This model was responsible for correctly identifying 73% of patients with pulmonary arterial hypertension 6 months prior to a confirmed diagnosis.
A new claims-based, machine-learning algorithm model has been shown to be effective in identifying patients with pulmonary arterial hypertension (PAH). This model was responsible for correctly identifying 73% of patients with PAH 6 months prior to a confirmed diagnosis.
The data on this classification model were presented at the American Thoracic Society 2020 International Conference in San Francisco.
Though early diagnosis and treatment initiation for PAH have always been the objective, PAH symptoms are non-specific, which has resulted in many patients experiencing delays (an average of 2 or more years) in symptom onset and a confirmed diagnosis.
However, investigators led by Katherine Bettencourt, PhD, of Actelion Pharmaceuticals, developed a classification model that could identify PAH prior to diagnosis based on retrospective healthcare claims data.
The machine-learning model analyzed data from the US-based Optum Clinformatics database from January 2015 to December 2019.
Patients from this database with shared early symptomology of PAH including dyspnea, fatigue, and chest pain were divided into 2 cohorts, those with PAG and non-PAH controls with cardiovascular or respiratory disease.
Propensity score matching was controlled for differences such as age, sex, and race, and the matched control cohort was randomly down sampled to create a control to PAH patient ratio of 3:1.
A total of 1724 patients were in the PAH cohort while the control group has 5352 patients. Mean age was similar (69 and 70 years old), and roughly two-thirds of patients were female and White.
Bettencourt and colleagues observed that the model was able to distinguish between patients with PAH and controls at 6 months prior to diagnosis, with a threshold of 0.43.
Additionally, the model was capable of accurately identifying 73% of the patients with PAH, and the precision of the model was 50%.
Regarding the implications of this device, investigators believed the high rate of imaging procedures prior to a confirmed diagnosis of PAH indicated a need to screen sooner. The team also noted that current hospitalization data has shown that shifting diagnosis 6 months sooner could provide an opportunity for earlier treatment and reduce the risk of hospitalization prior to diagnosis.
“The performance of our model indicates the feasibility of identifying patients at a population level who might benefit from PAH-specific screening by using existing data that are routinely collected in a claims database,” the team wrote.