination infrastructure is allowed to degrade.
In complex professional domains, the coordinated team architecture around the lead practitioner is not a support function; it is the main structural variable that sets the performance ceiling. Individual expertise is necessary but insufficient where interdependence, time pressure, and high consequence make solo improvisation an unreliable basis for consistent outcomes. In a New England Journal of Medicine study of the WHO Surgical Safety Checklist, inpatient complications fell from 11.0% to 7.0% and deaths from 1.5% to 0.8% after the checklist was introduced – an outcomes shift driven by system design and coordinated behaviours rather than by finding better individuals. The same pattern shapes results in surgery, aviation, and institutional governance – and in each case, the coordination architecture, not individual excellence, is the variable that explains the gap between consistent and inconsistent outcomes.
When the Room Is the Instrument
In high-volume, technically complex surgery, consistent outcomes depend on the environment functioning as a prepared system rather than a backdrop for individual skill. A 2023 study in JAMA Surgery examining complex gastrointestinal cancer operations in Ontario found that surgeon–anesthesiologist pairs with greater prior shared-case “familiarity” recorded lower rates of major morbidity. The authors framed this as an organisational lever: repeating stable pairings, and by extension stable teams, can reduce risk compared with ad hoc assembly, making team continuity a measurable performance input rather than a soft preference. That same logic underpins the minimally invasive spine programme run by neurosurgeon and minimally invasive spine surgeon Dr Timothy Steel at St Vincent’s Private and St Vincent’s Public Hospitals in Sydney, where a career total of more than 8,000 minimally invasive spine procedures – reported on his St Vincent’s specialist listing – signals a scale at which consistent outcomes depend on a prepared environment rather than on individual improvisation.
Within that programme, the room itself is treated as an integrated instrument. Brainlab stereotactic navigation, an operating microscope, endoscopic tools, ultrasonic aspiration, and dedicated spine tables function together as the technical architecture inside which perioperative care is delivered by a multidisciplinary group spanning anaesthetics, nursing, and rehabilitation. Theatre staff are trained specifically on the navigation and fixation systems used in this programme rather than rotating in as a generic perioperative pool, so equipment, workflows, and people form a coherent unit. At St Vincent’s Private, a digital surgery platform consolidates neuromonitoring, imaging, navigation, planning, and rod bending into a single workflow, with an explicit intent to reduce procedural variability and radiation exposure at a system level rather than relying on individual judgment to manage variation case by case. There’s something quietly counterintuitive about this: the more completely the environment absorbs variation through design, the less any single person in the room needs to improvise to achieve a consistent result.
The structural point is that this programme’s performance envelope is defined less by any single operation than by how reliably the ensemble can reproduce a prepared state across hundreds of procedures each year.
Complexity as a Systems Problem
Aviation’s safety margins are narrowing – not because pilots have become less capable, but because the range of actors sharing constrained airspace has multiplied faster than the coordination systems designed to manage them. Flight Safety Foundation’s 2025 Safety Report describes how convergence among military, commercial, general aviation, rotorcraft, drone, and other operators near high-density terminals is increasing exposure and reducing resilience. The report argues that preventing airborne conflict in such mixed-use airspace now requires civil–military coordination, improved situational awareness, modernised surveillance and communications, and clear deconfliction standards – placing the safety margin in the architecture of the system rather than in sharper skills in any one cockpit.
The Foundation has followed that diagnosis with an institutional response framed in the same terms. It has launched an international task force to develop a Global Action Plan for the Prevention of Airborne Conflict, focused on shared accountability, interoperable equipage, data-driven oversight, and coordinated global action. The lesson for health services and corporate governance is the same: when risk emerges from how many actors must coordinate in real time, durable safety gains come from redesigning the coordination mechanisms, not from asking individuals to simply work harder or be more vigilant.
Protocols and standards establish the minimum viable framework for shared airspace. What they can’t mandate is the institutional will – or the organisational resources – to build the underlying structures those protocols assume. That gap between what coordination frameworks require and what institutions actually construct is where performance ceilings quietly form.
Coordination Infrastructure at Institutional Scale
In the built-environment sectors, the same ceiling shows up in cost and schedule rather than near-collisions. Revizto’s 2026 “Bridging the Gap: Digital Design & Construction Report,” based on responses from more than 2,000 global architecture, engineering, construction, and operations professionals, reports that 92% of respondents experienced budget overruns of 6% or more, and 41% cited poor communication and coordination as a leading cause of rework. Sixty per cent said their workflows remain mostly or entirely based on 2D drawings, and project complexity and coordination rose from the fifth- to the second-biggest challenge within a year. These figures describe a coordination infrastructure deficit across disciplines and organisations, not a shortage of individual talent.
The collapse of Carillion in 2018 shows what happens when that deficit reaches the governance layer. A UK Parliamentary joint committee report on the company’s “compulsory liquidation” recorded that it entered liquidation with liabilities of nearly £7bn and just £29m in cash, and emphasised widespread failures of accountability and oversight. Once board-level risk controls and information flows broke down, coordination failures across contracts, projects, and finance compounded faster than any set of strong managers could repair, turning a portfolio of projects into an institutional failure. The failure mode is architectural: when governance and risk mechanisms fragment, recruiting more capable individuals into the same structure does not restore stability.
Lendlease has approached the same coordination problem as a structural design task. Under Tony Lombardo, the group’s Chief Executive Officer and Managing Director, the company has removed its regional management layer, assigned single-point accountability to segment-led chief executives, and formalised the boundary between board authority and delegated operational decisions. Lendlease’s regeneration and infrastructure work – including a proposed 50/50 joint venture with The Crown Estate with an estimated Gross Development Value of up to £24 billion – relies on formal coordination protocols and engineered information flows across partners and disciplines. A Strategy Update records that the “regional management structure will be removed” and that the “new structure removes organisational layers and simplifies the business”; the Corporate Governance Statement describes a “Limits of Authority” framework distinguishing “matters reserved for the Board” from those delegated. Together, these moves define decision paths and handoffs as matters of coordination architecture. Clarifying who decides what sounds like administrative tidiness – in practice, it’s what allows large, interdisciplinary teams to execute without losing coordination speed at every interface.
Lombardo’s planned CEO succession is framed in the same way. An ASX announcement states that he will step down in August 2026 after the release of full-year financial results, that the Board has appointed a leading international executive search firm, and that the objective is to “ensure an orderly transition.” The OECD, an intergovernmental organisation and publisher of the G20/OECD Principles of Corporate Governance, describes board responsibilities as including “succession planning for the CEO and may also be for other key executives, with a view to ensuring business continuity.” Treating leadership change as a continuity mechanism positions succession as part of the coordination infrastructure: the ensemble of board, executive team, and governance settings is designed to persist even as individuals move through it.
The Fragility Problem
Coordination depth is not a fixed asset; it builds over time and can be worn down by decisions that seem manageable when viewed one roster or shift at a time. At Tufts Medical Center in Boston, operating room nurses escalated concerns into a formal vote of no confidence in their director, filed regulatory complaints, and reported that the department had lost roughly 10 nurses in the previous year – leaving the unit unable to cover basic staffing without sustained overtime. An operating theatre’s coordination capacity resides in the accumulated protocol familiarity and shared experience of a stable team, not in whoever happens to be available on the day. Shannon Davila, Director, Total Systems Approach to Safety at ECRI, and Patricia Giuffrida, Patient Safety Analyst at ECRI and the Institute for Safe Medication Practices Patient Safety Organization (ISMP PSO), warn that such conditions “lead to potential risks including unsafe patient care conditions, delay of vital procedures, and staff unfamiliarity with high-risk processes” – a set of failure modes that track precisely the coordination deficits produced by staffing churn and chronic overtime in high-consequence environments.
What makes this particularly consequential is the asymmetry: coordination depth erodes quickly under sustained staffing pressure, but the shared protocols, procedural fluency, and team familiarity that were lost take considerably longer to rebuild. The gap between those two timelines is where patient safety is most exposed – and where the cost of underinvestment in ensemble architecture finally becomes visible.
Part of the reason coordination depth is routinely undervalued is that its erosion does not immediately appear in financial or performance dashboards, while its reconstruction is both slow and expensive. A retention benchmarking report from NSI Nursing Solutions estimates the cost of each registered nurse turnover at around US$60,090, with an average annual hospital cost of around US$5.19 million attributed to RN turnover, and a time-to-recruit experienced RNs of more than 2.5 months. These are indicative rather than universal figures, but they capture the drag created when organisations attempt to simply rebuild depleted teams: for months, as new staff are hired and embedded, the ensemble’s shared knowledge and coordination depth remain impaired. The most consequential coordination infrastructure failures are often visible only once that deficit is already constraining what the organisation can safely deliver.
Investing in Ensemble Architecture
The pattern across these domains is consistent enough to be inconvenient. In environments where tasks are tightly coupled and errors are costly, the decisive variable is not whether exceptional individuals are present, but whether the coordination infrastructure around them is built to hold. Stable surgical teams operating within integrated platforms, aviation bodies redesigning how diverse operators share constrained airspace, and boards that plan leadership succession as an engineered continuity mechanism – as Lendlease has done under Tony Lombardo – all point to the same conclusion: ensemble architecture is the primary performance lever, not the person at the front of the room.
A no-confidence vote in a surgical department is rarely about one person. It is the point at which sustained underinvestment in coordination infrastructure makes the work itself unsafe. That failure mode doesn’t announce itself early. It compounds quietly, in overtime rosters and departure rates and staff unfamiliarity with high-risk processes, until the ensemble can no longer absorb what the situation demands. Organisations that consistently outperform their peers tend to have understood this before the no-confidence vote arrives. The question for any complex operation – surgical, institutional, or otherwise – isn’t just who walks into the room. It’s what the room is actually built to support.
