
Race, ethnicity, Medicare insurance, and advanced age were linked to increased risk of canceling screening mammography during the COVID-19 pandemic, according to research published July 29 in Clinical Imaging.
Researchers led by Dr. Nita Amornsiripanitch from Harvard Medical School found that nonwhite individuals canceled mammograms more often than whites before and during the pandemic.
The team looked at three time periods: March through June 2020, when state-mandated shutdowns caused canceled appointments; June through August 2020, when screening mammography resumed; and June through August 2019, before the pandemic.
The group found that the overall cancellation rate after reopening was 46% during the shutdown, compared with 37% before the pandemic. The relative risk of cancellation after reopening also increased with age (1.20 vs. 1.27 vs. 1.36 for ages 53, 61, and 70 years, respectively, p < 0.001).
The risk was also higher among Medicare patients (1.41) than Medicaid patients and those with other providers (1.26 and 1.21, respectively, p < 0.001), as well as nonwhite compared with white people (1.34 vs 1.25, p = 0.03).
Risk of failing to reschedule missed mammograms was also higher in hospitals compared with outpatient settings both during shutdown and after reopening (0.62 vs. 0.54, p = 0.005 and 1.29 vs 1.03, p < 0.001, respectively).
The researchers said that "rigorous" rescheduling efforts could help decrease missed screening appointments for mammograms.


![A normal mammogram confirmed by three-year radiologic follow-up illustrates reader-marked regions of interest (ROIs) during (A) unaided (round 1) and (B) artificial intelligence (AI)–assisted (round 2) reading. Each colored dot represents an ROI for recall by a human reader. Readers could mark more than one ROI per case, represented by multiple dots of the same color. During AI-assisted reading, the AI system displayed three visible prompts: two with suspicion of malignancy scores of 35% (left mediolateral oblique [L MLO] and craniocaudal [L CC]) and one with a suspicion of malignancy score of 10% (right craniocaudal [R CC]), shown as polygonal overlays. Without AI, six of 10 readers (60%) marked a false-positive ROI. With AI assistance, this fell to two of 10 (20%). R MLO = right mediolateral oblique.](https://img.auntminnie.com/mindful/smg/workspaces/default/uploads/2026/07/2026-07-14-radiology-mammogram-ai-auto-bias.H0bYO8QlWs.jpg?auto=format%2Ccompress&fit=crop&h=100&q=70&w=100)







![A normal mammogram confirmed by three-year radiologic follow-up illustrates reader-marked regions of interest (ROIs) during (A) unaided (round 1) and (B) artificial intelligence (AI)–assisted (round 2) reading. Each colored dot represents an ROI for recall by a human reader. Readers could mark more than one ROI per case, represented by multiple dots of the same color. During AI-assisted reading, the AI system displayed three visible prompts: two with suspicion of malignancy scores of 35% (left mediolateral oblique [L MLO] and craniocaudal [L CC]) and one with a suspicion of malignancy score of 10% (right craniocaudal [R CC]), shown as polygonal overlays. Without AI, six of 10 readers (60%) marked a false-positive ROI. With AI assistance, this fell to two of 10 (20%). R MLO = right mediolateral oblique.](https://img.auntminnie.com/mindful/smg/workspaces/default/uploads/2026/07/2026-07-14-radiology-mammogram-ai-auto-bias.H0bYO8QlWs.jpg?auto=format%2Ccompress&fit=crop&h=112&q=70&w=112)







