A Study Evaluating the Accuracy of Triage for Breast Referrals During the Covid-19 Pandemic in a Tertiary Hospital.

Objective: To evaluate the accuracy of breast referral triage during Covid-19.

Design: Retrospective case study.

Setting: Primrose Breast Unit, Derriford Hospital, Plymouth. From March 17th to June 30th to encompass the height of the pandemic and the early enforced changes to practice.

Participants: All referrals received, triaged, and seen ( = 870) in the unit, identified by referral records.

Main Outcome Measures: The primary outcome measure of a positive disease state was of a histological diagnosis of cancer, with the absence of a cancer diagnosis representing a negative disease state. Accuracy has been determined by sensitivity and specificity calculations; thus defined by correctly triaging cancers to face-to-face clinics and benign cases to telephone or video clinics.

Results: Sixty-eight cancers (7.8% of referrals) were detected after initial triage and consultation, of which 51 (sensitivity = 75%) were triaged to one-stop-clinic; positive predictive value was 18.89%. Eight hundred two (specificity = 72.69%) of benign cases were triaged to phone or video clinic initially; negative predictive value was 97.15%. Comparing the study's incidence of cancer (7.8%) to the preceding year's (2019) of 6.8% with Yate's correction shows no significant difference ( < 0.05).

Conclusion: Triage accuracy is sufficiently robust to diagnose cancer promptly, which should reassure clinicians and decision makers within the cancer networks.