Budget impact

Estimate annual spending changes when an intervention replaces part of an existing treatment mix.

Method lesson · Updated October 6, 2026

Concepts and reasoning

Affordability is a different question from value

A budget-impact analysis asks how spending changes when an intervention enters a particular health system over a specified period. The central outcome is a monetary difference between a current scenario and a future scenario. A treatment may provide good value per QALY yet require a large near-term increase in expenditure when many people become eligible.

The relevant perspective is the budget holder whose spending is being assessed. A hospital, insurer, national program, and employer may pay for different resources. Savings outside that budget can be important to society without being available to finance the organization’s acquisition costs. Identify where costs occur and when money actually leaves or returns to the budget.

Eligibility is not uptake

Start with the population covered by the budget, then identify disease prevalence or incidence, diagnosis, clinical eligibility, access, and adoption. These are different filters. Multiplying the total population by a treatment price without identifying eligible users can overstate spending; assuming every eligible patient adopts immediately can do the same.

Adoption can also displace existing treatment rather than simply adding another cost. Some patients may switch from a more expensive option, while others receive treatment for the first time. The appropriate comparison is between complete treatment mixes in two scenarios. Population entry, discontinuation, mortality, and repeated treatment can make annual counts evolve.

Resource-impact work should document these local assumptions and allow decision makers to vary them. NICE’s resource-impact materials illustrate the practical budget-holder perspective.

Mathematical foundations

On a small screen, scroll wide equations and tables horizontally.

Build each scenario before subtracting

For year , let be the eligible population, the share receiving treatment , and its relevant per-person cost. A simple scenario total is

Here represents fixed implementation costs included in that scenario. Shares should cover the intended population, including an untreated category when appropriate. Adding costs for every option without applying shares would charge the same population multiple times.

For a simple replacement affecting fraction , the variable budget change reduces to . This compact expression is useful only if the new treatment replaces the old one on the same basis. It may fail when treatment duration, adherence, eligible population, or downstream events differ.

Keep recurring acquisition costs separate from one-time training, equipment, or service redesign. A cost offset should be counted only if it falls within the chosen budget and time horizon. Reducing hospital days may free capacity without reducing cash expenditure by the full accounting cost of those days.

Worked example

Hypothetical data for learning; independent of your analysis inputs.

Follow adoption over three years

Suppose 2000 people are eligible each year. Current annual treatment costs 500 per person. A new treatment costs 800, including the resources relevant to this budget. It replaces current treatment for 10%, 25%, and 40% of eligible people in years 1–3. A one-time implementation expense of 50,000 occurs in year 1.

The incremental variable cost is 300 per switched person. Therefore, year-1 variable cost rises by . Adding implementation gives 110,000. Year 2 adds 150,000 and year 3 adds 240,000.

Year	Uptake	Switched people	Incremental variable cost	Total budget impact
1	0.100	200	60000.000	110000.000
2	0.250	500	150000.000	150000.000
3	0.400	800	240000.000	240000.000

The undiscounted three-year increase is 500,000 under these assumptions. It is not the cost of treating all eligible people with the new therapy; it is the incremental difference caused by the stipulated adoption.

If the new treatment avoids 100 per switched person in costs that this same budget truly bears within the year, the net variable difference becomes 200. The total then becomes . If those savings occur after year 3 or accrue to another organization, they do not belong in this three-year budget total.

Interpreting the results

Separate affordability from value for money

Budget impact compares spending under a current treatment mix with spending under a proposed mix over specified calendar years. A positive incremental budget means additional expenditure under the proposed assumptions; a negative value means modeled savings. Neither sign alone establishes whether an intervention is cost-effective.

Cumulative impact adds annual differences. A small first-year effect can coexist with a large cumulative effect when uptake grows or the eligible population expands. Conversely, an upfront expense may be followed by savings, making the timing of cash flows important to the decision maker.

The relevant denominator is the population eligible for the budget, not necessarily the trial population. Report population size, uptake, unit costs, implementation expenses, and the payer perspective together. A credible per-person estimate multiplied by an unrealistic eligible population can still produce a misleading budget.

Assumptions and limitations

Define who enters the budget each year

Eligibility, market share, treatment duration, and service use must match the budget holder's responsibilities. Distinguish incident patients from prevalent patients and avoid counting the same treatment episode twice. A simple annual cohort calculation does not automatically track patients entering, stopping, or switching treatment across years.

Uptake is a share between zero and one, with a clearly defined alternative when the new treatment is not used. Unit costs should use a consistent currency and price year. Acquisition costs, administration, monitoring, adverse events, and displaced services may belong in different components rather than a single unexplained price.

Capacity constraints and implementation delays may limit feasible adoption. Cost offsets may occur later than treatment expenditure. Assumptions about those timings should be explicit; annual arithmetic cannot establish that the required staff or facilities are available.

Diagnostics and sensitivity analysis

Reconcile one year before projecting several

Recalculate a single year from eligible population, treatment shares, and annual per-person costs. Compare it with current expenditure if such data are available. Investigate differences caused by eligibility definitions, treatment duration, discounts, or missing resource categories.

Explore slower and faster uptake, alternative eligible-population estimates, and plausible unit-cost changes separately. Then combine coherent scenarios rather than mixing the highest cost from one setting with an incompatible uptake assumption from another.

Check whether cumulative impact is simply the sum of the displayed annual differences and whether a large change has a clear driver. Report consequences for staff, beds, or procedures alongside money when these affect implementation. A budget result without a feasible service plan may overstate what can actually be delivered.

References and other material

Read the methodological references and supporting tutorials below. The examples in this lesson are original teaching calculations, not findings from these sources.

NICE: assessing resource impact — Free resource-impact materials and examples in the UK health-system context.