Command CenterLive · Primary Care · PCN Live
Today's Appointments
42
6 urgent · 11 chronic disease · 25 routine
Overdue QOF Actions
847
AI prioritised by risk score & deadline
Notes Auto-Drafted
38/42
91% accepted unchanged today
High-Risk Patients
23
Proactive intervention queue
🤖 AI Agent Activity — Today
10 agents working across clinical, operational, and population health layers
Clinical Notes Agent38 drafts · 91% accepted
QOF Intelligence Agent847 actions prioritised
Risk Stratification Agent23 high-risk flagged
Prescribing Safety Agent3 interactions flagged
Appointment Demand AINext 14 days optimised
Referral Intelligence Agent5 referrals drafted
📡 Live Clinical Feed
Real-time AI agent actions across your practice
Why DoctorOS — The Primary Care AI Gap
📋 Documentation Burden
GPs spend 3+ hours/day on clinical documentation — 37% of their working day. DoctorOS Clinical Notes AI drafts structured SOAP notes from ambient consultation audio, reducing documentation time from 8 min to 90 seconds per patient. EMIS/SystmOne auto-populated.
💊 QOF & PCN Targets
The average GP practice leaves £34,000 / ₹36L / $43,000 per year of QOF points unclaimed due to incomplete disease registers and missed review windows. QOF Intelligence AI identifies every outstanding action by patient, deadline, and opportunity value — proactively closing gaps before year-end.
📞 Appointment Demand
GP appointment demand peaks are poorly predicted, causing 8am phone queue chaos and under-used afternoon slots. Demand Forecasting AI predicts daily and weekly demand from historical patterns, seasonality, and population data — enabling intelligent slot release and right-sizing of on-the-day capacity.
DoctorOS AI Agents — Primary Care Intelligence Layer
📝
Clinical Notes Agent
Ambient audio → structured SOAP note → EMIS/SystmOne auto-populated. SNOMED CT coding. Read code mapping. Consultation summary + follow-up tasks generated in 90 seconds.
Running · 38 drafts today
ReAct + RAG
🎯
QOF Intelligence Agent
Scans entire patient population against all QOF indicators. Identifies gaps by disease area, priority, and financial opportunity. Generates patient-specific call/recall lists and clinical letters. Tracks year-end performance vs target.
Running · 847 actions
Reflection + Planning
⚠️
Risk Stratification Agent
EMIS/SystmOne data → risk scoring for CVD, diabetes complications, COPD exacerbation, frailty, and mental health crisis. Produces weekly high-risk patient list for proactive GP call. NICE guideline compliance built-in.
Running · 23 flagged
RAG + Planning
💊
Prescribing Safety Agent
Real-time drug interaction checking. High-risk drug monitoring (anticoagulants, NSAIDs, opioids, DMARDs). MHRA safety alerts. Dose optimisation by renal/hepatic function. polypharmacy review for patients on 10+ medications.
Running · 3 alerts
ReAct + Tool Use
🔬
Diagnostic Support Agent
Symptom constellation → differential diagnosis with supporting evidence. NICE Clinical Knowledge Summaries integration. Safety-netting recommendations. Referral threshold guidance. Red flag identification. All presented as supporting evidence — clinician decides.
Processing · 8 consultations
Reflection + RAG
📅
Appointment Demand Agent
14-day appointment demand forecasting from historical patterns, seasonality, acute episode rates, and population data. Intelligent slot release recommendations. Telephone vs face-to-face triage suggestions. Reduces 8am call queue by 47%.
Running · Next 14d forecast
Planning + ReAct
📧
Referral Intelligence Agent
Referral letter drafting from consultation notes. BPAS / e-Referrals integration. Right-sizing: consultant vs physio vs self-management. Referral outcome tracking and feedback loop to improve future referral quality. Waiting time intelligence by specialty and hospital.
Running · 5 drafts today
RAG + Tool Use
🏘
Population Health Agent
Practice-level population health dashboard. Disease prevalence vs national benchmarks. Deprivation-adjusted outcome tracking. Unmet need identification. Proactive recall campaigns for cervical screening, bowel cancer screening, immunisation, and chronic disease review.
Running · PCN dashboard
Planning + RAG
🛡
Safeguarding Intelligence Agent
Safeguarding concern identification from consultation patterns, A&E attendances, and multi-agency data. Section 17/47 referral support. MARAC flagging. Non-accidental injury pattern recognition. Every flag requires GP review before any action.
Monitoring · 0 today
Reflection + HITL
📊
SNOMED / Read Code Agent
Free-text clinical entries → structured SNOMED CT coding. Read code harmonisation for QOF and enhanced service reporting. Missed coding identification. Diagnosis confirmation from clinical correspondence. Average coding time reduced from 3 min to 8 seconds per entry.
Idle · On demand
RAG + Tool Use
Risk Stratification — Population Health Intelligence
Very High Risk
23
Proactive call within 48h
High Risk
147
Review within 2 weeks
Medium Risk
682
Annual review due
Stable
3,847
Routine management
Unregistered Risk
284
Missing data — review needed
🎯 Top Risk Domains — This Week
AI-identified highest-impact intervention opportunities
❤️
Cardiovascular — QRISK3 ≥20%
47 patients above NICE threshold — not on statin. NICE NG203 recommends intervention offer. Estimated 12 preventable cardiovascular events per year at current rate.
47 patients · High priority
🩸
HbA1c — Uncontrolled Diabetes
23 patients with HbA1c >75 mmol/mol. NICE NG28 intensification pathway indicated. Not seen for review in 12+ months. Proactive recall letters drafted.
23 patients · Urgent
🫁
COPD — Exacerbation Risk
31 patients with 2+ exacerbations in last 12 months. NICE NG115 specialist referral criteria met for 8. Rescue pack and action plan review needed for remainder.
31 patients · Review
🧠
Mental Health — Relapse Indicators
12 patients with SMI showing early relapse pattern (missed medication, cancelled appointments, A&E attendances). CMHT liaison initiated.
12 patients · Urgent
📈 QOF Performance — Year to Date
AI continuously tracks all QOF indicators against achievement thresholds
Hypertension — BP <140/9084% · Target 80%
Diabetes — HbA1c ≤58 mmol67% · Target 75%
COPD — Review in last 12 months71% · Target 70%
Asthma — Asthma review79% · Target 75%
Mental Health — Annual physical54% · Target 60%
CKD — Urine ACR tested61% · Target 70%
💡 AI Insight: Closing the HbA1c gap alone would generate £8,400 / ₹8.8L / $10,500 additional QOF income at current practice list size. AI has drafted recall letters for the 23 patients identified.
Prescribing AI — Safety, Optimisation & MHRA Compliance
Interactions Flagged
3
Today · All reviewed by GP
Polypharmacy Reviews Due
67
10+ medications · Structured review
High-Risk Monitoring
124
Anticoag · Methotrexate · Lithium
Prescriptions Today
187
3 modified by AI · 184 unchanged
🚨 Active Safety Alerts — Review Required
⚠️
Drug Interaction — Patient #PT-4721
Warfarin + newly prescribed Clarithromycin. CYP2C9 inhibition risk — significant INR elevation expected. MHRA interaction severity: Contraindicated. Recommend alternative antibiotic or INR monitoring within 48h.
Flagged by Prescribing Safety Agent · 09:14 · Awaiting GP review
💊
Renal Dose Adjustment — Patient #PT-2847
Metformin 1g BD prescribed. eGFR 32 ml/min/1.73m² (last tested 3 months ago). NICE recommends dose reduction to 500mg BD at eGFR 30-45. Review renal function and adjust dose.
Flagged by Prescribing Safety Agent · 10:32 · Awaiting GP review
🔍
High-Risk Drug Monitoring Overdue — Patient #PT-1093
Methotrexate 15mg weekly. Last FBC + LFT: 4 months ago. NICE and NHSE guidance: monitoring every 12 weeks minimum. Patient recalled — appointment booked.
Flagged by Prescribing Safety Agent · 08:47 · Appointment booked
📊 Prescribing Intelligence — Practice Dashboard
Antibiotic stewardship — items/STAR-PU0.96 (below national avg)
High-risk prescribing safety score94.2% (top quartile)
Generic prescribing rate96.8%
MHRA safety alert compliance100%
Polypharmacy reviews completed (Q3)47 / 67 due
Unintentional opioid dose escalation0 this quarter
💡 AI Insight: Your practice's antibiotic prescribing is 18% below the national average — an excellent patient safety indicator. Continuing education on appropriate prescribing is a key driver per NHSE GP contract incentives.
AgentOps — Live Agent Observability

📡 Live Trace Feed

📊 Session Metrics (24h)

Total Sessions847
Avg Latency1.4s
P95 Latency3.1s
Error Rate0.3%
HITL Escalations12
RAGAS GatePASS ✓

💰 Cost & Tokens

Cost (24h)£124 / ₹13,000 / $155
Input Tokens8.2M
Output Tokens2.1M
Cache Hit Rate71%
Cost/Session£0.15 / ₹16 / $0.19

🎯 RAGAS Quality Scores

Faithfulness0.94 ✓
Answer Relevance0.91 ✓
Context Precision0.89 ✓
Hallucination Rate0.8%
Context Recall0.93 ✓

🤖 Agent Health

All agentsHealthy
OrchestratorActive
Tool registryOnline
MCP serversConnected
Memory storeHealthy
Guardrails — Responsible AI & Patient Safety

🛡 Active Safety Rails

✅ Clinician-in-the-Loop (HITL)
Every clinical AI recommendation requires authorised clinician review and sign-off before any patient-facing action. AI cannot autonomously update patient records, prescribe, or refer. GDPR Article 22 — no fully automated consequential healthcare decisions.
🔍 Patient Data — NHS DSPT / HIPAA
All patient identifiers (NHS number, name, DOB, address) stripped via Presidio before LLM processing. Data never leaves NHS-approved environment. No cross-patient data in AI context windows. DSPT Standards Met compliance.
⚠️ Clinical Safety — DCB 0129
All AI clinical recommendations carry confidence score + uncertainty disclosure. AI never overrules treating clinician. High-risk flags (suicide risk, safeguarding, emergency) automatically escalated to senior clinician within 60 seconds.
📝 Immutable Clinical Audit Trail
Every AI interaction logged with: patient pseudonym, input context, AI output, clinician response, and outcome. 10-year retention. Exportable for CQC, GMC, GDC, or NMC investigations.

📋 Regulatory Compliance Mapping

GDPR Art. 22 — Automated decisionsHITL enforced
NHS DSPT — Data SecurityCompliant
MHRA SaMD — Advisory onlyExempt
DCB 0129 — Clinical riskDocumented
EU AI Act — High-risk AIMapped
NIST AI RMFAligned
ISO/IEC 42001Compliant
0.8%
Hallucination Rate
Target <2%
100%
HITL Coverage
0
PII Leaks (30d)
A+
Security Grade
Documentation — Deployment Guide & Runbook

🚀 10-Week Deployment Guide

1
Week 1–2: Data Foundation & Integration
Connect to clinical systems (EMIS/SystmOne/Dentally/RiO). Establish NHS DSPT data governance. Configure RBAC roles (admin, clinician, approver). Run baseline evaluation on golden dataset. Confirm DCB 0129 clinical risk assessment.
2
Week 3–4: Core Agents Live
Deploy first 3 agents (highest-value use case for your setting). Wire HITL approval workflows in Temporal. Configure NeMo guardrails and patient data scrubbing. Set up Langfuse tracing and RAGAS eval gate. Clinician demo with real data.
3
Week 5–7: Full Agent Mesh
Deploy remaining agents. Configure Orchestrator routing. A/B test prompt variants. Enable drift detection. Train clinical staff on HITL workflow and approval interface. Clinical governance committee presentation.
4
Week 8–10: Production Hardening
Pen test + security scan. Load test to 3× peak concurrent users. Configure PagerDuty clinical alert routing. CQC/GMC compliance review. Caldicott Guardian sign-off. Produce runbook and clinical handover documentation. Go-live.

🏗 7-Layer Clinical AI Stack

L7PresentationClinical UI · SSO · RBAC · Mobile
L6API GatewayFastAPI · OAuth2 · Rate Limit · WAF
L5OrchestrationTemporal.io · LangGraph · HITL signals
L4Agent RuntimeNeMo Guardrails · RAGAS · Tools
L3Model + ToolsClaude API · MCP servers · Function calls
L2Data + IntegrationKafka · PostgreSQL · Pinecone · Redis
L1ObservabilityOTel · Langfuse · Prometheus · Grafana