Timeliness of an Expedited Diagnostic Pathway for Screening Mammography Recalls.

OBJECTIVE

To compare diagnostic and biopsy timeliness, malignancy yield, and cancer characteristics between expedited and standard screening mammography recall pathways.

METHODS

An expedited recall workflow implemented in Epic enabled radiologists to flag highly suspicious BI-RADS 0 findings for telephone scheduling into reserved appointments, targeting diagnostic evaluation within one week. We retrospectively reviewed recalls from 11/1/2024-1/31/2026. Primary outcomes were time from screening to diagnostic evaluation and malignancy yield (PPV1 and PPV3). Secondary outcomes included diagnostic-to-biopsy and screening-to-biopsy intervals, delayed diagnostic imaging (≥30 days from screening), delayed biopsy (≥30 days from diagnostic), and malignancy characteristics. Intervals were measured in calendar days.

RESULTS

Among 36,384 screening mammograms, 4,647 were BI-RADS 0 (recall rate 12.8%); 57 (1.2%) were expedited. After excluding 247 standard pathway cases without follow-up, 4,400 were analyzed. Expedited patients were older (median 64.6 vs 52.5 years, p<0.001) with shorter median screening-to-diagnostic (7 vs 25 days), diagnostic-to-biopsy (14 vs 19 days), and screening-to-biopsy intervals (24 vs 45 days; all p<0.001). Delayed diagnostic imaging (7.0% vs 38.2%, p<0.001) and biopsy (1.8% vs 14.5%, p=0.004) were less frequent. Expedited cases had higher BI-RADS 4/5 rates (96.5% vs 30.1%, p<0.001) and malignancy yield (PPV1 91.2% vs 5.2%; PPV3 94.5% vs 18.2%; p<0.001). Multivariable logistic regression among 278 cancers identified mass findings (OR 3.36, p<0.001) and imaging size (OR 1.03/mm, p=0.003) as predictors of expedited routing.

DISCUSSION: Radiologists identified highly suspicious screening mammography findings for expedited evaluation. Expedited routing was associated with shorter diagnostic and biopsy intervals and high malignancy yield, supporting diagnostic resource prioritization.