Best Ways to Prove a Health Program Works to Funders

Funders who wrote checks for the last decade of digital and community health work are asking a sharper question than they used to. They no longer want to know how many phones were distributed or how many workers were trained. They want to know whether people were healthier because the money was spent, and they want that claim backed by something more durable than a photo gallery and a headcount. For program leads, the practical challenge is to prove health program impact to funders in a way that survives scrutiny from evaluators, satisfies grant reporting requirements, and positions the work for renewal. This report walks through what a credible evidence package looks like, which methods carry the most weight, and where most programs quietly lose the room.

"GiveWell's cost-effectiveness analyses are the single most important input into its charity recommendations," the organization states in its published criteria. In 2023 it directed roughly $197 million to recommended programs, with 52 percent flowing to just four top charities that met its bar for rigorous, repeatedly studied evidence.

What it takes to prove health program impact to funders

The instinct of many program leads is to lead with activity counts. Activity data answers the wrong question. Funders distinguish between outputs (screenings delivered, workers trained) and outcomes (blood pressure detected earlier, antenatal visits completed, deaths averted). To prove health program impact to funders, you have to connect what you did to what changed, and then show that the change would not have happened anyway.

That connection rests on three things a strong evidence package always contains. First, a clear theory of change that maps activities to intermediate outcomes to final impact. Second, a measurement plan with a credible comparison, because a before-and-after number without a counterfactual invites the obvious question of what else changed. Third, honest cost data, since funders increasingly compare programs on impact per dollar rather than impact alone.

The theory of change is doing more work than most teams realize. Research by the CDC, which updated its Program Evaluation Framework in 2024 to emphasize collaborative engagement and continuous learning, treats the logic model as the backbone that determines what you measure in the first place. A vague theory produces a vague indicator set, and a vague indicator set produces a report a funder cannot act on.

Comparing the main evidence approaches

Not every program can or should run a randomized controlled trial. The right method depends on program stage, budget, and how skeptical the funder is. The table below compares the approaches program leads most often weigh when demonstrating program effectiveness.

Evidence approach	Strength of causal claim	Typical cost and effort	Best suited to	Common funder objection
Activity and output reporting	Very low	Low	Compliance and monitoring	"This tells us nothing about outcomes"
Pre/post outcome tracking	Low to moderate	Low to moderate	Early-stage programs	"How do you know it was your program?"
Quasi-experimental (matched comparison, difference-in-differences)	Moderate to high	Moderate	Scaled field programs	"Are the comparison groups really comparable?"
Randomized controlled trial	High	High	Flagship interventions seeking replication	"Does the result hold outside the trial setting?"
Cost-effectiveness analysis	Depends on underlying data	Moderate	Any program competing for scarce funds	"Your assumptions drive the whole number"

A few practical patterns follow from this:

• Match the method to the claim you actually need to make. A district screening program does not need a trial to show it shortened referral times; a well-constructed comparison group is often enough.
• Layer methods over the program lifecycle. Start with routine outcome tracking, then add a quasi-experimental design as the program scales.
• Never present a cost-effectiveness figure without the assumptions behind it. Funders who use these numbers, and many do, will interrogate the inputs before they trust the output.
• Pre-register your primary outcomes where possible. Choosing the metric after seeing the data is the fastest way to lose credibility.

Industry Applications

Community health worker deployments

Community health worker programs generate a firehose of field data, and the discipline is turning it into outcome evidence rather than activity logs. When a worker records a contactless vital sign reading, the data point matters to a funder only if it connects to a downstream action: a referral made, a follow-up completed, a condition caught earlier than it otherwise would have been. Programs that instrument this pathway can report the proportion of flagged cases that reached care, which is far more persuasive than the raw number of scans performed.

Grant reporting and donor relations

Grant reporting on health outcomes is where evidence either compounds or evaporates. A common failure is treating each report as a standalone deliverable rather than an installment in a longitudinal story. Funders who see the same indicators tracked consistently across reporting periods gain confidence that the measurement system is real. Evaluation specialists note that a defined evaluation plan, with named metrics and an analysis method stated up front, strengthens grant proposals before a single result is in.

Public health institutions and ministries

For district and national health offices, the audience is often a multilateral funder with its own results framework. The Global Fund's 2023 Results Report, which attributed 59 million lives saved to its investments as of the end of 2022, shows how large funders aggregate program-level data into headline impact claims. Programs that report in indicators compatible with these frameworks make themselves far easier to fund and to renew.

Current research and evidence

The methodological center of gravity among serious funders has moved toward rigor plus transparency. GiveWell's published process describes a funnel: shallow reviews of many programs, then intensive review of the most promising, prioritizing interventions that have been "rigorously and repeatedly studied" through randomized controlled trials and Cochrane Library meta-analyses. Its 2023 cost-effectiveness updates included granular adjustments such as medical costs averted by life-saving interventions and revised probability-of-death estimates for vaccine-preventable diseases. The lesson for program leads is not that every program must clear GiveWell's bar, but that the direction of travel among influential funders is toward defensible, auditable numbers.

Academic evaluation frameworks point the same way. The 2024 CDC framework foregrounds equity and stakeholder engagement alongside methodological quality, and work published in BMJ Public Health on building a shared theory of change across global entities argues that a jointly developed logic model makes capacity-strengthening efforts measurable in the first place. Meanwhile the US Department of Health and Human Services 2024 Evaluation Plan continues to distinguish process evaluations from outcome evaluations, a distinction program leads should make explicit in every funder report so that a monitoring metric is never mistaken for an impact claim.

What the evidence base does not yet resolve is external validity: a result proven in one district under trial conditions may not hold when a program scales across regions with different health systems. Funders know this, which is why replication and consistent longitudinal reporting increasingly matter more than a single striking result.

The future of proving program impact

Three shifts are reshaping how programs will demonstrate effectiveness to donors over the next several years. First, real-time outcome data is replacing the annual retrospective report. As field tools capture structured records at the point of care, funders will expect dashboards that update continuously rather than PDFs that arrive once a year. Second, cost-effectiveness is becoming a default screen rather than a specialist exercise, pushing programs to track spending against outcomes from day one. Third, the bar for a credible counterfactual is rising, so designs that can support a defensible comparison group will win funding against programs that offer only before-and-after numbers.

The programs that thrive will be the ones that build measurement into deployment instead of bolting evaluation on at the end. Evidence for donors is not a document you write in the final quarter of a grant; it is a system you stand up on day one and feed continuously.

Frequently asked questions

What evidence do funders actually require to prove a health program works? Most funders want to see outcomes rather than activities, a theory of change linking your work to those outcomes, a credible comparison that rules out other explanations, and cost data. The exact rigor depends on the funder, but the direction is consistently toward auditable, outcome-level evidence over headcounts.

Do we need a randomized controlled trial to satisfy funders? Rarely for routine grants. Trials are reserved for flagship interventions seeking to prove replicable impact. Most field programs can make a strong case with quasi-experimental designs, consistent longitudinal outcome tracking, and transparent cost-effectiveness figures.

How is a good grant report on health outcomes structured? Distinguish process metrics from outcome metrics, report the same indicators consistently across periods, state your measurement method up front, and present cost alongside impact. Treat each report as one installment in a longitudinal story rather than a standalone deliverable.

What is the most common mistake when demonstrating program effectiveness? Leading with activity counts and choosing outcome metrics after the data is in. Both signal to evaluators that the measurement system was designed to look good rather than to answer the funder's real question.

Building an evidence package that survives funder scrutiny is a design problem, not a reporting afterthought, and it is exactly the space Circadify is working in through field deployment research and outcomes reporting collaborations. Program leads, grant-making bodies, and institutions exploring an outcomes-reporting partnership can read the underlying research and field results at circadify.com/blog.