Sources of Failure Demand in Healthcare

Sources of Failure Demand in Healthcare

Nick Downham © Copyright Cressbrook Ltd 2025 – All Rights Reserved

First published 2018, updated March 2025

This article introduces failure demand, its systemic sources in healthcare, and the principles to avoid it. It is unashamedly focused on the assumptions and ideas that shape how work is done (the root causes) rather than on tools and techniques (which often focus on addressing symptoms).

Failure Demand
Improvement guru John Seddon describes failure demand as ‘The demand placed on the system, not as a result of delivering value to the ‘customer’, but due to failings within the system.’

In other words, based on my study and experience, we are essentially creating our own work – and alarming amounts of it in healthcare. This work is not produced by genuine patient demand, but rather by problems with our response to that demand. A helpful way to consider Failure Demand is to view it as one of the significant costs of poor quality – (Safety, Effectiveness, Patient-Centredness, Timeliness, Efficiency, and Equitability – IOM 1989).

Failure Demand examples – Downham 2024

When considering the size of most health and social care systems and providers, the scale is vast. In my modest studies, I have identified the following levels of failure demand:

19%* of all GP consultations (approximately 150 million+ consultations in NHS England annually).

89% of the demand for frequently attending patients is either partly or wholly due to failure demand (this group of patients will be accessing primary and secondary care).

15%** of ITU patients experience an extended stay because of failure demand.

55%** of in-patient psychiatry cases exhibit complexities caused by failure demand.

The practical reality is that this not only uses precious resources but also means that our improvement efforts often concentrate on the wrong thing. We frequently focus on making things more efficient (the general focus of most Quality Improvement initiatives), on temporary waiting list pushes, or asking our staff to give even more discretionary effort—but what if we are making the wrong thing more efficient?

Understanding failure demand is all about questioning the work in the first place.

*GP led demand categorisation analysis produced with my fellow collaborators at Develop Consulting.

**Results of clinician led audit I facilitated.

The Root Causes of Failure Demand in Healthcare

Failure demand is not an individual issue. It is not about staff doing the wrong or right thing – it is a systemic issue. It is about how the work has been organised, e.g. how our systems, processes and organisations have been designed. I use the word ‘design’ deliberately because the way things are now is, to a great extent, a result of a set of decisions and assumptions. As explored in my book, Improving Quality in Healthcare, co-authored with Murray Anderson-Wallace, systems thinking suggests front-line staff have far less influence over their work than commonly assumed. System conditions, which can be thought of as the unofficial and official ‘rules of the game’ (Anderson-Wallace & Downham 2024), shape the work. These are factors such as policies, standards, guidelines, incentives, structures, information, targets, measurement, custom and practice, equipment, resources, criteria, legacy models of care, peer pressure and opinion, to name just a few. Tellingly, leading leadership thinker Peter Senge (2006) states that individuals, however different, placed within the same structures (system conditions) will produce the same results.

As explored further in my book there is much more to this in terms of individual agency to act upon a system and the role of ideology (ideas) in shaping systems, but the point remains – there are systemic issues that cause failure demand. The diagram below details these issues across four powerful domains:

Systemic Sources of Failure Demand

Fragmentation by Design

Anywhere a patient is handed over, authorisation is sought, or part of a process / care is reliant on other teams or services is a prime source of failure demand. Important information falls between the cracks, and confusion and delay are created when we pass responsibility for care between individuals, teams and organisations. Each team, staff group, department or service tends to concentrate on ‘its bit’ rather than view the whole. The result is that individual departments can lower costs while the total system costs increase. Of course, some referrals to specialist services (like a GP referring a patient to a hospital for an operation) are required. However, in many, many cases, these handoffs are not a result of specialist service but rather a result of deliberate design decisions.

The number of handoffs driven by fragmentation in patient care can be staggering – and damaging for the service and the patient. For example, in the space of just a few months, a moderate to severely frail person can interact with tens of services and 100’s of different professionals. Each new service and professional requires a handoff, increasing the likelihood of failure demand. It also means that the individual cost of interventions can be low, while the overall end-to-end cost can be huge.

Why is it so fragmented??

The bewildering number of organisations, departments, services and staff groups that comprise our health services are a product of ideology (a system of ideas and ideals). As W. Edwards Deming suggests in his System of Profound Knowledge it is often the Theory of Knowledge (ideology) of leaders that governs how services are organised. These ideologies are often so normalised that they are unspoken and unrecognised. We often assume that this is how the organisational world works. We get taught it in many university courses and business schools. As a young manager, I assumed that services and organisations had to look a certain way (generally top-down). The product of these dominant leadership ideologies is often a fragmentation of our ability to understand and respond to patient needs holistically.

The dominant ideologies and how they fragment services:

The professional

The professions such as GPs, consultants, nurses, midwives, accountants and managers all have strong individual identities. They often have their own professional bodies and standards for entry and view of what their work should be and often protect their own interests. Some of this can be very useful, for example, in upholding standards, but in many cases, this strong identity can be a source of failure demand. A closed view of ‘my work’ increases the number of other professionals who need to be involved in a patient’s care. Increasing the likelihood of failure demand.

Economic / Industrial

The influence of economic and industrial thinking has been profound on our health services. It is in the DNA of our NHS. Much of this thinking concentrates on reducing labour costs. This is achieved through the division of labour and functional specialisation. This is the process of breaking work and job roles down into smaller chunks that can be simplified and specified. It means that someone lower skilled can do that piece of the work, and / or we can specialise in a specific type of work. This can bring benefits of specialism (such as higher quality in highly technical work), and it also leads to the notion of economies of scale. If we can concentrate on fewer things and do more of them, it will be cheaper. This may be the case in isolation, but not if it creates cost elsewhere in the system. This all leads to a proliferation of departments, roles, and staff groups and even outsourcing of some tasks and roles – increasing fragmentation.

Bureaucratic

Bureaucratic organisational design centres on limiting spans of control. It intends to limit change and maintain order so you do not lose control of an organisation. Almost every large institution, including the NHS, is based on these principles. A typical manifestation of this is the classic top-down organisational structure. At the level of the work, it has led to the introduction of the professional manager, the separation of decision-making from work and the increasing governance and specification of all work. This means that front-line staff have to concentrate on their bit, as defined by someone else, and only that bit. It all increases fragmentation.

A note on integration and IT systems

Integration and new IT systems are often seen as the magic bullet for solving issues caused by fragmentation. We often invest in IT to connect multiple staff groups and systems, reducing delays and stopping information from falling through the cracks. However, these systems are massively complex and expensive. You would not need such a complex IT system to cope with fragmented services if they were not fragmented in the first place.

Blind to the whole

Many services are consolidating to bring them under one roof in response to fragmentation. Unfortunately, in many cases, this tends to concentrate on pooled budgets, co-location, co-contracting, single points of access, and mergers. So, the same list of job roles, teams, services, and professional groups exists, just under the same banner. The fragmentation remains.

In summary and as stated in my book, despite the promise and often noble intention of the ideas in these ideologies, it has to be noticed that in complex systems of healthcare, there is a curious but profoundly damaging paradox: while massive levels of task specification, accountability and control have been created, in many systems, there has been a loss of control, accountability and visibility of the whole. In many cases, care has been atomised to the point where we are blind to the whole (adapted from Anderson-Wallace & Downham 2024).

A practical example of these ideas in action:

Let’s consider an example of a young person with mental health needs. The result of the fragmentation described is that the view of the young person’s needs, along with their context, is systemically fragmented. The pieces of the picture are held by Education, Social Care, VCS, and, of course, Health. Focusing on Health, specifically General Practice in this example, the care model has fragmented over the past five or six years. For instance, new professional groups have been introduced, new models of access (often driven by the push for a digital interface and triage systems) have been implemented, changes in GP working patterns have occurred as more GPs reduce their sessions due to workload pressure, and there has been a push for larger practices with bigger teams. The benefits of these changes are contested, but when considering our young person and the GP practice, this means:

– that continuity and forming an effective relationship can be more difficult,

– patients can often end up repeating distressing stories to different people,

– diagnoses can be delayed,

– care is delayed due to the increased level of gatekeeping by overstretched services.

A 10-month delay, which is typical for a young person requiring MH support, is a huge amount of time and damaging for someone in their early teens… They are likely to become more complex. This all creates failure demand.

Defensive Pressures

A set of pressures that force clinicians, managers and staff to act upon a short-term view or practice defensively rather than take the long-term best interests of the service and patients.

Medical / Legal Pressures

From a clinical perspective, this leads to overtreatment, excess diagnostics, depersonalisation, and avoidable referring and deferring. This defensive practice is often driven by perceived or actual medical/legal pressures. In some settings, up to 93% of clinicians report practising defensively.

The Social Systems Defense Against Anxiety

These defensive pressures are also driven by a social systems defence against anxiety among clinicians, which is caused by poor human factors, problems with culture, and a lack of support mechanisms for clinicians within organisations. This defence mechanism promotes depersonalisation, avoidance of decisions, denial of feelings, and the breakdown of the patient/clinician relationship.

Moral Ethical

Clinicians are often torn between doing no harm to a patient and doing what is in their best interests. Complicating this balance is the increasing workload and medical/legal pressure clinicians face. Picking up our example of frailty above, in some cases, it is in the patient’s best interests to remain at home, but they are admitted because of the potential risk of harm (they may fall). This can, in many cases, lead to over-treatment, complications and, of course, increased handovers – all sources of failure demand.

Political

Political ideologies come in all shapes and sizes, but there are dominant themes that lead to failure demand. These include the push for short-term results, often within the electoral cycle. This pressure for results can often lead to changes to organisations that may improve a specific process or service but, in fact, moves the problem elsewhere. For example, meeting the A&E four-hour wait, but moving the pressure elsewhere in the hospital or to community services. Another emerging theme is the drive for personal accountability – or some would say ‘someone to blame’. This forces leaders to look after their interests rather than see the whole.

These ideas in action:

Again, returning to our young person with MH needs, there are still strong schools of thought that advocate for distance and deliberate lack of connection in the patient relationship – with the view that it is more professional and protects professionals from the distress of medicine. Combine this with an increasing task and protocol-based approach to training and work design, and forming a relationship with patients can be marginalised – relational care being pivoltal for MH and much of healthcare. Further complicating factors include fear of litigation – there are many clinicians, at all levels, that are terrified of the coroner, and finally, in this melting pot, we commonly have pressure from families to intervene.

This all means it is deeply unclear if interventions and decisions to help our young people are in their best interest. This leads to delay, overtreatment and undertreatment – and a deeply unsatisfactory experience for professionals who join services to do ‘right’.

The Specialist, Generalist and Citizen Muddle

Doing generalist work in specialist settings, doing specialist work in generalist settings, and doing both when, in fact, the citizen is better placed all cause failure demand. This can play out at many levels. For example (Anderson-Wallace & Downham 2024):

– The interaction between a GP and a hospital.

– The relationship with the district hospital and the specialist hospital.

– The relationship between the consultant and nurse.

– The relationship between the GP and the citizen and family.

Perhaps driven by fear of litigation, staffing, and demand pressures, many care decisions and interventions are deferred and referred for delivery by specialists and specialist centres. For example, much of the pressure on the capacity of Paediatric Intensive Care Units (PICU) is caused by cases being sent to them that could be cared for in local high-dependency units (HDU). The Royal College of Paediatrics and Child Health states that, on average, 33% of all PICU bed days could have been cared for in an HDU.

The high-profile NHS Getting Right First Time (GIRFT) reports for Ear, Nose and Throat surgery, Spinal surgery, Oral and Maxillofacial surgery, Vascular surgery, Cranial Neurosurgery and General surgery found that inappropriate settings and referrals led to issues such as delays, avoidable admissions, re-admissions and the avoidable of valuable specialist response.

There are countless other examples. In primary care, there is the notion that a specialist must do physical assessments, and social care assessments cannot be done by healthcare workers. Patients also value a GP (a generalist clinician) who refers them to a specialist rather than managing their condition or presentation in the community.

Put simply, we frequently muddle skilled generalist work (which still requires skill and expertise) with genuinely specialist work. This increases fragmentation, as it requires a larger number of different professionals and, in the case of specialist centres, unnecessarily drags patients away from their homes.

These ideas in action:

Returning to young people with MH needs, this muddle plays out in a number of problematic ways. We have generalists routinely holding extremely complex and already harmed young people, driven by a lack of MH capacity and by pressure on MH services to discharge quickly.

Due to expectations of specialist input, the reverse can occur, pathologising arguably normal responses to times of distress. This can start a relationship with health services that is hugely difficult to disengage from. In many cases, the community might be better placed to support people with these needs.

Misidentification of Needs

The social determinants of health dwarf the healthcare determinants of health. Most clinicians accept, for many high-volume conditions, that a health and social response is generally required to help someone. However, our services are set up with an overwhelming bias toward health. Of course, it depends on the condition. For example, if a GP finds an indicator for cancer, then it is appropriate and desired to have a swift and hopefully effective response from the system. At least in the short term, this will be health-focused. However, other conditions, such as diabetes, are much more dependent on a balanced mix of health and social interventions. It can often be social contextual pressure (life pressure) that undermines health advice and interventions.

Through studies I have supported, GPs judged that 21% of patients were likely to be under pressure in their context (social stressors), and 4% had a turbulent social context. In other words, more than a quarter of the patients seen were likely to have a life context that would profoundly impact their ability to follow advice or make lifestyle changes. This dramatically limits the effectiveness of health-dominant approaches in many conditions. The practical implication of this is that in studies with GP practices, I have found 21% of GP consults are relational work – i.e. to fundamentally help, the patient requires a response that may include, but is largely concerned with, factors outside of the biomedical, such as the person’s social or environmental circumstances.

This dominant health-focused design reduces options available to clinicians even if they identify a social need, for example, prescribing anti-depressants for something with social root causes. It forces clinicians to treat the presentation rather than the underlying cause. Hence why, we watch so many continue along a journey of declining well-being while we treat their individual presentations. Many of these individuals end up escalating in needs and complexity, requiring a hugely expensive intervention when we could have avoided it if we had the means of intervening earlier.

These ideas in action:

We are set up to respond to labels and individuals. For mental health in young people, the patterns are often multigenerational and found within the family unit. Needs are driven by social and environmental context and sometimes past trauma. A GP friend of mine eloquently summarised this. They described that rarely, as a health service, do we truly understand or are able to retain what someone needs and hopes for, whether that is a young person or someone who is palliative.

The so what?

This piece is about raising the awareness of the implications of our design of structures and services when we lead teams, organisations and systems. This is also pertinent as we find ourselves in a position where we have massive financial pressure and continue to struggle to make lasting improvements. The problem may not be funding or efficiency; it may be effectiveness. We spend money on epic levels of failure demand rather than meeting actual patient needs – failure demand caused by our own deliberate design choices. The good news is that if we choose, we can make a different set of design decisions to transform how we deliver care radically.

Of course, it is a balance. It depends on the work. The current design may work for linearly bounded work, such as a common cancer intervention. However, for conditions that are reliant on patient context and relationships, the current design severely limits our capability to respond to patient needs, which is where much of our system’s rising costs lie.

With the proper analysis, engagement and design, all of these sources of failure demand can be reduced and, in some cases, avoided total – creating the system conditions for much greater levels of quality to be achieved. We must be open to viewing our work and resulting structures and systems differently. My new book describes, in detail,  four powerful domains to act upon failure demand to help with this reflection and rethinking of models of care. In summary they are:

Defragmenting to Integrate

Supporting Human Systems at Work

Avoiding the Specialist, Generalist, Citizen Muddle

Understanding Need