SMaRT-PD: a clinical decision support system for the management of Parkinson’s disease

<h3>Background</h3> Home Based Care (HBC) is a supported self-management pathway for Parkinson’s disease (PD) using remote monitoring data to inform person-centred care. Scalability is limited by manual processes. Digital clinical decision support systems (CDSS) have the potential to improve care delivery efficiency. <h3>Aim</h3> To develop a CDSS with high accuracy alert and patient-facing recommendations generation. <h3>Method</h3> Rule-based decision trees (RBDTs) were generated by clinical experts with input from patients and carer partners, integrating routine care record, wearable and patient- reported data. Retrospective off-line test-retest cycles and prospective testing proportionate to anticipated use informed iterations and accuracy evaluation. <h3>Results</h3> 58 RBDTs were generated: size ranged from 1 to 103 leaf nodes, with a median of 4.5 decision nodes. 48 inform patient-facing care recommendations (33.3%, 47.9% and 89.6% result in a further information request, alert for clinical contact or self-management advice respectively). 687 Retrospective and 963 prospective tests were undertaken, identifying 90 actionable issues that informed RBDT iteration; accuracy improved with iterated RBDT versions. <h3>Discussion</h3> We have created a prototype CDSS, which integrates multi-modal data to support <b>S</b>elf-<b>Ma</b>nagement, <b>R</b>emote monitoring and <b>T</b>imely review (SMaRT-PD). In next steps, we will produce a digital form for technical and clinical validation studies.