JMIR Publications

ABSTRACT

Background:

Digital self-management tools for chronic eye disease are evaluated on device or algorithm performance rather than on whether people already living with functional vision loss can reach and keep using them. Whether accessibility translates into sustained self-management, and whether clinical benefit follows, is largely untested. The role of nursing mediation in that translation is unmapped.

Objective:

We aimed to map where evidence sits along the pathway from disease severity to clinical outcome for patient-facing digital tools in eye care, and to test whether the apparent absence of nursing-mediation evidence is real or an artefact of database coverage.

Methods:

Following PRISMA-ScR, we searched PubMed, Web of Science, Scopus and CINAHL Ultimate for English-language records published 2015–2026. Eligibility required eye disease or functional vision impairment, a patient-facing digital tool, and a user-side outcome. Studies were coded on causal-chain position, tool class, vision status, outcome type and evidence grade.

Results:

Of 7,536 records, 97 studies were included. Evidence concentrated in one link: 85 studies (87.6%) addressed interface accessibility → self-management behaviour, against 9 and 3 for the adjacent links; 76 stopped at an accessibility or usability endpoint and only 6 reached a clinical outcome. Only 21 (22%) measured any causal edge. The nursing-mediation layer is a genuine absence within our criteria — no study treated nursing attribution as an examined variable. A severity gradient runs in the same direction but is not significant (Fisher P=.12).

Conclusions:

Research has documented whether users can operate these tools, leaving untested whether operation changes outcomes or whether nursing mediation makes it possible. We term this the accessibility paradox and specify four measurable breakpoints in the assess–adapt–teach–follow-up sequence, so that the zero finding is refutable by design.