Predictive Demand & Capacity Modelling
Integrating geo‑spatial, demographic and epidemiological data, we forecast patient cohorts, disease burden and service demand across multiple time horizons.
Forecasts are stress‑tested under funding and workforce scenarios and linked to costed investment options so decision‑makers can plan resilient capacity and budgets with confidence.
Current State
The government has committed to a 92% 18 week Referral to Treatment (RTT) standard by the end of the parliament in 2029. While there have been incremental improvements - for example RTT performance moved from 58.8% to 61.4% between July 2024 and January 2026 - current trends indicate further progress will be required to reach that ambition.
Other system measures continue to face pressure: A&E four hour performance was 74.1% (Feb 2026) against the 95% standard, diagnostics met roughly 75% of the six week standard (Jan 2026), and many hospitals are operating with limited spare capacity.
Planning remains primarily focused on responding to acute demand and reducing backlogs, creating an opportunity to complement this with greater emphasis on upstream prevention and pathway optimisation to improve resilience and reduce emergency admissions.
Our approach
Data driven insights
Integrating geo spatial, demographic and epidemiological data to explore regional trends and patient level patterns that directly inform targeted capacity and service planning.
Predictive modelling
Forecasting future patient cohorts, chronic disease burden and service demand (e.g. A&E, outpatient, elective surgery) across short and long term horizons to plan capacity and prioritise interventions.
Scenario planning
Stress testing workforce and funding allocations under alternative population and health need scenarios to quantify risk and prioritise resilient options
Funding optimisation
Linking demand forecasts to cost models to identify efficient funding choices and reduce the risk of over or under investment.
Dynamic monitoring
Reconcile forecasts with live hospital activity and regional data so planning and operational teams have current, actionable intelligence to track performance and manage services across provider networks.
Illustrative Example: Managing the 18 Week RTT Target
How Predictive Demand & Capacity Modelling could change the outcome
Background
Referral to treatment (RTT) (18 week) is a core operational and financial benchmark for elective care. Early progress relied on short term surge programmes rather than sustainable capacity balance. Population ageing, rising long-term care (LTC) prevalence and post COVID referral growth increased demand materially. By 2024-25, performance fell to c.60%6 and waiting lists expanded across multiple pathways. The result: growing backlogs, higher cancellations and strained theatre, diagnostics and bed capacity.
Why is it failing?
Demand capacity mismatch: demand grew faster than workforce, theatres and diagnostics could expand. Reactive planning: emphasis on clearing backlogs after breaches rather than anticipating referrals. Target-driven behaviours: urgent focus on short term threshold management rather than pathway optimisation. Siloed decision making: poor coordination between specialties, system partners and funding levers Limited surge resilience: insufficient seasonal buffers and no timely early warning triggers.
What can be done?
Predictive demand forecasting: time series and advanced statistical models to anticipate referrals, seasonal peaks and backlog growth. Actuarial capacity modelling: quantify uncertainty and costed workforce/theatre/diagnostic requirements under multiple scenarios. Queue & pathway simulation: test how prioritisation, productivity and capacity changes affect RTT and flow. Early warning systems: Forward projections of RTT breaches enabling proactive intervention, rather than retrospective performance management. Process & funding optimisation: link forecasts to costed investment choices and pilot targeted, high impact interventions
A systematic deployment of predictive analytics and actuarial capacity planning provides early warning, realistic targets and prioritised investments to prevent sustained backlog growth
Future State
The system moves from backlog management to measurable recovery: targeted, data driven capacity plans and predictive models accelerate elective throughput, reduce cancellations and shorten waits, with RTT and diagnostic performance improving as capacity is matched to demand
A&E flow stabilises when upstream capacity, discharge pathways and seasonal buffers are coordinated; bed use is managed with defined surge capacity so hospitals retain resilience through peak periods. Diagnostics backlogs fall through smarter scheduling, targeted investment and pragmatic use of additional capacity where it delivers fastest throughput.
Real time management: Forecasts are linked to costed investment options and regularly recalibrated against recent activity, enabling earlier detection of pressure and prioritisation of high impact interventions.
Align capacity plans to workforce strategy: once models identify required staff types and volumes, we support targeted attraction and retention so capacity gains are sustainable.
The result is better patient experience, more efficient use of NHS budgets and a durable, data driven basis for longer term population health improvement.
© 2026 Marsh LTD. All Rights Reserved. Designed & built by the Impact team.