JMIR Publications

ABSTRACT

Background:

Rare diseases (RDs) affect over 300 million people globally, with only 5% having approved therapeutic solutions, and are predominantly chronic and complex conditions that necessitate integrated health system responses. Lysosomal storage disorders (LSDs) illustrate the diagnostic and management challenges typical of RDs. Advances in Digital Health Technologies (DHTs), especially Artificial Intelligence (AI) and Connected Care (CC) applications, are increasingly reported as tools to support LSDs management.

Objective:

This scoping review provides the first comprehensive mapping and descriptive analysis of evidence and applications of DHTs, specifically AI and CC, for the management of LSDs. It seeks to clarify scope boundaries between AI, CC, and other DHTs; chart the distribution of evidence across populations, care-journey phases, and outcome domains; and identify gaps, methodological limitations, and priorities for future research and policy.

Methods:

Using a PICO-based framework, we conducted a PRISMA-ScR–conformant scoping review of Medline/PubMed, Google Scholar, ClinicalTrials.gov, and grey literature (last decade). Searches were iteratively extended with >80 DHT-specific keywords and AI-assisted tools. Of 1633 records screened, 245 were included. Evidence was charted by LSD population, intervention class (AI, CC, other DHTs), outcome domains (patient, healthcare-delivery, societal), and care-journey phase; peer-reviewed and grey sources were summarised separately, and no formal effectiveness or risk-of-bias assessment was undertaken.

Results:

Among the 245 records, more than 90% were peer-reviewed and 7.8% grey literature; none were completed randomized controlled trials or LSD-specific systematic reviews. 46 records involved AI-driven DHTs, 93 CC technologies, and 163 other DHTs (with overlaps). Evidence was mostly concentrated in Gaucher disease and Fabry disease, with sparse signals for other LSDs. Care-journey mapping showed a front-loaded digital footprint, with almost half of mapped literature focused on screening and diagnosis, and fewer in treatment intensification, rehabilitation, and end-of-life care. Outcome mapping revealed a predominance of healthcare delivery performance measures (eg, diagnostic accuracy, workflow coordination), with relatively few patient-reported or societal outcomes. AI applications were mainly used for diagnostic support, phenotyping, and risk prediction and CC interventions focused on telemedicine, remote patient monitoring and care coordination. Across all classes, evidence came mostly from small, single-centre observational studies.

Conclusions:

The mapped evidence base is appreciable in size for a niche field, reflecting growing interest in digital health applications for LSD care. However, the quality and heterogeneity of available studies remain insufficient to infer effectiveness or recommend routine implementation. The evidence mapping across populations, intervention types, outcome domains and care-journey phases, highlights areas with relatively stronger evidence as well as clear gaps. These insights can help inform future, higher-quality evidence generation, expert consensus-building needed to guide clinical practice as evidence and real-world applications rapidly evolve, and contribute to the development of timely policies and governance frameworks—consistent with ongoing EU and WHO priorities for rare diseases.