The inquiry that ate the hospital

On the resource cost of accountability, and the feedback loop nobody designs out

When three patients die at Lady Sybil Free Hospital during the water failure, the City Watch opens an inquiry. This is the correct response. There were deaths. There was a systemic failure. The city has an obligation to understand what happened and to prevent it from happening again. Nobody argues against the inquiry.

The inquiry requires things. It requires the senior physicians to attend formal hearings on the third Tuesday of each month. It requires the ward administrator to pull records going back eighteen months. It requires the head of surgical to give depositions, review transcripts, correspond with the Watch’s investigative division. These are not unreasonable demands. An inquiry that did not examine the people and records involved in the failure would be an inquiry in name only.

While the inquiry runs, Lady Sybil Free Hospital runs two doctors short in the surgical ward. The head of surgical is available on alternate weeks. The ward administrator is producing documentation rather than managing the ward. The conditions that contributed to the original deaths, understaffing, interrupted water supply, inadequate backup systems, are not fixed during the inquiry period. The budget that might fix them is under review pending the inquiry’s findings. The inquiry will take four months. Its eventual report will find systemic failure and recommend, among other things, enhanced audit requirements and an additional oversight body for hospital infrastructure. The enhanced audit will require the ward administrator to produce monthly compliance reports. The oversight body will require quarterly documentation from senior clinical staff.

The accountability mechanism designed to prevent future failures has made future failures more likely.

Operational principles

The accountability trap is a reinforcing loop that runs through a structural feature of almost every oversight mechanism: it draws its resource from the same pool as the operation it is overseeing.

An inquiry into a hospital draws on hospital staff time. An audit of a public body draws on the public body’s management capacity. A fiscal commission imposed on a regional health system draws on the health system’s administrative resource. In each case, the accountability mechanism and the remediation effort are competing for the same thing: the attention and capacity of the people who are simultaneously responsible for compliance and for fixing the underlying problem.

This is not a design failure in the sense that it results from error. It is a design failure in the sense that it results from not designing. Accountability mechanisms are created by people focused on the accountability objective: establish what happened, assign responsibility, prevent recurrence. The question of what the oversight will cost the overseen organisation, and whether that cost competes with the organisation’s ability to remediate, is rarely central to the design. It should be.

The loop runs as follows. A system fails. An accountability mechanism is activated. The mechanism consumes resource from the failing system. The failing system has less capacity to address its underlying problems while the mechanism is running. The mechanism eventually finds systemic failure and recommends additional oversight. The additional oversight creates additional compliance requirements. The compliance requirements accumulate, because new requirements are added after each failure but old ones are rarely removed. Over time, the total compliance burden on the organisation grows with each incident, gradually displacing the operational capacity that the oversight was meant to protect.

Why this is easy to miss

The accountability trap is genuinely counterintuitive. Accountability is normatively correct. When a hospital fails and patients die, the impulse to demand rigorous examination of what happened is not a mistake. It is a legitimate and important social function. The argument here is not against accountability. It is against accountability mechanisms designed without awareness of their structural effect on the systems they oversee.

The trap is also obscured by timing. The resource cost of the inquiry runs in parallel with the period when the organisation is trying to recover. The competing pressures are visible to the people inside the organisation, who are simultaneously managing the inquiry, managing the remediation, and managing ongoing operations. They are much harder to see from the outside. The inquiry produces a report. The report makes recommendations. The recommendations look like improvements. The connection between the inquiry’s resource cost and the organisation’s reduced capacity to improve during the inquiry period is not documented anywhere, because nobody is measuring it.

The loop also has a longer-term dimension that reinforces invisibility. Accountability findings accumulate as requirements. Each incident produces new oversight. The new oversight is added to the existing oversight. The existing oversight is rarely reviewed or removed when new oversight is added. The total compliance burden is the product of decades of incidents, each generating its own reporting requirements, its own governance bodies, its own audit schedules. No single requirement is obviously unreasonable. The aggregate is never assessed as a whole. The organisation is, at any given moment, complying with requirements generated by incidents that took place under different leadership, in a different operating environment, addressing problems that may have long since been resolved. The compliance with those requirements continues to consume resource that could address current problems.

The Italian commissariamento

Italy’s mechanism for addressing financially failing regional health systems is the commissariamento: the appointment of an external fiscal commissioner, typically from the Ministry of Economy and Finance or from the higher ranks of public administration, with a mandate to restore financial equilibrium to the system.

The regions of Lazio, Campania, Molise, and Calabria were placed under health system commissariamento at various points from 2007 onwards after years of accumulated deficits. The commissioners’ primary mandate was financial: reduce the deficit, demonstrate compliance with national spending protocols, implement cost reduction plans that could be audited against targets. These were legitimate objectives. The health systems involved had structural financial problems that could not simply be allowed to compound indefinitely.

The effect on operational capacity was documented across multiple systems. Clinical management time was redirected to fiscal reporting, audit preparation, and regulatory correspondence. Hospital directors who had previously spent the majority of their time on clinical governance, staff management, and service improvement began spending the majority of their time on commissioner meetings, compliance documentation, and response to audit queries. The commissioners required detailed monthly reports on spending in formats that the existing administrative systems were not built to produce, requiring additional administrative work to reformat data that existed but not in the required form. Senior clinical staff in Campania reported, in survey data collected during the commissariamento period, that administrative burden had increased substantially while clinical resource had been reduced as part of the cost programme.

The health systems placed under commissariamento continued to demonstrate the clinical quality failures that had, alongside financial mismanagement, been identified as institutional problems. The oversight mechanism was producing financial compliance and consuming the management capacity that clinical improvement required. The accountability structure added to the burden of the systems it was meant to fix.

The Calabrian health system, which entered commissariamento in 2010, remained under it more than a decade later, cycling through commissioners while producing successive reports of financial improvement alongside persistent documented problems with waiting times, specialist access, and clinical outcomes. The mechanism was functioning exactly as designed. The system it was overseeing was not improving in the dimensions that mattered most.

Colectiv and its aftermath

On 30 October 2015, a fire at the Colectiv nightclub in Bucharest killed 27 people at the scene. At least 37 more died in the weeks that followed, many of them in Romanian hospitals, from infections that should have been treatable. The hospital infection scandal that emerged, one of the most serious public health revelations in recent Romanian history, showed that disinfectant solutions supplied to hospitals across the country had been systematically diluted to a fraction of their stated concentration for years. The infections that killed many of the Colectiv survivors were not a consequence of the fire alone. They occurred in hospitals that had been functionally unable to control nosocomial infection, and in a regulatory and procurement system that had not detected the fraud that enabled it.

The accountability response was extensive and politically intensive. The government fell. Multiple ministers resigned. Criminal investigations were opened against the disinfectant supplier, against procurement officials, and eventually against hospital administrators. Parliamentary committees conducted hearings. A new hospital accreditation authority, ANMCS, was created with a mandate to assess hospital quality standards.

The hospitals treating the most severely burned survivors were, during this period, simultaneously managing critically ill patients with complex infection risks, responding to investigations that required clinical staff to give statements and produce records, preparing for new accreditation assessments, implementing revised infection control protocols under external supervision, and managing the institutional and staff consequences of intense public scrutiny. Each of these demands was individually legitimate. Together, they competed for the same clinical management time and administrative capacity at a moment when that capacity was urgently needed for patient care.

Accounts from the period indicate that clinical staff who participated in exposing the failures faced professional consequences, and that the burns hospital where the worst cases were treated underwent only superficial changes despite the scrutiny. The accountability process, however necessary, created conditions in which the organisations most under pressure were also the most resource-constrained, because the accountability process itself consumed part of the capacity that recovery required.

The accumulation problem

There is a compounding version of the trap that operates over decades rather than months. Each significant failure in a regulated sector generates new requirements. Reporting obligations are added. New oversight bodies are created. Governance requirements are extended. Audit schedules are made more demanding. These requirements are layered onto the existing compliance structure; they are very rarely substituted for it. The principle that guides accumulation is simple: something went wrong, so we need more oversight. The principle that would counteract accumulation, that existing oversight which no longer serves its original purpose should be removed, has no equivalent institutional engine driving it. Oversight is easier to add than to remove.

The result, visible in healthcare systems across Europe that have experienced multiple major incidents over decades, is a compliance burden that has grown with each incident and which, in aggregate, is consuming organisational capacity at a scale that no single oversight requirement would justify. Senior management in hospital systems with long regulatory histories routinely report that compliance and reporting functions account for a substantial and growing proportion of total administrative time, and that this proportion has increased consistently regardless of whether clinical outcomes have improved or worsened.

The oversight is generating its own demand for resource. That resource comes from the organisations being overseen. The organisations being overseen are the ones responsible for clinical outcomes. The connection between these facts is documented in operational experience and almost entirely absent from oversight design.

What different design looks like

The accountability trap is not inevitable. It follows from accountability mechanisms that are designed to draw on the resources of the overseen organisation rather than being resourced externally. The alternative design principle is: if accountability is a social function, it should be funded as one, not extracted as a cost from the organisation that has already failed.

This means: inquiry teams staffed externally rather than pulling staff from the organisation under inquiry. Audit functions provided with dedicated resource rather than placing the burden of audit preparation on the operational staff whose time is already constrained. Compliance requirements assessed against their total burden, not just their individual justification. Oversight functions with sunset clauses requiring periodic re-evaluation rather than accumulating indefinitely.

These designs are more expensive. External inquiry teams cost money. Dedicated audit support costs money. Reviewing accumulated compliance requirements takes time and generates political friction, because every existing requirement was created in response to a real failure and its removal looks like a retreat from accountability. The political economy of oversight accumulation is strong. The political economy of oversight rationalisation is weak, because the costs of accumulation are diffuse and invisible and the costs of removal are concentrated and politically legible.

The report finds systemic failure

Back in Ankh-Morpork, the inquiry into the Lady Sybil Free Hospital deaths runs its course. Its report is thorough, its findings are accurate, and its recommendations are largely sensible. Enhanced audit requirements will improve financial visibility. Additional oversight of infrastructure maintenance will catch future failures earlier. The Guild of Physicians is satisfied that accountability has been served.

In the surgical ward, the head of surgical returns to full availability. The ward administrator begins managing the ward again rather than producing documentation. The new monthly compliance reports take six hours to prepare. The quarterly infrastructure documentation takes a day and a half. These are added to the existing reporting requirements, which already consume a day each week. The total is not measured as a total. It is the sum of previous inquiries, each of which found systemic failure and recommended additional oversight, none of which asked what the previous inquiry’s oversight was still costing.

The next failure, when it comes, will generate another inquiry. The inquiry will find that the hospital is struggling with administrative burden, understaffing, and inadequate resources. It will recommend enhanced oversight of staffing levels. The enhanced oversight will require monthly staffing reports. The staffing reports will be prepared by the ward administrator, who will have six fewer hours each month to manage the ward.

The loop is patient. It does not announce itself. It waits in the accumulated weight of legitimate requirements, each justified, none of them, on its own, unreasonable.