Evaluation of Referral Accuracy for Suspected Retinal Tears and Detachments: A Retrospective Cohort Study

Abstract

This is a retrospective study that looked at referral rates of retinal tears and detachments seen in rapid access eye care. Data was collected over a period from 2022 to 2024 and assessed the referral patterns and accuracy of diagnoses to hospital eye services. The primary aim was to evaluate how accurate referral sources were in correctly diagnosing retinal tears and detachments. The sources of referrals included WECS/Optometrists, self-referrals, hospitals, and primary care. Methods: Data was collected from clinical records and entered into a database using a pre-prepared proforma. This took place between January 2022 and April 2024. A total of 1089 patients were identified, and accuracy was calculated by confirming or denying the referring differential diagnosis to the official diagnosis. Data was excluded if there was no identifying referral source. Referral sources included WECS/Optometrists, self-referrals, medical teams/wards, and others such as GPs and A & E. Statistical analysis was performed using one-way ANOVA and Fisher’s exact test. Results: Optometrists constituted the majority of referrals (64.3%), with a sensitivity of 40.3% and a specificity of 81.2% for detecting retinal breaks/tears or detachments. Self-referrals demonstrated the highest accuracy at 71.4%. Statistical analysis revealed no significant differences in referral accuracy among the sources (p-value > 0.05). Retinal breaks/tears were the most common diagnosis (36.7%), followed by retinal detachments (26.6%). Conclusions: While WECS/Optometrists were the predominant referral source, their sensitivity was moderate, indicating potential for improvement. The lack of statistically significant differences in referral accuracy suggests that majority of referral sources could benefit from enhanced training and clearer referral guidelines. The study also highlights the need for better documentation and larger sample sizes to refine referral pathways and improve patient outcomes. Limitations: The study is limited by the retrospective nature of the data and the substantial proportion of referrals with unspecified sources (14.6%). Additionally, small sample sizes for certain referral groups reduce the ability to compare fairly and extrapolate the data to wider practice. Future research should focus on prospective data collection and the integration of advanced diagnostic tools to enhance referral accuracy.