What we will and will not do.
These principles run every inquiry that comes in. They are the structural protection that assure Summit Epi Analytics' work is independent. They are public so that prospective clients can self-select — and so that we can be held to them.
Last revised June 2026. Substantive revisions are tracked in a visible changelog.
The six principles
- 01
Independence is the product.
What clients pay for is the willingness to tell them what they don't want to hear. We protect that structurally — through what we publish, what we co-author, what we refuse, and how we are paid.
- 02
Accompaniment over delivery.
Modeling, evaluation, and trial work are co-developed with the people closest to the problem. Co-authorship and capacity transfer are part of the deliverable, not a nice-to-have. Where we can, client-affiliated statisticians are named on every brief.
- 03
Rigor is respect.
Sloppy causal reasoning, ignored interference, hidden assumptions, and overstated certainty are failures of respect toward the populations the work is meant to serve. They are treated as such here.
- 04
Methodology is open.
Our evaluation frameworks, modeling code, and analysis templates are published openly under permissive licenses. Only the client-specific application is confidential. Open methodology is what makes independence verifiable.
- 05
Findings are publishable.
Every engagement reserves the right to publish findings, with a reasonable embargo negotiated up front (typically 6–12 months). No contract requires permanent suppression of methodology or findings.
- 06
Loyalty is to the populations the work touches.
When a tool's claims, a trial's design, or a program's evaluation would mislead the populations it is deployed against, we say so in the headline — not the appendix. This is the line that distinguishes the practice from a marketing arm.
Work we will not take on.
A practice is defined as much by what it refuses as by what it accepts. The list below is binding, not aspirational.
- Tobacco-, firearm-, or extractive-industry-funded health work intended primarily to delay regulation.
- Modeling for state actors with documented histories of weaponizing health data against minority populations, dissidents, or migrants.
- "Validation" engagements where the deliverable is presupposed.
- Contracts that require permanent suppression of methodology or findings.
- Subcontracts under primes whose analysis plan we have not been allowed to review first.
- More than two active engagements plus one in scoping at any time, while the practice runs at part-time capacity.
For conflict-of-interest disclosure on biosecurity evaluation engagements, a disclosure policy is provided before any engagement begins.
If these principles fit your work, we should talk.
Tell us what you're trying to decide, who else is at the table, and the deadline you're working against.
Request a scoping call