Polycystic ovary syndrome (PCOS) is a heterogeneous disorder with reproductive, metabolic, and psychological features, often underdiagnosed due to diagnostic inaccuracies and inconsistent knowledge among providers. These gaps highlight the need for improved diagnostic approaches to identify patients at risk earlier. This pilot study aimed to evaluate the validity of the PCOS risk algorithm (PriskA), a digital tool designed to assess PCOS risk in symptomatic women. A total of 144 women were referred for standardized endocrine screening at the Erasmus Medical centre and were included in the study, after excluding six women with inconclusive diagnoses. Of the 95 women with PCOS, 91 (96%) received a high PriskA score. Among the 49 without PCOS, 35% received a low score, 53% a high score, and 12% a moderate score. ROC analysis (high versus low scores) showed a sensitivity of 0.97 and specificity of 0.40, which improved to 0.45 when Elecsys AMH ≥ 3.2 ng/mL replaced ultrasound for diagnosing polycystic ovarian morphology (PCOM). Refining the algorithm increased specificity to 0.77 (AMH for PCOM) and 0.67 (ultrasound for PCOM). In this pilot study, PriskA proved to be a promising tool for assessing PCOS risk, particularly in primary care settings. It showed high sensitivity and minimal missed cases, making it especially useful where timely diagnosis is critical yet often challenging. By enabling earlier referrals, PriskA has the potential to reduce diagnostic delays, improve PCOS management, and help prevent associated comorbidities.
Ham et al. (Fri,) studied this question.