Table 3.
Cumulative average radon and breast cancer risk stratified by region of residence in NHSII
| Invasive breast cancer | ER+/PR+ | ER−/PR- | |||||||
|---|---|---|---|---|---|---|---|---|---|
| Stratification variablea | Cases/p-years | HR (95% CI)b | p int. | Cases/p-years | HR (95% CI)b | p int. | Cases/p-years | HR (95% CI)b | p int. |
| Region of residence | 0.0002 | 0.13 | 0.45 | ||||||
| Northeast | |||||||||
| Radon tertile 1 | 198/110,392 | Referent | 94/110,506 | Referent | 22/110,559 | Referent | |||
| Radon tertile 2 | 665/376,841 | 1.05 (0.90, 1.23) | 356/377,134 | 1.19 (0.95, 1.50) | 72/377,377 | 0.96 (0.60, 1.56) | |||
| Radon tertile 3 | 494/341,830 | 0.83 (0.70, 0.99) | 262/342,049 | 0.96 (0.76, 1.23) | 69/342,242 | 0.91 (0.56, 1.49) | |||
| Midwest | |||||||||
| Radon tertile 1 | 158/105,523 | Referent | 87/105,584 | Referent | 14/105,661 | Referent | |||
| Radon tertile 2 | 441/311,230 | 0.96 (0.80, 1.15) | 241/311,433 | 0.95 (0.74, 1.21) | 73/311,597 | 1.77 (1.00, 3.14) | |||
| Radon tertile 3 | 616/391,334 | 1.09 (0.91, 1.29) | 301/391,619 | 0.96 (0.76, 1.22) | 96/391,793 | 1.86 (1.06, 3.26) | |||
| West | |||||||||
| Radon tertile 1 | 504/308,674 | Referent | 279/308,883 | Referent | 62/309,064 | Referent | |||
| Radon tertile 2 | 94/48,620 | 1.19 (0.95, 1.49) | 59/48,653 | 1.41 (1.06, 1.89) | 8/48,713 | 0.72 (0.34, 1.53) | |||
| Radon tertile 3 | 53/20,194 | 1.47 (1.10, 1.97) | 23/20,223 | 1.22 (0.80, 1.88) | 7/20,239 | 1.52 (0.69, 3.37) | |||
| South | |||||||||
| Radon tertile 1 | 464/300,872 | Referent | 224/301,096 | Referent | 50/301,245 | Referent | |||
| Radon tertile 2 | 158/102,885 | 0.95 (0.79, 1.14) | 76/102,968 | 0.95 (0.73, 1.23) | 23/103,014 | 1.27 (0.77, 2.10) | |||
| Radon tertile 3 | 110/77,355 | 0.91 (0.74, 1.13) | 64/77,389 | 1.11 (0.84, 1.47) | 17/77,433 | 1.26 (0.72, 2.19) | |||
Abbreviations: p int. p-value for interaction, p-years person-years
a Radon tertile 1: <33.3 Bq/m3; tertile 2: ≥33.3–57.4 Bq/m3; tertile 3: ≥57.4 Bq/m3
b Models are not adjusted for the stratifying variable. Models are adjusted for age, race, Census tract median home value, Census tract median income, marital status, living arrangements, individual-level income, family history of breast cancer, screening mammography, personal history of biopsy-confirmed BBD, age at menarche, parity, age at first birth, lactation, menopausal status and hormone use (among postmenopausal women only), height, BMI at age 18, change in BMI since age 18, smoking status, physical activity, adult alcohol consumption, PM2.5, population density