Skip to content
Recovrly

Evidence

What we know, and what we're still finding out

Recovrly is early-stage. This page is the honest version of our evidence base: published research, what's actively being evaluated, the governance that keeps personalisation inside safe bounds, and a register mapping every specific claim on this site back to its source.

A regulated, governed foundation

Recovrly is a UK-registered Class I medical device (Recovrly Health Ltd), built to support, not replace, the judgement of your care team. Before a single outcome claim, this is the part that's already rigorous:

  • Formal risk management

    Developed following ISO 14971 risk management principles and IEC 62304 software lifecycle principles, with a maintained risk management file.

  • Governed content, not ad hoc pages

    Every clinical content item moves through author → draft → clinical review → approval before publishing, with a named owner, reviewer and review-due date.

  • Personalisation stays inside clinical bounds

    _kim, our supporting personalisation technology, only surfaces or prioritises content that already exists within a patient's clinician-approved pathway. It doesn't generate new clinical content, diagnose, triage, or make clinical decisions.

Published research

Primary evidence (about Recovrly)

The Hospital to Home study (H2H): smartwatch technology-enabled rehabilitation following hip fracture in older adults, a feasibility non-randomised trial

Hewage K, Fosker S, Leckie T, Venn R, Gonçalves AC, Koulouglioti C, Hodgson LE. Future Healthcare Journal, 2023;10(1):14–20. View on PubMed →

A UK feasibility study in patients recovering from hip fracture (aged 65+, n=66, median age 78). One hospital gave patients a smartwatch and the CUSH Health© app, Recovrly's technology under its founding name, while a second hospital continued with usual care as a comparison.

  • • Hospital stay: 10 days (IQR 7–16) in the app hospital vs. 12 days (IQR 10–18) in the comparison hospital (p=0.05).
  • • Falls-related readmissions: none in the app hospital vs. 15 in the comparison hospital, including 11 fractures (p=0.016).
  • • Median daily step count in the app group rose from 477 (in-hospital) to 5,352 at 12 weeks (p=0.001).

Read this with

This was a feasibility study, not a randomised controlled trial: the two groups were different hospitals, not randomly assigned patients, so other differences between the sites could contribute to the result. The sample was small (n=66 across both hospitals), and 12 people in the app group withdrew, which the authors flag as a digital-exclusion risk still to be addressed. The paper discloses that the founder of CUSH Health had no role in analysing the results.

Design-context research

The key values and factors identified by older adults to promote physical activity and reduce sedentary behaviour using co-production approaches: a scoping review

Ioannou E, Chen HL, Bromley V, Fosker S, Ali K, Fernando A, Mensah E, Fowler-Davis S. BMC Geriatrics, 2023;23:371. View on PubMed →

A scoping review of 16 studies asking what older adults themselves identify as the reasons they stay physically active. This is not a study of Recovrly. Recovrly's founder co-authored it, and it informs the design principles behind Recovrly's approach rather than evidencing Recovrly's own effectiveness.

The consistent themes: activity needs to feel safe, accessible and enjoyable, preserve independence, keep people connected to others, and show tailored, observable progress.

Currently being evaluated

Colorectal perioperative vignette study

Phase 1: development & pilot underway

A blinded study design comparing recovery plans generated by _kim against plans independently written by experienced colorectal surgical registrars, scored by consultant colorectal surgeons who don't know which plan is which.

The question it's designed to answer: are _kim-generated recovery plans clinically acceptable, compared with plans written by experienced clinicians? Right now we're in the development and pilot stage: building the case set with a panel of consultants and registrars and testing the scoring approach, ahead of a larger, definitive evaluation. No results exist yet; this page will be updated when they do.

Claims register

Every specific number or outcome claim on this site, mapped to its source. This is the only place recovrly.com currently states figures; nothing above appears anywhere else on the site without also appearing here.

ClaimSource
Shorter hospital stay in the app hospital (10 vs. 12 days)Hewage et al. 2023, Future Healthcare Journal
Fewer falls-related readmissions in the app hospital (0 vs. 15)Hewage et al. 2023, Future Healthcare Journal
Median daily steps rose to 5,352 at 12 weeks in the app groupHewage et al. 2023, Future Healthcare Journal

Last reviewed: 8 September 2026. This page is updated as new research, pilots or evaluations complete.

Curious how this translates into care? For Patients · For Clinicians