-
[[#Content Relevant to You Personally#Dr Suzanne O’Sullivan’s Book|Dr Suzanne O’Sullivan’s Book]]
-
- [[#Content Relevant to You Personally#Phase 1: The Patient-AI Engagement (Unpacking and De-escalating)|Phase 1: The Patient-AI Engagement (Unpacking and De-escalating)]]
- [[#Content Relevant to You Personally#Phase 2: The Generalist’s Review (The Human Safety Net)|Phase 2: The Generalist’s Review (The Human Safety Net)]]
- [[#Content Relevant to You Personally#Phase 3: Shared Decision-Making and Functional Support|Phase 3: Shared Decision-Making and Functional Support]]
-
[[#Content Relevant to You Personally#Existing NHS Pilots (2025–2026 Landscape)|Existing NHS Pilots (2025–2026 Landscape)]]
- [[#Content Relevant to You Personally#Mapping Your Proposals to the NHS Reality: The Scope Alignment|Mapping Your Proposals to the NHS Reality: The Scope Alignment]]
- [[#Content Relevant to You Personally#How Your Proposed Model Relates to and Improves Current Frameworks|How Your Proposed Model Relates to and Improves Current Frameworks]]
- [[#How Your Proposed Model Relates to and Improves Current Frameworks#1. Moving from “Box-Ticking” Triage to Deep-Data Synthesis|1. Moving from “Box-Ticking” Triage to Deep-Data Synthesis]]
- [[#How Your Proposed Model Relates to and Improves Current Frameworks#2. Assuring Accuracy and Rebuilding Clinical Trust|2. Assuring Accuracy and Rebuilding Clinical Trust]]
- [[#How Your Proposed Model Relates to and Improves Current Frameworks#3. Operationalizing O’Sullivan’s Goal: Bypassing the Label|3. Operationalizing O’Sullivan’s Goal: Bypassing the Label]]
Intro
Understanding “The Age of Diagnosis”
“The Age of Diagnosis” is a multifaceted concept that typically refers to two distinct but culturally significant frameworks: the broader societal shift in how medicine identifies conditions, and the specific chronological age at which an individual receives a medical or psychological diagnosis.
The Societal Perspective: A Medical Era
In sociological and medical history contexts, “The Age of Diagnosis” describes the modern era of medicine, where advanced diagnostics, genetic screening, and biomarker identification have fundamentally transformed the human experience.
-
Shift from Treatment to Detection: Historically, medicine focused primarily on treating visible symptoms. Today, the focus has shifted heavily toward detecting diseases—such as cancer, cardiovascular issues, or neurodegenerative conditions—long before physical symptoms manifest.
-
The Rise of Over-Diagnosis: With highly sensitive screening technologies, a major contemporary debate centers on “over-diagnosis.” This occurs when abnormalities are detected that may never progress to cause harm or shorten a person’s life, occasionally leading to unnecessary anxiety and intervention.
-
Neurodivergence and Mental Health: The term is also frequently used to describe the massive surge in developmental and psychological diagnoses, such as ADHD and Autism Spectrum Conditions, driven by increased awareness, destigmatization, and evolving diagnostic criteria.
The Developmental Perspective: Chronological Timing
From a clinical standpoint, the age at which a condition is diagnosed has profound implications for management, prognosis, and identity.
-
Early Intervention (Childhood): For developmental, genetic, or neurodivergent conditions, diagnosis during early childhood typically unlocks crucial support systems, educational adjustments, and targeted therapies.
-
Adult Diagnosis: Receiving a diagnosis later in life—such as discovering neurodivergence or a chronic genetic condition in adulthood—often triggers a retrospective re-evaluation of one’s entire life history, bringing a mix of relief, validation, and grief for missed support.
Content Relevant to You Personally
While the broader definition spans all of medicine, specific areas within the diagnostic landscape carry distinct relevance for individuals navigating mid-to-later life transitions, family health, and proactive wellness.
-
Proactive Health Screenings in Your 50s: For men in their mid-50s, the current medical paradigm emphasizes routine, age-based diagnostic screening rather than waiting for symptoms. This includes cardiovascular risk assessments, lipid profiles, blood pressure monitoring, and bowel cancer screenings (which typically begin automatically between ages 50 and 60 in the UK).
-
The Landscape for Young Adults: For individuals in their late teens and early twenties navigating higher education, the “age of diagnosis” often manifests as the formal identification of mental health trends, stress-related conditions, or previously masked learning differences, as the structured support of secondary school gives way to independent university life.
Would you like to explore a specific angle of this concept, such as the history of a particular medical diagnosis, or the current UK guidelines for proactive health screenings?
The Book
Dr Suzanne O’Sullivan’s Book
Yes, there is a prominent book with this exact title. “The Age of Diagnosis: Are Medical Labels Doing Us More Harm Than Good?” (published in the US with the subtitle How Our Obsession with Medical Labels Is Making Us Sicker) was written by Dr. Suzanne O’Sullivan and published in March 2025.
Dr. O’Sullivan is an NHS consultant neurologist specializing in epilepsy, widely acclaimed for her compassionate non-fiction books exploring the intersection of the brain, mind, and body (such as her previous award-winning book, The Sleeping Beauties).
Key Themes of the Book
The book acts as a critical but deeply empathetic examination of modern medicine’s shift toward hyper-classification. Dr. O’Sullivan explores what she calls “diagnosis creep” and “over-medicalization.”
-
Medicalizing the Normal: She questions whether ordinary human differences, natural life stages, and emotional variations are being unnecessarily turned into pathologies.
-
The Double-Edged Sword of Labels: While acknowledging that a clear diagnosis can bring immense relief and valid treatment pathways, she argues that some labels can inadvertently lock individuals into a fixed identity of “illness,” causing more harm, anxiety, and disability than the underlying process itself.
-
Case Studies Explored: The chapters meticulously analyze specific conditions that have seen modern explosions in diagnostic numbers or controversy, including:
-
ADHD and Autism (the massive rise in mild-to-moderate diagnoses).
-
Long Covid and Chronic Lyme Disease.
-
Hypermobile Ehlers-Danlos syndrome (hEDS) and PoTS (Postural Orthostatic Tachycardia Syndrome).
-
The impact of preemptive genetic and cancer screening.
-
Content Relevant to You Personally
For readers tracking modern healthcare trends in the UK, the book has drawn massive acclaim across British medical and mainstream media, including being featured as a Radio 4 Book of the Week and becoming a Sunday Times bestseller.
-
Perspective on Screenings: If you monitor residential wellness or age-related healthcare guidance, O’Sullivan’s chapters on genetic and preventative screening provide a fascinating, cautionary perspective on how “catching things early” sometimes leads to treating abnormalities that would never have caused harm in a person’s lifetime.
-
Navigating the Modern Medical System: The book is highly relevant for anyone looking to understand the contemporary NHS landscape, particularly how social media, patient self-diagnosis, and clinical boundaries intersect in the 2020s.
Are you interested in a specific chapter or theme from her book, such as her perspective on preventative screenings or how she approaches the rise in modern neurodivergent diagnoses?
Recommendations
Dr. O’Sullivan’s Structural Critiques and Proposed NHS Changes
Yes, Dr. Suzanne O’Sullivan addresses structural flaws within the NHS and broader medical systems. While The Age of Diagnosis is heavily focused on the philosophical, cultural, and psychological impacts of medical labeling, she outlines several practical changes and shifts in approach that she believes are vital for the sustainability of healthcare.
1. Rebalancing the System: A Return to Generalism
A core improvement O’Sullivan advocates for is reversing the extreme hyper-specialization that has come to dominate modern medicine.
-
The Problem: The current NHS framework frequently routes patients through multiple isolated specialty pipelines. For example, a patient experiencing complex, overlapping autonomic symptoms might simultaneously see a cardiologist, a gastroenterologist, and a neurologist, with none of them coordinating. This lack of continuity leads to redundant testing, conflicting advice, and fragmented care.
-
The Proposed Change: She calls for a significant reinvestment in “generalist” consultants and multi-disciplinary clinicians who can oversee a patient’s care holistically. A central coordinator ensures that individuals are not subjected to a revolving door of diagnostic tests for the same underlying issue, which often fuels anxiety and over-medicalization.
2. Overhauling Joined-Up Infrastructure
O’Sullivan directly points out that administrative and technical fragmentation directly worsens the overdiagnosis epidemic.
-
The Problem: Disconnected IT systems across different NHS trusts and regional care boards mean that medical histories, test results, and scan images do not seamlessly follow the patient.
-
The Proposed Change: True infrastructure integration. She argues that if clinicians had instant, unified access to a patient’s complete cross-boundary testing history, it would dramatically cut down on the “overdetection” cycle driven by repeating scans and blood panels simply because old data is inaccessible.
3. Decoupling Support Resources from Formal Medical Labels
One of her most systemic and controversial arguments addresses how public services, including the NHS and the education system, allocate aid.
-
The Problem: Currently, the system uses a diagnosis as a gatekeeper. To access extra time in exams, workplace adjustments, specialized school support, or specific disability benefits, an individual must secure an official diagnostic label from a doctor. This forces families and individuals into a desperate race for mild or borderline diagnoses just to unlock basic practical help.
-
The Proposed Change: O’Sullivan argues for a complete cultural and administrative overhaul where support is allocated based on functional need and vulnerability, rather than the possession of a medical label. She believes this would relieve immense pressure on diagnostic waiting lists and stop turning healthy people into lifelong “patients” just so they can access societal accommodations.
4. Shifting Beyond Biological Determinism to Social Solutions
O’Sullivan contends that by treating every human struggle as a purely biological or chemical “disease,” the medical establishment unintentionally shields society from addressing root causes.
-
The Problem: Attributing issues like mild depression, behavioral challenges, or sleep disturbances entirely to genetic or neurological deficits ignores environmental realities like childhood trauma, isolation, or poor social conditions.
-
The Proposed Change: She advocates for pushing resources toward social prescribing, community support, and systemic societal changes. Instead of medicalizing a natural life stage or stress response, the focus should shift toward building resilience and community networks that remind people what they can do, rather than reinforcing an inescapable identity of illness.
Content Relevant to You Personally
For anyone observing the current evolution of the NHS, O’Sullivan’s insights hit at the heart of the modern British healthcare debate—balancing resource constraints against an explosion in demand.
-
A Non-Austerity Perspective: O’Sullivan explicitly defends her stance against political misuse. She actively clarifies that her arguments against overdiagnosis are not designed to support austerity-driven budget cuts or the reduction of benefits. Rather, she argues that “diagnosis creep” causes the most severely disabled patients to have to vie for the scarce, essential resources they cannot live without, because the system is swamped with mild cases.
-
Managing Personal Health Expectations: Her emphasis on avoiding unnecessary medicalization for natural life transitions—such as normal age-related changes in sleep patterns or bone density—serves as a practical guide for how individuals can interface with the NHS today: focusing heavily on controllable lifestyle factors (exercise, nutrition) rather than demanding diagnostic interventions for expected biological shifts.
To hear Dr. O’Sullivan discuss these concepts firsthand, you can watch her WIRED Health Presentation on the Diagnosis Boom. In this talk, she reflects on her 35-year career as an NHS neurologist and explains why the modern drive to label every symptom can inadvertently compromise patient well-being instead of improving it.
NHS 2.0 with HITL GPs
- Phase 1: The Patient-AI Engagement (Unpacking and De-escalating)
- Phase 2: The Generalist’s Review (The Human Safety Net)
- Phase 3: Shared Decision-Making and Functional Support
Integrating a Human-in-the-Loop (HITL) service model into the NHS offers a highly practical solution to the exact challenges Dr. Suzanne O’Sullivan identifies in The Age of Diagnosis. By using an advanced AI system as the patient’s primary interface, and positioning an NHS generalist clinician as the “human in the loop,” the health service can balance the demand for answers with clinically assured, non-medicalized guidance.
This framework moves away from treating AI as an autonomous diagnostic machine and shifts it into a tool for holistic navigation, triage, and social prescribing.
Phase 1: The Patient-AI Engagement (Unpacking and De-escalating)
In a modern HITL framework, the patient’s initial interface is a conversational AI designed to gather information without immediately jumping to a medical label.
-
A Non-Pathologizing Interface: Unlike traditional symptom checkers that output a terrifying list of worst-case diagnoses, the AI is programmed to use neutral language. If a patient inputs symptoms of burnout, mild attention issues, or age-related biological shifts, the AI does not immediately suggest ADHD, PoTS, or early-stage disease. Instead, it maps out the environmental, situational, and physiological context.
-
Deep-Dive Virtual Journaling: The AI acts as an ongoing, highly accessible touchpoint. Over weeks or months, the patient can log variations in sleep, energy, stress, and physical sensations. The AI synthesizes this data into a comprehensive, longitudinal overview, removing the “recency bias” that often distorts a patient’s memory during a brief doctor’s appointment.
-
Differentiating “Signals” from “Noise”: Advanced algorithms process the patient’s inputs against rigorous clinical data to strip away the “overdetection” anomalies that Dr. O’Sullivan warns against. The AI distinguishes between acute markers that require medical intervention and standard bodily fluctuations driven by life circumstances or aging.
Phase 2: The Generalist’s Review (The Human Safety Net)
The AI does not issue diagnoses or treatment plans. Instead, it prepares a structured “Holistic Patient Brief” for the NHS Generalist (such as an advanced General Practitioner or a General Internal Medicine consultant).
-
Mitigating Hyper-Specialization: Before the clinician even sees the patient, the AI aggregates data across disparate physiological systems. If the patient has overlapping gut, heart, and neurological complaints, the generalist sees them mapped together on a single dashboard, preventing the patient from being fragmented across three separate specialty waiting lists.
-
The “Bullshit Filter” for Self-Diagnosis: The AI flags if a patient’s self-reported symptoms are heavily mirroring social media trends or algorithmic echo chambers. This alerts the generalist to focus the consultation on the actual functional distress the patient is experiencing, rather than debating a specific label the patient has found online.
-
Clinical Quality Assurance: The generalist reviews the AI’s synthesized data, exercising the irreplaceable human elements of clinical judgment, empathy, and contextual understanding. The clinician validates the accuracy of the AI’s data mapping, ensuring no red-flag symptoms have been overlooked.
Phase 3: Shared Decision-Making and Functional Support
Once the generalist reviews and adjusts the AI-generated brief, the interaction culminates in an assured, holistic plan delivered to the patient.
-
Decoupling Support from Labels: Aligning with O’Sullivan’s major structural recommendation, the generalist utilizes the AI’s analysis of functional impairment rather than diagnostic criteria to unlock support. The system generates a personalized care plan based on what the patient cannot do (e.g., struggling with focus at work or physical mobility at home) rather than waiting for a formal consultant diagnosis to grant accommodations.
-
Automated Social Prescribing: If the AI and generalist determine that the root causes are environmental, psychological, or lifestyle-driven, the system bypasses pharmaceuticals. The AI automatically matches the patient with local community resources, walking groups, exercise schemes, or psychological resilience hubs, while the generalist formally signs off on the plan.
-
The Closed-Loop Feedback: The patient returns to the AI interface to track their progress with the social or lifestyle interventions. If the functional impairment decreases, the loop successfully closes without the patient ever entering a chronic “illness identity.” If the data shows stagnation or worsening, the system alerts the generalist to escalate to targeted clinical interventions.
Content Relevant to You Personally
Implementing an AI-assisted, generalist-led HITL model fundamentally changes how individuals interact with public health systems, moving from a reactive “find the disease” approach to proactive, stable longevity management.
-
Optimizing Personal Health Navigation: In this model, routine monitoring—such as tracking metabolic changes, cardiovascular markers, or dietary responses—is entirely handled via the AI interface. For individuals managing their own wellness through high-protein nutrition, fasting, or active lifestyles, this system acts as a personalized data-gathering tool. It allows you to present a clean, verified data portfolio to an NHS GP, bypassing the standard 10-minute rushed consultation and enabling a high-level, collaborative discussion on long-term vitality.
-
Protecting Young Adults in Transition: For university students experiencing the intense cognitive and environmental pressures of higher education, this model provides an immediate safety net. Instead of entering multi-year waiting lists for formal neurodivergent or mental health diagnoses, an interactive AI can immediately assess their functional needs (such as sleep disruption or executive dysfunction). The generalist can then swiftly approve university accommodations and lifestyle scaffolding based on immediate need, completely avoiding the harmful “diagnosis creep” or the anxiety of a lifelong medical label during their formative years.
Would you like to explore how this HITL model could handle a specific medical scenario, or look into the current real-world pilot frameworks the NHS is using for AI clinical decision support?
Current NHS AI Pilots
Existing NHS Pilots (2025–2026 Landscape)
The NHS is currently moving out of an experimental “isolated pilot” phase and into systemic, policy-driven integration. Under the NHS 10-Year Health Plan and the newly published Neighbourhood Health Framework, the health service has mandated a decisive shift toward a unified digital front door.
Key real-world pilots and operational updates currently running include:
-
The NHS App “Total Triage” Shift: By mandate, the NHS is transitioning to a unified access model. The policy dictates that an overwhelming majority of primary care appointments must only be accessible after passing through an automated triage engine embedded directly within the NHS App.
-
Algorithmic Routing Engines (e.g., Klinik, eConsult): Real-world rollouts of these CE-marked, clinically regulated AI triage systems are active across hundreds of GP surgeries. Rather than a human reading every incoming request, the AI takes a structured patient questionnaire and automatically routes the patient. Musculoskeletal issues go directly to a physiotherapist, medication queries to a pharmacist, and urgent cases to a GP. In pilot practices, this has slashed telephone queues from 99% of entries down to 30%, drastically cutting down administrative waiting times.
-
Ambient AI and Automated Review Tools: In regions like Nottingham and London, pilots focus heavily on administrative de-escalation—using AI auto-review rules for low-risk pathology results, automated documentation generation, and ambient voice technology to record consultations.
Mapping Your Proposals to the NHS Reality: The Scope Alignment
The Human-in-the-Loop (HITL) model you outlined fits squarely into parts of the current NHS trajectory, but it also fundamentally diverges from current pilots by addressing the exact psychological and philosophical vulnerabilities highlighted by Dr. Suzanne O’Sullivan.
| Feature | Current NHS 2026 Pilot Scope | Your Proposed HITL Model Extension |
|---|---|---|
| Primary Goal | Operational Efficiency: Routing patients quickly to the cheapest/most appropriate staff member to clear backlogs. | Holistic De-escalation: Preventing over-medicalization and unpicking complex, cross-boundary symptom webs. |
| AI Interface Style | Transactional & Diagnostic: Asks structured, narrow medical check-box questions to assign an urgency tier. | Conversational & Contextual: Longitudinal virtual journaling over time, tracking life context alongside physical data. |
| The Human Role | The End-Point Specialist: The clinician receives a parsed task (e.g., a physiotherapist receives a back-pain referral). | The Central Conductor: An empowered Generalist sits in the middle, reviewing systemic data to provide unified, assured advice. |
| Allocation of Care | Label-Driven: Uses strict diagnostic markers and risk tiers to decide who gets an appointment. | Function-Driven: Uses AI-mapped functional impairment data to immediately deploy social prescribing and support. |
How Your Proposed Model Relates to and Improves Current Frameworks
If the NHS integrated your exact proposals into its current rollout, it would solve the three greatest bottlenecks currently threatening the success of the 2026 digital transition:
1. Moving from “Box-Ticking” Triage to Deep-Data Synthesis
Current NHS tools ask a static series of medical questions. If a patient logs multiple vague symptoms (fatigue, brain fog, joint pain), the current system often panics, marks them as high-risk, or generates fragmented tasks for multiple doctors. Your proposal changes the AI’s objective from sorting to synthesizing. By tracking the patient over weeks and evaluating the data across all bodily systems simultaneously, it presents the Generalist with a unified baseline, completely eliminating the “recency bias” of a single bad day.
2. Assuring Accuracy and Rebuilding Clinical Trust
A major anxiety among NHS clinicians is the rise of unverified, generalist AI black-boxes that pull information from open-access web sources, fueling patient self-diagnosis. Your proposal utilizes a closed-loop system where the AI is strictly bound to peer-reviewed, clinical-grade evidence guidelines, while leaving the final interpretive judgment exclusively to the Generalist. The AI provides the traceable data mapping and evidence logs, and the human provides the accountability. This creates an environment where doctors can actually trust the AI as a digital teammate rather than viewing it as a liability.
3. Operationalizing O’Sullivan’s Goal: Bypassing the Label
Right now, the NHS App triage pilots route patients to appointments to get diagnosed. Your proposal shifts the mechanism to route patients to adjustments. Because the AI tracks functional impairment (e.g., “Patient cannot sit for more than 20 minutes” or “Cognitive focus drops sharply post-midday”), the Generalist can immediately sign off on localized workplace or university accommodations and deploy automated social prescribing (community groups, lifestyle interventions), long before a formal, multi-year secondary care diagnostic pipeline is ever triggered.
Content Relevant to You Personally
Understanding how these current pilots map to an ideal HITL model has practical implications for managing healthcare expectations under the current system.
-
Navigating the Current “Total Triage” Reality: Because the NHS is actively rolling out these App-based triage engines, when you or your family interface with a GP practice, you will increasingly encounter these automated questionnaires. Knowing that the current system is looking for narrow “signals” to route tasks means that providing clear, objective data about daily operational function (rather than speculative labels) will yield the fastest, most accurate routing through the live system.
-
Advocating Within the System: If navigating care for a young adult at university, understanding that current NHS pilots are moving toward “routing to the right professional” means you can actively request access to a General Practitioner or an Internal Medicine generalist to look at a holistic picture, rather than allowing the current siloed IT infrastructure to split care across completely separate, uncoordinated hospital trusts.
Would you like to analyze a specific clinical scenario under this proposed model—such as how it would manage chronic lifestyle markers, or how it would handle a young adult presenting with complex fatigue and academic stress?