← Back to blog

Healthcare Savings Calculation for Employers: A Step-by-Step Method

November 7, 2026
Healthcare Savings Calculation for Employers: A Step-by-Step Method

To calculate healthcare savings, compare what your plan actually paid after a program started with what it would likely have paid without it. Set a baseline from 12 to 24 months of claims, trend it forward, subtract actual cost per member per month (PMPM), then divide the net result by program cost to get ROI. The rest of this guide shows each step with a worked example, the data you need, and the mistakes that inflate savings numbers.

Hadaco
Make Healthcare Savings Measurable
Hadaco complements existing health plans with evidence-based programs and quarterly reporting to help employers track workforce health and healthcare costs.
Visit Hadaco

TL;DR

  • Build the baseline from at least 12 months of fully paid medical and pharmacy claims, and 24 months if you can get them.
  • Savings equal the trended baseline PMPM minus actual PMPM, multiplied by member months. Net it against program cost.
  • ROI is (benefits − costs) ÷ costs × 100. Payback is the time until cumulative savings cover your outlay.
  • Count only savings you can defend to your CFO: adjust for plan changes, large claims and enrollment shifts.
  • Keep HSA and Section 125 payroll tax effects on a separate line from claims savings.

Table of Contents

Why a clear savings calculation matters

Most employers cannot say with confidence what a health initiative returned. A Wharton working paper notes that most companies cannot accurately measure the benefits and costs of health improvement programs. Leaders then fall back on intuition. The same paper cites a survey in which two-thirds of CFOs said they would consider a program with a return of at least 8%. A CFO will fund what can be measured and defended.

A good calculation does three jobs. It tells you whether a program is working. It gives finance a number built on the same logic as the rest of the budget. And it makes next year's decisions easier, because you know which levers moved cost.

Step 1: Set your baseline

Everything else depends on the baseline. The SHRM and HERO/PHA measurement guide says baseline data should include a minimum of 12 months, or 24 months for a more solid baseline, of membership, eligibility, medical and pharmacy claims. Lab or biometric results help if you have them. Two requirements are easy to miss:

  • You must be able to link the same individuals in the baseline to those in the program years.
  • A dollar-based analysis needs fully adjudicated claims, so allow for claims runout before you calculate anything.

Chapman Institute recommends preserving a baseline of at least 12 months, preferably 24 to 36 months, with appropriate privacy protections. Use employer-paid medical and pharmacy claims net of rebates, not billed charges.

Cap large claims

One transplant or neonatal intensive care stay can swamp a year of data. A Milliman review describes capping claims at a selected large-claim truncation point before calculating the baseline PMPM, then trending it with a published trend rate to create a benchmark for each program year. Pick the truncation point before you see results, and apply it to both periods.

Step 2: Express cost per member per month

Total spend is a poor yardstick because headcount moves. PMPM fixes that. HealthPartners' technical guidelines define the calculation using all medical and pharmacy claims in a 12-month period plus three months of paid claims runout, with a minimum of nine months of enrollment for a member to count. They risk-adjust by dividing PMPM by the relative risk score.

For an employer calculation, the working version is:

PMPM = (medical paid + pharmacy paid) ÷ member months

Member months are the number of members enrolled in each month, added across the year. A plan with 1,000 members all year has 12,000 member months.

Step 3: Trend the baseline forward

Healthcare costs rise every year without any program. If your baseline PMPM was $450 and you simply compare it with this year's actual, you will misstate the result. Trend the baseline using a published rate, or your own pre-program history, so you compare the actual to what was expected.

The HERO/PHA guide recommends five savings methodologies for directly monetized claims:

  1. Cost trend compared with industry peers
  2. Adjusted-expected versus actual cost trend
  3. Chronic versus non-chronic trend comparison
  4. Participant versus non-participant trend comparison
  5. Comparison with matched controls in a non-exposed population

Most mid-sized employers start with the second. It needs only your own claims and a defensible trend assumption.

Step 4: Calculate the savings, then ROI and payback

Here is a hypothetical employer with 1,000 covered members. The figures are illustrative only, not a result from any client.

LineFigure
Baseline PMPM (12 months before program)$450
Assumed annual trend6%
Trended benchmark PMPM$477
Actual PMPM in program year$455
Savings PMPM$22
Member months12,000
Gross savings$264,000
Program cost$90,000
Net savings$174,000

Now apply the standard formulas.

ROI is (program benefits − program costs) ÷ program costs × 100 in the usual form. Here that is ($264,000 − $90,000) ÷ $90,000 × 100, or about 193%.

Benefit-cost ratio is benefits ÷ costs. SHRM's guide offers a worked example: a blood-pressure control program producing $50,000 in savings against a $20,000 annual intervention cost gives a ratio of 2.5 to 1. In our example, $264,000 ÷ $90,000 is about 2.9 to 1.

Payback period is how long cumulative savings take to cover what you spent. Benely describes a three-part model: an initial balance for year-one outlay and immediate savings, an annual balance for later years, and time to recovery. If savings accrue at about $22,000 a month and the outlay is $90,000, you recover the cost in roughly four months of full-rate savings. Savings usually ramp up, so model the real monthly curve.

A Milliman review specifies ROI as employer net PMPM savings divided by employer PMPM program cost, looking at both year-specific and cumulative figures. Report both. Year one often carries startup effort, and cumulative ROI tells the longer story.

Step 4: Calculate the savings, then ROI and payback

Savings per employee

Divide net savings by average covered employees, not members, if your CFO thinks per head. In the example, $174,000 across 1,000 members is $174 per member. Say which denominator you used.

Published studies give a sense of scale. A 2024 study of a cardiovascular risk-reduction program reported annualized cost savings of $1,224 per individual and a $4.90 return per dollar spent, with much higher returns in at-risk subgroups. Results depend on who is enrolled and what the program targets, so treat any published figure as context for your own numbers, not a forecast.

Step 5: Separate realized savings from projections

The key difference is between money that stopped leaving your plan and money that a model says you avoided. Chapman Institute frames the central question as what would probably have happened without the program. It lists randomized rollout, matched comparison, difference-in-differences and interrupted time series as ways to answer it, and advises counting an avoided claim only when the analysis supports causation.

In practice, keep three buckets:

  • Realized: actual PMPM below the trended benchmark, after risk and plan-design adjustments.
  • Projected: modeled from risk reduction, such as fewer expected hospital admissions, valued with published evidence. Label it as an estimate.
  • Soft: retention, absenteeism and productivity. Real, but kept apart from claims dollars.

The HERO/PHA guide also suggests reporting the monetized impact on potentially preventable hospitalizations, ER visits and procedures. These are a useful bridge between participation data and claims results.

Step 6: Decide which costs and factors belong in the model

On the cost side, include fees, incentives, staff time and any vendor charges. On the benefit side, ask what the intervention has to achieve. A JAMA Health Forum analysis of prevention economics says three things matter: the intervention's cost, how expensive the downstream care it averts is, and how many people must be treated to avert one event. The authors give examples such as roughly $1,000 for a Medicare emergency department visit in 2021 and almost $22,000 for heart failure treatment in the first 90 days in 2018.

That is a practical filter. Programs aimed at chronic conditions that lead to expensive events, such as heart failure, are where your claims data is most likely to show movement.

Adjustments to apply

Chapman Institute lists the adjustments to make: inflation, benefit changes, enrollment duration, risk differences and unusually large claims. Add a short note for each in your report. If you changed deductibles in the program year, say so and quantify the effect.

Step 7: Account for participation and group size

Savings follow participation. The HERO/PHA guide gives minimums for a rigorous study: 50% health assessment or screening participation, 30% enrollment into targeted coaching and 15% or more in a population-wide program, typically with two years of data. It also notes that when membership is well below 25,000, running all five primary savings metrics is generally not advisable, though valid results are sometimes possible with far fewer members.

Program participation benchmarks and study guidance

For a mid-sized employer, that argues for choosing one or two methods and reporting a range rather than a single point. It also argues for tracking participation every quarter, since it explains most of the variation in results.

Program type and employer size also shape returns. A scoping review found that disease-management components produced higher ROI than general wellness components, and that ROI rose with program length. It also found larger companies had higher ROI than companies with 500 or fewer employees. The same review found that more rigorous evaluations tend to report more conservative figures. That is a reason to build your model carefully: a number that survives scrutiny is the one finance will keep funding. Benchmarks for related benefits, such as EAP ROI for employers, show the same pattern of results varying with program design.

Step 8: Keep HSA and Section 125 savings separate

Plan design can produce savings that never appear in claims. When employees fund HSAs through payroll under a cafeteria plan, employers benefit through reduced employer FICA and FUTA taxes. RSM lists 2026 HSA contribution limits of $4,400 for self-only coverage and $8,750 for family coverage, plus a $1,000 catch-up at age 55 and over.

The cafeteria plan matters. Thomson Reuters' EBIA analysis explains that HSA contributions made through a Section 125 plan are excluded from employee gross income and from FICA, FUTA and RRTA taxes, while payroll deductions without one are treated as taxable compensation. Have your tax advisor confirm the calculation. Then report it as its own line, so it does not blur the claims-based result.

Step 9: Report on a schedule and keep the audit trail

A calculation is only useful if it repeats. A workable cadence:

  1. Monthly: pull paid claims and enrollment to track PMPM.
  2. Quarterly: compare against the trended benchmark, update participation, and note claims runout still to come.
  3. Annually: finalize with fully adjudicated claims, show year-specific and cumulative ROI, and refresh the baseline assumptions.

Show your assumptions on one page: baseline period, truncation point, trend rate, adjustments and the denominator for per-employee figures. When a number changes, the reader should see why.

Our perspective

We built Hadaco around one idea: an employer should be able to see the savings math before and after a program starts. That is why we pair a transparent savings estimator with quarterly reporting of outcomes.

The calculation in this guide is what we think every employer should expect from a health partner: a clear baseline, a benchmark that accounts for trend, and reporting that finance can audit. A program with a plausible estimate and a quarterly checkpoint beats a program with a glossy promise and no accountability.

We also think the model should cover the whole person: chronic disease, preventive care, mental health and engagement. Savings come from several places at once, and our programs complement your existing benefit plans instead of replacing them. Our service fees are performance-based, contingent on demonstrated savings and engagement outcomes, with no upfront fees.

Gene

Put the method to work

If you want an estimate before you commit, our savings estimator applies this logic to your group. In their first year, companies we work with often see an average of $451 in savings per employee, along with better engagement and retention. That is our reported average, and your own number depends on your baseline and participation.

Hadaco

Our population health programs deliver evidence-based support for chronic disease, preventive care and mental health, with quarterly reporting and support for both employers and brokers. For a spreadsheet version of the steps above, read our guide to estimating wellness savings. For how claims data feeds the model, see integrating claims data with wellness. You can also review how wellness programs reduce healthcare costs and what HR leaders need to know about corporate wellness ROI.

FAQ

How many months of claims do I need for a baseline?

Use at least 12 months, and 24 if possible. The SHRM and HERO/PHA guide recommends these periods, with fully adjudicated medical and pharmacy claims. Chapman Institute suggests 24 to 36 months where available, which gives a steadier benchmark.

What is the simplest formula for healthcare ROI?

ROI equals program benefits minus program costs, divided by program costs, times 100. For example, $264,000 in savings against $90,000 in cost gives about 193%. Pair it with the benefit-cost ratio, which here is about 2.9 to 1, so finance sees both views.

Should I count avoided claims as savings?

Count them only when your analysis supports that the program caused the difference. Keep realized savings, where actual PMPM beat the trended benchmark, apart from projected savings based on modeled risk reduction. Label each clearly in your report.

How often should we recalculate?

Track PMPM monthly and report quarterly. Finalize annually, once claims runout is complete. HealthPartners' guidelines, for example, use a 12-month period plus three months of paid claims runout.

Do HSA tax savings count in the same calculation?

Track them as a separate line. Payroll tax effects from HSA contributions through a Section 125 plan come from tax treatment, not from claims. Your tax advisor should confirm the figures.

Sources