The 140,615 patients analyzed overall had a median follow up of 57.7 months, and 47.5% of patients had ≥ 60 months of follow-up. Mean (SD) patient age was 58.6 (11.9) years, ranging 18–90. Postoperatively, 123,947 (88.1%) remained pathologically DCIS, whereas 16,668 (11.9%) were found to have invasive disease (Fig. 1).
Five-year OS was 95.8% (95% CI 95.7–96.0%). Following adjustment, greater time to surgery (TTS), with multiple other factors, was independently associated with poorer survival (eTable 2). A total of 39,364 (28.0%) patients had > 1 surgery date recorded. A sensitivity analysis (eTable 3) excluding these women from the adjusted analysis found surgical delay significantly associated with OS (HR 1.12; 95% CI 1.09–1.16; P < 0.0001).
There were 123,947 patients analyzed after excluding those with invasive disease. Mean (SD) age for these patients was 58.7 (11.9) years (range 18–90); 7422 (6.0%) patients died, and adjusted 5-year OS was 96.0% (Fig. 2). Adjusted associations between analyzed variables and OS in patients with noninvasive disease are shown in eTable 4.
Among the 16,668 patients found postoperatively to have invasion, mean (SD) age was 57.3 (12.4) years (range 19–90). Within this cohort, 2730 (16.4%) patients died and the adjusted 5-year OS was 94.9% (Fig. 2). eTable 5 demonstrates the adjusted associations between analyzed variables and OS in patients with invasion.
Delay to Surgery
Unadjusted median delay from diagnosis to surgery was 38 (IQR: 24–58) days. Patients with intervals of ≤ 30, 31–60, 61–90, 91–120, 121–180, 181–240, and 241–365 days accounted for 37.6%, 38.9%, 14.6%, 5.2%, 2.8%, 0.6%, and 0.3% of the total number, respectively. Patient characteristics are summarized by preoperative interval in eTable 6. After adjustment, characteristics independently associated with increased TTS included Black and Asian race, Hispanic ethnicity, lack of high school diploma, metropolitan setting, greater treating facility distance, transfers of care, and increased Charlson comorbidity score, among others (Table 1). Added risk of death from all causes for each 30-day interval delay increase was 7.4% (HR 1.07; 95% CI 1.05–1.10; P < 0.0001) for the entire cohort. Survival point estimates by delay group are listed in eFig. 1 and eTable 7.
Among those without invasion, individuals with intervals of ≤ 30, 31–60, 61–90, 91–120, 121–180, 181–240, and 241–365 days accounted for 38.7%, 38.6%, 14.1%, 5.0%, 2.6%, 0.6%, and 0.3% of the cohort, respectively. The added risk of death from all causes for each 30-day interval increase in delay among the noninvasive cohort was 7.3% (HR 1.07; 95% CI 1.05–1.10; P < 0.0001) (eTable 4).
Among the invasive patients, the progressive delay categories accounted for 28.8%, 41.0%, 18.5%, 6.8%, 3.8%, 0.7%, and 0.4% of the cohort, respectively. The added risk of death from all causes for each 30-day interval increase in delay among the invasive cohort was 6.8% (HR 1.07; 95% CI 1.01–1.13; P = 0.0306; eTable 5).
As shown in Fig. 3 and Table 2, invasive cancer was increasingly found with greater delay, as well as associated with insurance status, facility distance and volume, transfers of care, year of diagnosis, comorbidities, grade, and ER status. After adjustment, increasing delay to surgery in the entire cohort was an independent predictor of invasion (OR 1.13; 95% CI 1.10–1.15; P < 0.001). Among patients with postoperative diagnosis of invasive disease, median invasion measured 5.0 mm (range 0–480). Additional independent predictors of invasion are elaborated in Table 2. A sensitivity analysis in which women having > 1 surgery date were excluded from the adjusted analysis (eTable 8) found that increasing delay to surgery remained an independent predictor of invasion (OR 1.12 per month delay; 95% CI 1.10–1.15; P < 0.001). There was no significant difference in the survival effect from delay between women with or without invasion on pathology (P = 0.507).