03·Evidence

The proof that a clinical or operational lead needs in order to recommend Sylva internally.

Published research, the framework behind it, the economics, and an honest account of what we do and don't claim.

Published & peer-reviewed

Our single strongest proof point.

Nature

Brown et al. (2026), Nature Scientific Reports. Open access.

4.8/5

Mean rating from an independent evaluation by NHS clinical psychologists on clinical appropriateness, accuracy and safety.

100%

Safeguarding detection accuracy in independent evaluation.

How we built it

A four-layer clinical framework

Built with the same rigour you would expect from a clinical intervention. The details are here if you want it.

Layer 1 Grounded in NICE guidance Started with all NICE relevant guidance: NICE CG128, CG142, CG170 and NG87.

Layer 2 Reviewed the literature 49 literature reviews referencing ~460 systematic reviews and meta-analyses across 1,738 primary sources.

Layer 3 Mapped what makes it work Mapped the behaviour change techniques behind effective support, using the BCT Ontology (Marques et al., 2024).

Layer 4 Targeted the mediators Focused on the psychological mediators that drive real outcomes (Hayes et al., 2022).

The economics

The argument that answers "we already run a programme".

£240 per family / year versus £952–£5,000 for comparable programmes. 4-20× more families for the same budget.

4–18% attend group programmes 30–68% decline. Sylva removes every attendance barrier: no transport, childcare, fixed schedule or group setting challenges, making the commitment easier for families to stick with.

£109m cumulative saving / 5 yrs versus scaling a group programme to 20,000 families. Net cost-saving from year two, even against a waitlist.

Cost-effectiveness. The ICER remains below £25,000/QALY under every tested scenario. Probability of cost-effectiveness exceeds 88% at the NICE threshold. Sylva's cost would need to rise by 474% before it ceased to be cost-effective.

Source: SM-SYLVA-HE-001 health economics technical report. Pre-clinical model, CHEERS 2022 compliant, NICE PMG36 reference case.

Outcomes we measure

Named instruments. Real measurement.

Measured at baseline, 4, 8 and 12 weeks. 134 completers targeted. Service evaluation in progress. RCT is planned.

ECBIEyberg Child Behaviour Inventory: primary outcome for child behaviour change.

SDQStrengths & Difficulties Questionnaire: co-primary for child wellbeing.

PSS-10Perceived Stress Scale: parental stress.

PSOCParenting Sense of Competence: confidence and efficacy.

EQ-5D-5LHealth economics: the NICE-preferred instrument.

CompACT-10Psychological flexibility.

⅔

Two thirds of parents reported reduced stress within six weeks. PSS-10 data from the service evaluation will replace this early figure when available. Our post-market clinical follow-up programme is in progress, and we'll update this page as results come in.

What we do not claim

We do not claim Sylva produces equivalent outcomes to face-to-face therapy. We do not claim technology-delivered techniques are identical to facilitator-delivered ones.

What we do claim is that Sylva was designed with the same rigour you would expect from a clinical intervention.

We'll send the full evidence pack ahead of the call.

Published paper, health economics report and the clinical rigour we've used to develop Sylva: everything your governance team needs to review before they say yes.