Abstract
Early and accurate diagnosis of HIV remains a cornerstone of public health strategies. This study aimed to evaluate the predictive value of signal-to-cutoff (S/CO) ratios from two fourth-generation HIV screening assays (Abbott Architect and Alinity) and to analyze diagnostic trends across pre-pandemic, pandemic, and post-pandemic periods in a low-prevalence setting. We retrospectively analyzed 197,642 unique HIV screening tests conducted at Bursa Uludağ University Hospital from 2015 to 2024. Receiver operating characteristic (ROC) analysis was used to determine optimal S/CO thresholds for distinguishing true-positive results. Of the 197,642 samples screened, the overall HIV prevalence was 0.5%, with 196 cases (0.1%) confirmed as new diagnoses. The Architect assay showed an optimal S/CO threshold of ≥11.8 (sensitivity 98.3%, specificity 97.3%). The Alinity assay demonstrated 100% sensitivity and specificity at an S/CO threshold of ≥19.1. Although a temporary decline in test volume occurred in 2020, there was no statistically significant difference in confirmation rates across years. During the pandemic, newly diagnosed individuals were significantly older and had lower CD4 counts, indicating delayed diagnosis (p = 0.026 and 0.008, respectively). Men who have sex with men (MSM)-related transmission significantly increased post-pandemic (p = 0.032). S/CO ratio-guided interpretation enhances diagnostic accuracy and may reduce unnecessary confirmatory testing, especially in low-prevalence and resource-limited regions. Selecting the optimal threshold can help to ensure a timely diagnosis and optimize HIV screening algorithms.