Cost-Effectiveness vs. Budget Impact: Why “Good Value” Does Not Always Mean “Affordable”

A healthcare intervention can represent good value for money and still create a budget problem.

That statement can sound contradictory until two different questions are separated: is the intervention worth its additional cost relative to its additional health benefit, and can the healthcare system absorb the resulting expenditure over the next few years?

The first is primarily a cost-effectiveness question. The second is a budget impact question. They are related, but they are not interchangeable.

This distinction matters whenever a payer, hospital, health system, employer, or public agency evaluates a new treatment, diagnostic strategy, care pathway, or technology. ISPOR describes budget impact analysis as an important component of comprehensive economic assessment alongside cost-effectiveness analysis and emphasizes that BIA should reflect the population, treatment mix, resource use, costs, and information needs of the particular healthcare decision maker.

Understanding the difference prevents a common analytical mistake: treating “cost-effective” as if it meant “inexpensive.” It does not.

Cost-Effectiveness Asks Whether Additional Benefit Justifies Additional Cost

A cost-effectiveness analysis compares alternatives in terms of both costs and outcomes.

When health outcomes are measured in quality-adjusted life years, or QALYs, the result may be expressed as additional cost per additional QALY.

The new intervention costs an additional $8,000 and produces an additional 0.40 QALYs.

This number does not tell the decision maker whether the intervention is affordable. It tells the decision maker about the incremental relationship between cost and health benefit under the model assumptions. Whether that represents acceptable value depends on the relevant decision framework, jurisdiction, perspective, uncertainty, and other considerations. NICE, for example, uses cost-utility analysis with incremental analysis in its reference-case framework and specifies QALYs as its measure of health effects.

Budget Impact Asks a Different Question

Now imagine the same intervention will be offered to 100,000 eligible patients. If the incremental treatment cost remains $8,000 per treated patient, immediate treatment of all eligible patients would create gross incremental treatment expenditure of:

The intervention’s ICER has not changed. It is still $20,000 per QALY in our simplified example. Yet a payer suddenly faces the possibility of hundreds of millions of dollars in additional expenditure. That is the difference between value and affordability.

The Denominator Changes the Decision

A simplified cost-effectiveness model asks what happens to costs and health outcomes for patients receiving Strategy A versus Strategy B. A budget impact model asks how many eligible people there are, how quickly the new strategy will be adopted, what it will replace, and what all of that will cost the decision maker over the budget horizon.

That introduces variables that may have little effect on the ICER but enormous effects on affordability:

• eligible population size;
• uptake rate and market share;
• implementation speed and treatment displacement;
• acquisition, administration, and monitoring costs;
• downstream resource use, discontinuation, and capacity constraints.

ISPOR recommends that budget impact analyses reflect the size and characteristics of the population, current and new treatment mix, efficacy and safety, resource use, and costs relevant to the decision maker.

A Worked Three-Year Budget Impact Example

Consider a hypothetical health plan with 100,000 eligible patients. Assume an incremental intervention cost of $8,000 per treated patient, adoption of 10% in Year 1, 25% in Year 2, and 40% in Year 3, and downstream healthcare savings of $1,500 per treated patient in each modeled year.

Year 1

Patients treated: 100,000 × 10% = 10,000. Additional intervention expenditure: 10,000 × $8,000 = $80 million. Estimated downstream savings: 10,000 × $1,500 = $15 million. Net budget impact: $65 million.

Year 2

Patients treated: 100,000 × 25% = 25,000. Additional intervention expenditure: $200 million. Estimated downstream savings: $37.5 million. Net budget impact: $162.5 million.

Year 3

Patients treated: 100,000 × 40% = 40,000. Additional intervention expenditure: $320 million. Estimated downstream savings: $60 million. Net budget impact: $260 million.

Cumulative illustrative net budget impact is $65M + $162.5M + $260M = $487.5 million. Nothing about this calculation contradicts the earlier cost-effectiveness result. The intervention can simultaneously have an attractive cost per QALY and require substantial incremental spending under these assumptions.

Cost-Effectiveness and Budget Impact Can Point in Different Directions

Intervention B is especially important: a large eligible population can turn a reasonable per-patient incremental cost into substantial aggregate expenditure. Neither metric substitutes for the other.

Perspective Determines Which Costs Belong in the Model

Economic models are not built from a universal list of costs. The relevant costs depend partly on perspective. A payer may focus on reimbursed healthcare expenditures. A hospital may care about staffing, bed capacity, procedure costs, transfers, and operational resources. A societal analysis may consider additional consequences outside the healthcare budget, depending on the methodological framework.

This is why perspective should be specified before calculations begin. NICE’s health technology evaluation framework explicitly specifies perspectives for health outcomes and costs, while ISPOR recommends constructing BIA around the information needs of the specific budget holder.

Time Horizon Can Change the Story

Cost-effectiveness models often need sufficiently long horizons to capture important differences in costs and outcomes. NICE states that the time horizon should be long enough to reflect all important differences between the technologies being compared.

Budget holders may face a different operational reality. Annual budgets are finite. An intervention might create large expenditures today while generating health or resource benefits years later. A long-run economic evaluation can incorporate those future consequences, while the finance team still needs to know what happens to next year’s budget. This is where simulation and decision modeling can help make uncertainty explicit. Neither view is inherently more correct; they answer different decisions.

Budget Impact Is Not Simply ICER Multiplied by Population

A tempting shortcut is ICER × number of patients = budget impact. That is generally not the right calculation. An ICER contains incremental costs and incremental health effects in its denominator. Budget impact requires actual monetary flows under current and future treatment scenarios.

1. Population
2. Eligibility
3. Current Mix
4. Future Mix
5. Resource Use
6. Cost

A basic structure is closer to: Budget Impact = Cost under New Treatment Mix − Cost under Current Treatment Mix. If 20% of patients adopt a new intervention, what were those patients receiving before? Did the new treatment replace another expensive therapy, get added on top of existing care, reduce hospitalization, or require new monitoring? Those questions directly affect the budget calculation.

Scenario Analysis Matters Because Uptake Is Uncertain

Suppose the base case assumes Year 3 adoption of 40%, but actual adoption could plausibly be 20% or 60%. Using the same simplified assumptions, the net annual impact is $130 million at 20% uptake, $260 million at 40%, and $390 million at 60%.

The range is enormous. That does not mean the model failed; it means adoption is a decision-relevant uncertainty. ISPOR recommends scenario analyses for budget impact models and emphasizes assumptions relevant to the decision maker.

Sensitivity Analysis Asks Which Assumptions Drive the Decision

Not every uncertain parameter matters equally. A model can vary eligible population by ±20%, intervention cost by ±10%, adoption by ±50%, and downstream savings by ±30%. The resulting budget impact might be most sensitive to adoption and eligible population.

Instead of reporting only “our estimate is $260 million,” the analysis can say that the base-case Year 3 estimate is $260 million, while the modeled low- and high-uptake scenarios range from $130 million to $390 million. Economic modeling should expose uncertainty rather than hide it behind decimal precision.

Affordability Is Not the Same as Causal Effectiveness

Budget models inherit the strength and limitations of their clinical evidence. If observational data show that patients receiving a new care pathway have fewer hospitalizations, that association does not automatically establish that the pathway caused the reduction. Patients selected into the pathway may differ in disease severity, access to care, clinician behavior, or other characteristics.

If the budget model assumes hospitalization savings caused by the intervention, the evidence supporting that treatment effect matters. A transparent model should distinguish observed association from an assumed causal effect used in the economic model.

Cost Savings Are Not Required for an Intervention to Be Cost-Effective

A new intervention may cost more and improve health enough to represent acceptable value under the applicable decision framework. That is different from being cost-saving, where the new strategy both improves outcomes and reduces costs.

• Cost-saving: Better outcomes and lower costs.
• Cost-effective: Additional benefits may justify additional costs.
• Affordable: Aggregate spending fits the relevant budget constraints.

An intervention can satisfy one, two, or all three. These terms should not be used interchangeably.

Budget Impact Becomes Especially Important for Large Populations

The CDC reports that more than 795,000 people in the United States experience a stroke each year, and stroke remains a major cause of serious long-term disability. Current CDC reporting also describes heart disease and stroke as major drivers of health and economic burden.

For conditions affecting large populations, even modest per-patient changes can scale into substantial system-level financial consequences. An average incremental cost of only $500 across 500,000 patients becomes:

Scale therefore belongs explicitly in healthcare decision models. Per-patient economics alone do not describe the financial consequences of widespread implementation.

Decision Models Should Show Both Value and Affordability

A useful healthcare economic dashboard should present both a cost-effectiveness result and a budget forecast. A decision-oriented summary might include:

• clinical outcome difference;
• incremental cost per patient;
• incremental cost-effectiveness;
• eligible population and expected uptake;
• annual budget impact;
• uncertainty range and capacity implications.

This moves the discussion from “Is it cost-effective?” to “What value does it create, what will it cost us, how certain are we, and can we implement it?” That is a better decision question.

From Economic Evaluation to Healthcare Decisions

Health economics is most useful when it clarifies trade-offs rather than compressing them into one number. An ICER can provide information about incremental value. A budget impact model can provide information about affordability. Scenario and sensitivity analyses can show how much the answer depends on uncertain assumptions.

The decision is therefore rarely simply whether an intervention is economically good or bad. More useful questions are: Does it improve outcomes? What additional resources does it require? Is the incremental benefit worth the incremental cost? What happens to the actual budget as adoption grows? Which assumptions could change that conclusion?

When those questions are modeled together, economic evaluation becomes more than a reimbursement calculation. It becomes a framework for making healthcare resource decisions under uncertainty.