Budget Impact Models 101: Proving Your Device Saves the Health System Money

Ask a payer or a hospital value analysis committee for the single document that most influences an adoption decision and the answer is rarely the pivotal trial. It is the budget impact model: a short, transparent calculation of what happens to a specific budget holder's spending over the next one to three years if they adopt your device.

It is also the document most early-stage medtech companies either skip or outsource too late. This is what one contains, and how to build a credible first version yourself.

What a budget impact model is — and is not

A budget impact model answers an affordability question: given this population, this adoption rate and this price, what does total spending look like with the device versus without it, year by year, for the entity paying.

A cost-effectiveness analysis answers a different question: is the health gain worth the extra cost, usually expressed as cost per quality-adjusted life year. It is comparative and often societal, with a lifetime horizon.

Both are legitimate and they are not substitutes. We compared them directly in cost-effectiveness analysis versus budget impact model. The short version is that HTA bodies in Europe usually want both, while US hospital committees and commercial payers almost always start with budget impact, because their planning horizon is the fiscal year in front of them.

The six inputs every model needs

A budget impact model is built from six blocks. Everything else is presentation.

The eligible population. Start from the budget holder's own denominator — covered lives, admissions per year, tests performed per year — then narrow it to the indication, and narrow again to the subgroup where your device actually gets used. Overstating this is the most common credibility failure.

Current practice and its cost. What the pathway costs today, including the comparator device or service, staff time, follow-up and complications. Sourced from the buyer's own data where possible, published costing where not.

The new pathway and its cost. Your device price, plus everything it adds: training, integration, consumables, monitoring.

Clinical effect. What the device changes, stated as a rate difference with a source. Fewer readmissions, shorter stay, avoided imaging, earlier diagnosis. Each effect must trace to a study, not an assumption.

Uptake over time. Realistic adoption curve — rarely one hundred percent in year one, and often capped by capacity or clinician preference.

Time horizon and perspective. One to three years, and stated explicitly from whose budget. This single line resolves most arguments about whether a saving is real.

Building a first version

You do not need specialist software. A spreadsheet with one tab per block and every assumption on a visible input sheet is better than a black box, because buyers change assumptions and a model they can edit is a model they can believe.

Build the no-device scenario first: population times event rates times unit costs, summed per year. Then build the with-device scenario with the same structure and the modified rates. The budget impact is the difference, reported per year and cumulatively, gross and net of your device cost.

Then run one-way sensitivity analysis on each input. Change each by a plausible range and record how the result moves. This produces the most useful single output in the whole exercise: the short list of variables the case actually depends on. Very often it is not price. If your result is driven by a readmission-rate assumption you have not measured, you have just found the endpoint your next study needs, which is precisely the argument for planning HEOR before launch.

Framing it for the right budget holder

A saving only counts if it lands in the budget of the person you are asking to buy. A device that reduces post-discharge costs saves a payer money and may cost a hospital money, unless the hospital is at risk for readmissions. A tool that saves clinician time frees capacity but does not reduce the payroll line unless that capacity is redeployed to billable activity.

The discipline is to run the model separately for each stakeholder you sell to, and to be explicit when the benefit sits elsewhere. Buyers respect that far more than a blended figure that quietly attributes someone else's saving to them. The underlying payment structure matters too: inside a bundled rate, a hospital keeps what it avoids, which is why total cost of ownership framing and budget impact modelling reinforce each other.

Common mistakes

Modelling against an outdated comparator rather than current practice. Assuming instant full adoption. Using list price rather than net contracted price. Ignoring the one-off implementation costs, which makes year one look implausibly good and undermines everything after it. Hiding assumptions inside formulas. And presenting a single point estimate with no range, which signals either overconfidence or inexperience to anyone who reviews these for a living.

One more: treating the model as a marketing asset. Independent assessors now review device value claims publicly, and a model that cannot survive scrutiny becomes the thing reviewers cite against you. We wrote about that pattern in why digital therapeutics are failing independent HTA reviews.

Where to start this quarter

Draft the model before your next clinical protocol is locked, using literature values where you have no data. Identify the two or three inputs that drive the answer. Add those endpoints to the study you are already running. Then refresh the model with real data and localise it per account and per market.

MedTech Compass scores reimbursement route and evidence readiness alongside regulatory route across 25+ markets, so you can see where an affordability case is the binding constraint on entry. To see that applied to your product, book a walkthrough or read the market access guide first.

A budget impact model is not a finance exercise bolted on at launch. It is the cheapest early diagnostic you have for whether your evidence plan is pointed at the right question.

Sources

1. ISPOR — Principles of Good Practice for Budget Impact Analysis II: https://www.ispor.org/heor-resources/good-practices/report/principles-of-good-practice-for-budget-impact-analysis-ii 2. ISPOR — About Health Economics and Outcomes Research: https://www.ispor.org/heor-resources/about-heor 3. NICE — Health technology evaluations manual (PMG36): https://www.nice.org.uk/process/pmg36 4. NICE — HealthTech programme manual (PMG48): https://www.nice.org.uk/process/pmg48 5. CMS — Acute Inpatient Prospective Payment System: https://www.cms.gov/medicare/payment/prospective-payment-systems/acute-inpatient-pps 6. CMS — Medicare Coverage Database: https://www.cms.gov/medicare-coverage-database/search.aspx 7. AHRQ — Healthcare Cost and Utilization Project (HCUP): https://www.ahrq.gov/data/hcup/index.html 8. Peterson Health Technology Institute — independent assessments of health technologies: https://phti.org/assessments/

This article is general information about health economic modelling, not legal, regulatory or financial advice.