Aims Junior Medical Officers (JMOs) play an important role in postoperative care in ENT OPD. Accurate evaluation of grommet position, patency and presence of infection is necessary to ensure optimal patient outcomes and reduce complications. This study aims to evaluate inter-rater reliability of JMOs evaluating grommets. Method Otoscopic images of children with tympanostomy were collected from telehealth programs in NT and QLD from 2011-2019. Retrospective analysis was conducted of images featuring grommets, which were independently reviewed by 3 JMOs and 1 accredited registrar (reference standard). Grommet conditions were classified into 3 categories: in-situ and patent, insitu and blocked, and extruded. Agreement metrics, including % agreement and kappa statistics, were calculated to assess the accuracy and consistency of JMOs’ detection of grommet status in comparison to the reference standard. Results Among 300 grommets reviewed, 61% were identified by a senior registrar (reference standard) as in situ and patent, 26% in situ and blocked, and 13% extruded. Overall percent agreement between JMO and registrar assessment for grommets in situ was 82% (95% CI: 78 - 85%), with kappa value of 0.77, indicating substantial agreement. For grommets identified as in situ and patent, agreement rose to 89% (95% CI: 85 - 92%) with kappa of 0.88. However, the detection of blocked grommets showed lower agreement of 66% (95% CI: 58 - 73%) and kappa of 0.53, indicating moderate agreement. Conclusions JMOs demonstrate high proficiency in detecting patent grommets but show lower accuracy in identifying blocked grommets, potentially leading to missed opportunities for postoperative aural toileting, topical therapy, or specialist review. Integrating artificial intelligence-driven analysis into digital otoscopy could offer JMOs real-time feedback on grommet status, support learning opportunities, enhance patient education, and improve competency in outpatient care settings.