RESUMEN
OBJECTIVE: To characterize the patient population using weekend and evening appointments for screening mammography versus standard appointment times across four outpatient facilities in our academic health system. METHODS: In this institutional review board-approved retrospective cohort study, there were 203,101 screening mammograms from 67,323 patients who had a screening mammogram performed at outpatient centers at a multisite academic institution from January 1, 2015, to December 31, 2022. Screening appointments were defined as "standard appointment time" (between 8 am and 5 pm on Monday through Friday) or "weekend or evening appointment time" (scheduled after 5 pm on Monday through Friday or at any time on a Saturday or Sunday). Associations between appointment group and patient characteristics were analyzed using univariate and multivariate logistic regression. RESULTS: Most screening mammograms (n = 185,436, 91.3%) were performed at standard times. The remainder (n = 17,665, 8.7%) were performed during weekends or evenings. As we created additional weekend and evening appointments after the coronavirus disease 2019 pandemic, the annual percentage of all screening mammograms performed on evenings and weekends increased. On multivariate analysis, when compared with standard appointment times, we found that patients who were younger than age 50 (P < .001), a race other than non-Hispanic White (P < .001), non-English speakers (P < .001), and from less advantaged zip codes (P < .03) were more likely to use weekend and evening appointment times compared with those aged 70 and above, non-Hispanic White patients, English speakers, and those from the most advantaged zip codes. CONCLUSIONS: Weekend and evening appointment availability for screening mammograms might improve screening access for all patients, particularly for those younger than age 50, those of races other than non-Hispanic White, and those from less advantaged zip codes.
RESUMEN
Background Digital breast tomosynthesis (DBT) has higher diagnostic accuracy than digital mammography, but interpretation time is substantially longer. Artificial intelligence (AI) could improve reading efficiency. Purpose To evaluate the use of AI to reduce workload by filtering out normal DBT screens. Materials and Methods The retrospective study included 13 306 DBT examinations from 9919 women performed between June 2013 and November 2018 from two health care networks. The cohort was split into training, validation, and test sets (3948, 1661, and 4310 women, respectively). A workflow was simulated in which the AI model classified cancer-free examinations that could be dismissed from the screening worklist and used the original radiologists' interpretations on the rest of the worklist examinations. The AI system was also evaluated with a reader study of five breast radiologists reading the DBT mammograms of 205 women. The area under the receiver operating characteristic curve (AUC), sensitivity, specificity, and recall rate were evaluated in both studies. Statistics were computed across 10 000 bootstrap samples to assess 95% CIs, noninferiority, and superiority tests. Results The model was tested on 4310 screened women (mean age, 60 years ± 11 [standard deviation]; 5182 DBT examinations). Compared with the radiologists' performance (417 of 459 detected cancers [90.8%], 477 recalls in 5182 examinations [9.2%]), the use of AI to automatically filter out cases would result in 39.6% less workload, noninferior sensitivity (413 of 459 detected cancers; 90.0%; P = .002), and 25% lower recall rate (358 recalls in 5182 examinations; 6.9%; P = .002). In the reader study, AUC was higher in the standalone AI compared with the mean reader (0.84 vs 0.81; P = .002). Conclusion The artificial intelligence model was able to identify normal digital breast tomosynthesis screening examinations, which decreased the number of examinations that required radiologist interpretation in a simulated clinical workflow. Published under a CC BY 4.0 license. Online supplemental material is available for this article. See also the editorial by Philpotts in this issue.
