An expedited recall pathway cut time to biopsy for highly suspicious mammograms

Design

Retrospective single-centre comparison of recall pathways

Population

4,400 screening mammography recalls; 57 expedited

Primary outcome

Time from screening to diagnostic imaging, and malignancy yield

Effect

Median 7 vs 25 days to diagnostic imaging; 24 vs 45 days to biopsy; PPV1 91.2% vs 5.2%

At one US academic centre, screening radiologists could flag a highly suspicious BI-RADS 0 finding in the electronic record, and the patient was telephoned and booked into a reserved diagnostic appointment, aiming for work-up within a week. The study reviewed recalls from November 2024 to January 2026.

Of 36,384 screens, 4,647 were recalled and 57 (1.2%) were flagged. Flagged patients had diagnostic imaging at a median of 7 days against 25, and biopsy at 24 days against 45. Delays of 30 days or more were far less common (7% vs 38% for imaging; 2% vs 15% for biopsy). Malignancy yield was very high — 91% of flagged recalls proved to be cancer, against 5% of standard ones. Mass findings and larger size predicted flagging.

The pathway is cheap: a flag, a phone call and a few held slots. It works because screening radiologists can already identify the small group of recalls that are almost certainly cancer. The data are retrospective and from one centre, and the study measures time, not cancer outcomes, but a three-week gain to diagnosis for these women is worth having where diagnostic capacity is stretched.

• Consider a flag on screening reports for recalls you judge highly suspicious for malignancy.
• Reserve a small number of diagnostic slots each week for flagged recalls.
• Contact flagged patients by telephone rather than letter to book them quickly.
• Keep the flag for findings you would expect to be BI-RADS 4 or 5 — over-use will erode the benefit.
• Audit time to diagnosis and positive predictive value for flagged recalls.

Why it matters

A simple scheduling change shortened time to diagnosis for the patients most likely to have cancer.

Don't overread it

A single-centre retrospective workflow study — it measured time to diagnosis, not stage or survival.

The statistics, in plain English

A PPV1 of 91% means more than nine in ten flagged recalls were cancer. The comparison with standard recalls (5%) is not like-for-like, because flagging was meant to select the most suspicious cases. The time savings are medians, so half of patients waited longer than the stated figure.

Read the rest in the app

You have read your two free briefings this month. The app carries all 27 specialties, every morning, free — and this finding is waiting in it.

Get Daily Doctor — free

Scan to keep reading on your phone. No account needed to start.