(SLC-037) Implementing and Validating an Algorithm for Identifying ATTR-CM patients in the PHARMO Data Network

Background: Transthyretin amyloid cardiomyopathy (ATTR-CM) is a rare, progressive, and often underdiagnosed condition affecting the heart. It is characterised by restrictive cardiomyopathy and can lead to progressive heart failure. Currently, diagnostic codes specific to ATTR-CM do not exist, posing challenges to the identification of this debilitating disease in real-world data research and in its management in routine clinical care.

Objectives: To implement a literature-based algorithm for identifying ATTR-CM patients in Dutch real-world data from the PHARMO Data Network.

Methods: The PHARMO Data Network links anonymous, patient-level electronic healthcare data from primary and secondary healthcare settings in the Netherlands. Hospital and out-patient pharmacy records from 2013 to 2022 were used to identify ATTR-CM patients based on a combination of diagnostic, procedure, and medication codes indicative of amyloidosis and cardiac involvement. Patients with codes indicative of light chain amyloidosis, secondary amyloidosis, or amyloid angiopathy were excluded. This algorithm is based on a previously published algorithm by Damy et al. on a French healthcare database and adapted for use with Dutch healthcare data.

Results: Of 3,054,433 patients eligible for linked hospital and out-patient pharmacy data between 2013 and 2022, there were 2,365 patients (0.08%) with an amyloidosis diagnosis. Of these, 871 amyloidosis patients (37%) had a relevant cardiovascular condition, including heart failure, arrhythmia, cardiomyopathy, or conduction disorders. Among amyloidosis patients with a relevant cardiovascular condition, 504 patients (58%) were subsequently excluded because they had at least one diagnostic, procedure, or medication code indicative of light chain amyloidosis, secondary amyloidosis, or amyloid angiopathy. This resulted in a final sample of 367 patients classified as ATTR-CM cases, indicating a prevalence of 12 per 100,000 patients.

Conclusions: The estimated prevalence of ATTR-CM identified herein is in line with that of the original literature-based algorithm, but higher than reports from other countries. Further case validation of the identified cohort by independent clinicians is needed and will rely on general practitioner, in-patient pharmacy, laboratory, and pathology data. A clinically rich database like the PHARMO Data Network can provide a comprehensive platform to identify, validate, and manage high-risk rare disease patients including ATTR-CM.