Background Self-sampling has been shown to be a non-invasive and cost-effective method for the diagnosis and screening of sexually transmitted infections (STI). This study aims to evaluate the accuracy of detection of HPV and other STI on self-collected vaginal samples as compared to clinician-collected cervical samples in women with a recent diagnosis of cervical dysplasia. Methods Self-collected vaginal (VS) and physician-administered cervical samples (CS) were collected from 130 women attending the Colposcopy Clinic, San Gerardo Hospital, Monza, Italy with a diagnosis of cervical dysplasia. VS and CS were collected using FLOQSwabs and L-Shaped FLOQSwab (Copan) respectively and transported to the Microbiology Laboratory of the University of Milano-Bicocca. Samples' nucleic acid extraction was performed using NucliSENS ® easyMAG (bio-Mérieux). HPV and STIs detection was evaluated using Anyplex II HPV28 and STI-7 (Seegene), respectively. Sample cellularity adequacy through human CCR5 gene assessment was performed using an 'in house' Real-time PCR assay. Results Demonstrated a very good overall concordance for HPV and STI detection on self and clinician-collected samples (gold standard). Very good agreement for the detection of one or more HPV types was demonstrated (Kappa = 0.915) with HPV positivity rates of 75% and 72% for VS and CS respectively. Similarly very good agreement was demonstrated for the detection of one or more of the 7 STIs understudy (Kappa = 0.899). Overall a higher positivity for STIs was found in VS (48%) compared to CS (43%), with Ureaplasma parvum being most frequently detected. Adequate sample cellularity was demonstrated for all samples types; mean values of 2.07E+06 and 3.16E+06 cells/sample for VS and CS respectively. Conclusion Self-collected samples showed a high degree of concordance with CS for both HPV and STIs detection with comparable sample adequacy. These results are promising for the introduction of self-collected samples in sexually transmitted and cervical cancer screening programs. Disclosure No significant relationships.