Study type

Study type

Non-interventional study

Scope of the study

Disease epidemiology
Other

If ‘other’, further details on the scope of the study

Using digital biomarkers within EMR (OPCRD) to detect potentially undiagnosed patients with rare or difficult-to-diagnose disease in population-scale data
Study drug and medical condition

Medical condition to be studied

Fabry's disease
Glycogen storage disease type II
Population studied

Age groups

Children (2 to < 12 years)
Adolescents (12 to < 18 years)
Adults (18 to < 46 years)
Adults (46 to < 65 years)
Adults (65 to < 75 years)
Adults (75 to < 85 years)
Adults (85 years and over)

Estimated number of subjects

19993084
Study design details

Main study objective

The key objectives: 1. Ensuring patients get the right diagnosis faster 2. Increasing awareness of rare diseases among healthcare professionals 3. Better coordination of care

Outcomes

The primary outcome of this QI study is to use digital biomarkers for the purpose of identifying patients at risk of rare and difficult to diagnose diseases and help clinicians to diagnose disease in population-scale data. (1) To provide educational material within the QI reports that targets the importance of diagnosing rare disease and includes specific content about individual cases. (2) To highlight patients at risk of the disease to GPs for consideration of further evaluation and investigation, therefore, helping patients to access the specific and coordinated care and treatment that they need early on.

Data analysis plan

The QI programme runs the machine learning algorithm through EMR data to identify patients flagged as at risk of either Fabry disease or Pompe disease against 500 features within the EMR. The key analysis metrics: • Number of patients flagged at risk of Fabry or Pompe disease at baseline • Number of patients with a diagnosis of Fabry or Pompe disease at baseline • Number of patients flagged at risk of Fabry or Pompe disease at quarterly intervals • Number of patients with a diagnosis of Fabry or Pompe disease at quarterly intervals • Number of practices that opt in • Number of referrals made for a diagnosis The analysis runs the algorithm monthly on the EMR data to learn lessons from the initial cycle of QI and barriers are addressed by the study team and the programme adapted going forward. A learning healthcare system is created that combines cyclical QI reports for GPs for external validation and to generate knowledge that can be used to achieve change in practice.