Abstract
BackgroundEstimation of incidence in rare diseases is often challenging due to unspecific and incomplete coding and recording systems. Patient- and health care provider-driven data collections are held with different organizations behind firewalls to protect the privacy of patients. They tend to be fragmented, incomplete and their aggregation leads to further inaccuracies, as the duplicated records cannot easily be identified. We here report about a novel approach to evaluate the incidences of Duchenne muscular dystrophy (DMD) and spinal muscular atrophy (SMA) in Germany.MethodsWe performed a retrospective epidemiological study collecting data from patients with dystrophinopathies (DMD and Becker muscular dystrophy) and SMA born between 1995 and 2018. We invited all neuromuscular centers, genetic institutes and the patient registries for DMD and SMA in Germany to participate in the data collection. A novel web-based application for data entry was developed converting patient identifying information into a hash code. Duplicate entries were reliably allocated to the distinct patient.ResultsWe collected 5409 data entries in our web-based database representing 1955 distinct patients with dystrophinopathies and 1287 patients with SMA. 55.0% of distinct patients were found in one of the 3 data sources only, while 32.0% were found in 2, and 13.0% in all 3 data sources. The highest number of SMA patients was reported by genetic testing laboratories, while for DMD the highest number was reported by the clinical specialist centers. After the removal of duplicate records, the highest yearly incidence for DMD was calculated as 2.57:10,000 in 2001 and the highest incidence for SMA as 1.36:10,000 in 2014.ConclusionWith our novel approach (compliant with data protection regulations), we were able to identify unique patient records and estimate the incidence of DMD and SMA in Germany combining and de-duplicating data from patient registries, genetic institutes, and clinical care centers. Although we combined three different data sources, an unknown number of patients might not have been reported by any of these sources. Therefore, our results reflect the minimal incidence of these diseases.
Highlights
Estimation of incidence in rare diseases is often challenging due to unspecific and incomplete coding and recording systems
Movement of patients between regions and health care providers, and data capture through different organizations and for different purposes lead to a duplication of records for the same individual, which cannot be corrected for if the personal information of the individual is protected in accordance with data protection regulation
Highlighting the subgroup of spinal muscular atrophy (SMA) patients entered by a genetic institute and a neuromuscular center, 94.6% of patients diagnosed within the first 6 months of life were clinically classified as SMA type 1 by the neuromuscular center
Summary
Estimation of incidence in rare diseases is often challenging due to unspecific and incomplete coding and recording systems. Patient- and health care provider-driven data collections are held with different organizations behind firewalls to protect the privacy of patients They tend to be fragmented, incomplete and their aggregation leads to further inaccuracies, as the duplicated records cannot be identified. The International Rare Disease Research Consortium (IRDiRC) has convened a task force that developed principles and concepts for privacy protecting record linkage (PPRL) for rare diseases [2], similar to what has been applied in the cancer field (EUPID) [3] These systems rely in part on hash codes that allow de-duplication of records while not revealing personal identifiable information (PII) of the individual. They have not been utilized for rare neuromuscular disease on a larger scale
Sign-in/Register to view highlights & summary Sign-in/Register to access full text options 
Free) 
Talk to us
Join us for a 30 min session where you can share your feedback and ask us any queries you have
Schedule a call
