Underuse of Radiation in Younger Women With Breast Cancer
Eighteen thousand one hundred twenty (86.25%) of the 21 008 patients received RT within a year of BCS. Table 1 shows that the majority of the patients (89.86%) had no comorbidity. More than half of the patients (59.51%) were the primary holders of the insurance policy. Among the study cohort, 4.25% had at least one child aged less than 7 years, and the rate of RT was 80.61% for this group. Geographic variations in RT compliance were also observed, ranging from less than 80% in the Pacific Census division to close to 90% in the West-North-Central Census division.
Results from the logistic regression (Table 1) showed that compared with patients aged 50 years of age or younger, the odds ratio (OR) of receiving radiation therapy was 1.22 (95% confidence interval (CI) = 1.08 to 1.37), 1.38 (95% CI = 1.22 to 1.55), and 1.28 (95% CI = 1.11 to 1.47) for those in the age ranges 51 to 55 years, 56 to 60 years, and 61 to 64 years, respectively. Having at least one child aged less than 7 years resulted in statistically significantly lower odds of radiation therapy than having no children or older children. Among young patients (aged 20–64 years), compared with those with at least one young child (aged <7 years), patients with children aged 7 to 12 years, patients with children aged 13 to 17 years, and patient with no children or children aged 18 years or older were more likely to receive RT (children 7–12 years: OR = 1.32, 95% CI = 1.05 to 1.66, P = .02; children 13–17 years: OR = 1.41, 95% CI = 1.13 to 1.75, P = .002; no children or children ≥18 years: OR = 1.38, 95% CI = 1.13 to 1.68, P = .001). Perturbation analyses (Supplementary Table 2, available online) showed that although women's age group and the family structure variable were correlated, the correlation was not high enough to cause incorrect inferences. Statistically significantly lower odds of RT were observed among patients who were enrolled in a HMO or a capitation PPO (OR = 0.70; 95% CI = 0.63 to 0.77), patients who travelled across Census division to receive BCS (OR = 0.72; 95% CI = 0.60 to 0.86), or patients who received BCS in an outpatient setting (OR = 0.73; 95% CI = 0.57 to 0.94). In addition, patients who were the primary holder of insurance policy were more likely to have RT (OR = 1.20; 95% CI = 1.10 to 1.31). Those living in areas with lower percentages of the population with college education were statistically significantly less likely to receive RT. In addition, DCIS was found to be statistically significantly negatively associated with the receipt of RT. Patients with chemotherapy, imaging, and mammogram screening were more likely to receive RT after BCS.
Analyses stratified by age group (Table 2) showed that the association between young children and lower likelihood of RT was statistically significant only in the youngest age group of women (aged 20–50 years). This suggested that the association observed in the full sample was primarily driven by women in the youngest age group. Other associations observed above largely persisted across all age groups. In addition, sensitivity analyses (Figure 2) for women in the youngest age group showed that our hypothesized relationship was supported regardless of patients' DCIS status or whether they were primary holders of an insurance policy.
(Enlarge Image)
Figure 2.
Adjusted odds ratios in sensitivity analyses for family structure variables, by subgroups. Error bars represent 95% confidence intervals. All statistical tests were two-sided. DCIS = ductal carcinoma in situ.
Results
Eighteen thousand one hundred twenty (86.25%) of the 21 008 patients received RT within a year of BCS. Table 1 shows that the majority of the patients (89.86%) had no comorbidity. More than half of the patients (59.51%) were the primary holders of the insurance policy. Among the study cohort, 4.25% had at least one child aged less than 7 years, and the rate of RT was 80.61% for this group. Geographic variations in RT compliance were also observed, ranging from less than 80% in the Pacific Census division to close to 90% in the West-North-Central Census division.
Results from the logistic regression (Table 1) showed that compared with patients aged 50 years of age or younger, the odds ratio (OR) of receiving radiation therapy was 1.22 (95% confidence interval (CI) = 1.08 to 1.37), 1.38 (95% CI = 1.22 to 1.55), and 1.28 (95% CI = 1.11 to 1.47) for those in the age ranges 51 to 55 years, 56 to 60 years, and 61 to 64 years, respectively. Having at least one child aged less than 7 years resulted in statistically significantly lower odds of radiation therapy than having no children or older children. Among young patients (aged 20–64 years), compared with those with at least one young child (aged <7 years), patients with children aged 7 to 12 years, patients with children aged 13 to 17 years, and patient with no children or children aged 18 years or older were more likely to receive RT (children 7–12 years: OR = 1.32, 95% CI = 1.05 to 1.66, P = .02; children 13–17 years: OR = 1.41, 95% CI = 1.13 to 1.75, P = .002; no children or children ≥18 years: OR = 1.38, 95% CI = 1.13 to 1.68, P = .001). Perturbation analyses (Supplementary Table 2, available online) showed that although women's age group and the family structure variable were correlated, the correlation was not high enough to cause incorrect inferences. Statistically significantly lower odds of RT were observed among patients who were enrolled in a HMO or a capitation PPO (OR = 0.70; 95% CI = 0.63 to 0.77), patients who travelled across Census division to receive BCS (OR = 0.72; 95% CI = 0.60 to 0.86), or patients who received BCS in an outpatient setting (OR = 0.73; 95% CI = 0.57 to 0.94). In addition, patients who were the primary holder of insurance policy were more likely to have RT (OR = 1.20; 95% CI = 1.10 to 1.31). Those living in areas with lower percentages of the population with college education were statistically significantly less likely to receive RT. In addition, DCIS was found to be statistically significantly negatively associated with the receipt of RT. Patients with chemotherapy, imaging, and mammogram screening were more likely to receive RT after BCS.
Analyses stratified by age group (Table 2) showed that the association between young children and lower likelihood of RT was statistically significant only in the youngest age group of women (aged 20–50 years). This suggested that the association observed in the full sample was primarily driven by women in the youngest age group. Other associations observed above largely persisted across all age groups. In addition, sensitivity analyses (Figure 2) for women in the youngest age group showed that our hypothesized relationship was supported regardless of patients' DCIS status or whether they were primary holders of an insurance policy.
(Enlarge Image)
Figure 2.
Adjusted odds ratios in sensitivity analyses for family structure variables, by subgroups. Error bars represent 95% confidence intervals. All statistical tests were two-sided. DCIS = ductal carcinoma in situ.