Skip to content
Heyrafiki

Our Science-Backed, and Honest Approach on Care & Technology.

The Approach

What Science-First means here.

Four standing practices, and the standard everything we publish is held to.

Validated Instruments Only

Every Check on Heyrafiki is a published, validated Instrument: the same questions researchers published, the published cutoffs, and scoring anyone can check. From the WHO-5 to the PHQ-9, every Instrument links back to its source literature.

See every Instrument and its Citation

Safety Outranks Score

Risk protocols are built into the Screening flows themselves. A concerning answer routes to Crisis Support immediately, before any score is computed, because a number should never stand between a Person and help.

The Crisis Centre

Evidence, Cited

Health guidance we publish cites the research behind it, study by study. Where the evidence is thin or contested, the article says so instead of rounding up.

Read the Articles

Published Either Way

When we test something, the result gets published whether it flatters us or not. Methods, samples and limitations are named in every write-up, so our claims can be checked rather than taken on trust.

Three Tracks.

Every study runs with licensed Experts, and every result gets published, including the ones that prove us wrong.

For Experts

How Practitioners use the tools: Scribe accuracy reviews, workflow studies and note-quality audits.

For Clients

How People experience Care: matching satisfaction, assessment usefulness and accessibility studies.

For the Technology

How the Technology holds up: privacy protections, matching accuracy and safety-threshold testing.

Where We Are

Honest about Stage.

Heyrafiki has not yet been field-tested at population scale, and we will not write as if it has. What exists today: validated Instruments running in the product, safety protocols live in every screening flow, and the first studies with licensed Experts being scoped. As the evidence grows, it publishes here, with the methods attached.

Standing Commitments.

  • Every publication names its Method, its Sample and its Limitations
  • Study protocols are registered before results exist, so findings cannot be reshaped after the fact
  • Negative results get published too
  • Claims trace back to validated Instruments and reproducible Protocols