Phil Whatling is a GP partner in North Devon with interests in continuity, system behaviour and the everyday realities of general practice. He is on LinkedIN
General practice remains the point at which pressures within the wider health and social care system first become visible. In recent years, many clinicians have noted a shift in the pattern of presentations: demand appears less predictable, escalation more rapid, and neighbourhood variation more pronounced. Patients present late, re‑attend frequently, or deteriorate unexpectedly. These patterns can appear abrupt, yet when examined closely, they reflect the interaction between individual circumstances, system interfaces, and the small steps that fail quietly along the way.
We need a practical language for describing these observations. It should draw on established work in systems thinking, safety science and the social determinants of health,1-6 but be grounded in the day‑to‑day reality of general practice. The intention is not to offer a comprehensive theoretical model, but to outline a clinically useful way of understanding why some problems escalate and why neighbourhoods differ in their ability to absorb strain.
From demand to underlying conditions
Demand is often discussed as if it were an external pressure acting upon services. In practice, it is the downstream expression of upstream conditions. Marmot’s work on the social determinants of health highlights how income, housing, transport, education and chronic stress shape health outcomes3. These factors together with individual experiences influence how and when patients seek help, how they interpret risk, and how reliably they can follow a plan. Importantly clinician behaviour is influenced by similar factors (e.g. cognitive load).
Demand is often discussed as if it were an external pressure acting upon services. In practice, it is the downstream expression of upstream conditions.
In primary care, these influences are visible long before they appear in formal activity data. Late presentation, reassurance‑seeking, crisis‑driven contact, missed appointments and incomplete follow‑through are not signs of disengagement. They are adaptive responses to the conditions in which people live.
Small failures as the unit of system behaviour
Safety science has long recognised that systems succeed or fail through the reliability of everyday steps. Reason’s work on organisational accidents and Hollnagel’s Safety‑II perspective both emphasise the cumulative effect of small failures.2,5 In general practice, these steps include safety‑netting advice, referrals, follow‑up arrangements, medication changes, and actions required of schools, employers or carers.
A misfire is simply a step that does not complete as intended. Examples include:
- a plan the patient is unable to carry out
- a referral rejected because thresholds differ
- a follow‑up that does not occur
- safety‑netting advice that is misunderstood
- an external action (e.g. school or employer) that does not take place
Such events are common. The key question is not whether they occur, but whether the surrounding conditions allow them to be absorbed without escalation.
Propagation: why some cases deteriorate
In supportive conditions, small failures are corrected: someone notices, someone adjusts, someone intervenes. When conditions are less favourable — digital exclusion, fragmented interfaces, limited continuity, high cognitive load, or low trust — these failures travel.
Propagation describes the process by which an unresolved step amplifies across time and interfaces: repeat contact, deterioration, crisis, escalation. It explains why demand can feel non‑linear and why similar clinical problems behave differently in different neighbourhoods.
This concept is consistent with established systems literature. Senge, Meadows and Sterman all describe how delays, feedback loops and amplification shape system behaviour.4,6,7 A synthesis of these of ideas can be applied to the everyday work of general practice and to identify misfires as the practical unit of analysis.
Inequality as propagation length
Inequality is usually described in terms of disease burden or service utilisation. From a GP perspective, it often appears as how far a missed step travels before the system can stabilise it — what might be termed propagation length.
In some neighbourhoods, a missed step is corrected early. In others, the same missed step progresses through multiple sectors before intervention is possible.
Propagation length is influenced by:
- trust and psychological safety
- relational continuity
- digital access
- transport
- employment patterns
- interface reliability
- cognitive load on staff and patients
This helps explain why two patients with similar clinical needs can have markedly different trajectories, and why neighbourhood variation persists despite apparently uniform service provision.
Neighbourhoods as early‑signal layers
Neighbourhoods are often described as delivery units. They can also be understood as early‑signal layers. This is where small failures first appear, where repeat contacts cluster, where digital exclusion becomes evident, and where differences in risk language or interface reliability are most visible.
Bate and Robert’s work on experience as data highlights that lived experience often provides the earliest indication of system strain.1 General practice encounters these signals daily. Neighbourhood structures, if designed with this in mind, could support earlier identification and mitigation of emerging problems.
Why this language is introduced now
I would like to introduce three terms:
- Misfire — a simple, non‑blaming way to describe a step that does not complete.
- Propagation — a way of explaining why problems escalate across time and sectors.
- Propagation length — a way of describing inequality in operational, rather than purely descriptive, terms.
Other components of the wider framework — particularly, a structured approach to diagnosing upstream conditions (structural, trust, reality‑of‑place, intelligence, delivery capability and experience) — are intentionally not explored here. They require more detailed development.
General practice sits at the interface between people’s lives and the wider system. It is here that the earliest signs of strain appear.
Similarly, the potential for measurement of propagation — including repeat‑contact rates, plan‑not‑executed events, interface friction and predictive modelling — is acknowledged but not expanded upon. These represent a phase of work that tests and builds upon an established conceptual language.
Implications for practice
If small failures and propagation are taken seriously, several shifts follow:
- The focus moves from managing demand to understanding the conditions that generate it.
- Inequality is viewed not only as disease burden but as variation in the system’s ability to absorb small failures.
- Early‑signal metrics become as important as activity data.
- Neighbourhoods are understood not only as delivery structures but as sites of early detection and adaptive capacity.
This perspective does not replace existing models. It provides a way of articulating what GPs observe daily: that misfires, shaped by local conditions, accumulate into the patterns we recognise as demand and inequality.
Conclusion
General practice sits at the interface between people’s lives and the wider system. It is here that the earliest signs of strain appear. A living‑systems perspective, grounded in the routine work of the consultation, may help clinicians and system leaders describe these patterns more clearly and intervene earlier.
Neighbourhoods offer a promising layer for this work. Their value lies not only in service delivery but in their potential to detect and respond to emerging problems. This article introduces some language required to begin that conversation. The next step is to test it, measure it, and develop the operational tools that may follow.
References
- Bate P, Robert G. Bringing User Experience to Healthcare Improvement. Radcliffe Publishing; 2007.
- Hollnagel E. Safety-I and Safety-II: The Past and Future of Safety Management. Ashgate; 2014.
- Marmot M, Allen J, Goldblatt P, Boyce T, McNeish D, Grady M, et al. Fair Society, Healthy Lives. The Marmot Review; 2010. Available from: https://www.instituteofhealthequity.org/resources-reports/fair-society-healthy-lives-the-marmot-review [Accessed 25 June 2026]
- Meadows D. Thinking in Systems. Chelsea Green; 2008.
- Reason J. Managing the Risks of Organisational Accidents. Ashgate; 1997.
- Senge P. The Fifth Discipline. Doubleday; 2006.
- Sterman JD. Business Dynamics. McGraw-Hill; 2000.
Featured Photo by Robert Stump on Unsplash
