Table 1.
Participant baseline characteristics in the NHS development and validation cohorts and in BWHS
| NHS development cohort | NHS validation cohorta | BWHS (Crude) | BWHS (Age-standardized)a | |
|---|---|---|---|---|
| N | 55,553 | 27,777 | 17,380 | 17,380 |
| Factors in our final model | ||||
| Age, mean (SD) | 70.1 (7.0) | 70.2 (7.0) | 63.7 (6.7) | – |
| 55–59 years, % | 7.1 | 6.7 | 36.5 | – |
| 60–64 years, % | 21.4 | 21.5 | 28.3 | – |
| 65–69 years, % | 22.4 | 22.3 | 17.1 | – |
| 70–74 years, % | 20.9 | 20.8 | 9.9 | – |
| 75–79 years, % | 18.1 | 18.8 | 5.8 | – |
| 80+ years, % | 10.1 | 10.0 | 2.4 | – |
| Highest self-reported Body Mass Index (BMI) in past 10 years kg/m2, mean (SD) | 28.1 (5.7) | 28.1 (5.8) | 32.1 (7.0) | 31.6 (6.5) |
| Highest self-reported BMI in past 10 years | ||||
| < 20 kg/m2, % | 2.7 | 2.7 | 0.5 | 0.5 |
| 20–22.4 kg/m2, % | 10.9 | 11.2 | 3.0 | 3.0 |
| 22.5–24.9 kg/m2, % | 19.1 | 19.4 | 8.5 | 9.2 |
| 25–29.9 kg/m2, % | 36.5 | 36.3 | 31.6 | 33.7 |
| 30–34.9 kg/m2, % | 19.5 | 18.9 | 28.3 | 28.0 |
| 35–39.9 kg/m2, % | 7.1 | 7.3 | 14.5 | 13.5 |
| 40+ kg/m2, % | 4.1 | 4.1 | 12.6 | 10.7 |
| Unknown, % | 0.1 | 0.2 | 1.0 | 1.3 |
| Average alcohol use per day (highest average use in past 10 years)b | ||||
| None, % | 37.0 | 37.0 | 50.4 | 54.1 |
| 1–4.9 g/day, % | 22.5 | 22.7 | 26.0 | 23.9 |
| 5–14.9 g/day, % | 17.5 | 17.1 | 13.7 | 12.6 |
| 15+ g/day, % | 13.1 | 13.4 | 9.9 | 9.3 |
| Unknown, % | 10.0 | 9.8 | – | – |
| Cigarette use | ||||
| Never | 44.7 | 44.8 | 53.4 | 50.5 |
| Current | 7.7 | 7.9 | 10.1 | 8.0 |
| Past | 47.4 | 47.1 | 36.4 | 41.5 |
| Unknown | 0.2 | 0.2 | – | – |
| Limited from walking several blocksc | ||||
| Not at all | 61.8 | 61.5 | – | – |
| A little or a lot | 32.8 | 33.1 | – | – |
| Unknown, % | 5.4 | 5.4 | – | – |
| Limited in bathing or dressing oneselfc | ||||
| Not at all | 88.2 | 87.9 | – | – |
| A little or a lot | 6.5 | 6.8 | – | – |
| Unknown, % | 5.3 | 5.3 | – | – |
| Usual walking pace outdoors | ||||
| Unable to walkc | 2.6 | 2.8 | – | – |
| Slow or average (less than 3 mph) | 73.1 | 72.6 | 62.0 | 65.1 |
| Brisk/very brisk (≥ 3 mph) | 19.7 | 20.0 | 27.5 | 23.3 |
| Unknown, % | 4.6 | 4.6 | 10.6 | 11.6 |
| High Blood pressured | 60.3 | 60.3 | 68.7 | 75.4 |
| Depression | 20.4 | 20.5 | 22.2 | 19.4 |
| Hip fracture | 2.1 | 2.2 | 0.8 | 1.1 |
| Parkinson’s disease | 0.7 | 0.6 | 0.2 | 0.3 |
| Myocardial infarction | 5.5 | 5.7 | 4.3 | 6.5 |
| Congestive heart failure | 3.3 | 3.4 | 3.2 | 4.6 |
| Stroke/transient ischemic attack | 7.5 | 7.5 | 4.2 | 5.7 |
| Emphysema/Asthma | 18.6 | 18.5 | 18.3 | 18.0 |
| Diabetes | 12.2 | 11.7 | 24.2 | 28.5 |
| Dementia | 1.2 | 1.2 | 0.2 | 0.5 |
| Kidney disease | 0.5 | 0.6 | 1.9 | 2.3 |
| Cancer (excluding non-melanomatous skin cancers) | 12.3 | 12.2 | 5.7 | 7.4 |
| Age at menopause (years) | ||||
| < 45, % | 10.8 | 10.4 | 19.3 | 21.2 |
| 45–49, % | 23.4 | 23.9 | 20.1 | 19.2 |
| 50–54, % | 56.1 | 56.2 | 27.4 | 25.9 |
| 55+, % | 8.8 | 8.5 | 8.3 | 10.6 |
| Hysterectomy, age unknown, %e | – | – | 22.2 | 21.9 |
| Unknown, % | 1.0 | 1.0 | 2.7 | 1.2 |
| Mammogram in past 2 yearsf | ||||
| No | 12.1 | 12.2 | 14.0 | 15.2 |
| Yes | 79.4 | 79.5 | 86.0 | 84.8 |
| Unknown | 8.5 | 8.3 | – | – |
| Number of breast biopsies | ||||
| 0, % | 73.2 | 73.2 | 67.0 | 65.7 |
| 1, % | 23.5 | 23.6 | 21.3 | 21.3 |
| 2+, % | 3.3 | 3.2 | 11.7 | 13.1 |
| Postmenopausal hormone use | ||||
| Never, % | 22.5 | 22.9 | 40.7 | 37.5 |
| Current estrogen plus progestin user < 5 years, % | 0.4 | 0.5 | 0.7 | 0.2 |
| Current estrogen plus progestin user 5 + years, % | 2.7 | 2.7 | 1.1 | 0.8 |
| Current estrogen-alone user < 5 years, % | 0.9 | 0.8 | 1.6 | 0.7 |
| Current estrogen-alone user 5 + years, % | 9.4 | 9.1 | 6.3 | 5.6 |
| Past estrogen plus progestin user < 5 years, % | 14.9 | 14.9 | 12.8 | 12.5 |
| Past estrogen plus progestin user 5 + years, % | 15.8 | 16.1 | 5.6 | 7.4 |
| Past estrogen-alone user < 5 years, % | 7.1 | 7.2 | 12.2 | 12.2 |
| Past estrogen-alone user 5 + years, % | 13.8 | 13.8 | 13.8 | 18.1 |
| Unknown, % | 12.6 | 12.2 | 5.3 | 4.8 |
| Age at first birth (years) and parityg | ||||
| Nulliparous | 5.3 | 5.4 | 16.0 | 12.7 |
| < 25, 1–2 children | 14.2 | 13.9 | 32.2 | 35.4 |
| < 25, 3+ children | 35.2 | 34.9 | 27.8 | 28.3 |
| 25–29, 1–2 children | 14.8 | 15.3 | 11.9 | 10.8 |
| 25–29, 3+ children | 20.1 | 20.3 | 2.9 | 4.0 |
| 30+, 1–2 children | 5.8 | 5.8 | 7.1 | 6.3 |
| 30+, 3+ children | 2.8 | 2.8 | 0.5 | 0.6 |
| Unknown | 1.8 | 1.7 | 1.6 | 1.9 |
| Number of first-degree relatives with history of breast cancer and age at diagnosish | ||||
| None, % | 82.1 | 81.9 | 80.1 | 79.8 |
| 1 and age < 50, % | 4.3 | 4.1 | 3.8 | 4.2 |
| 1 and age 50+, % | 11.6 | 11.8 | 14.2 | 14.0 |
| 2+ and at least one age < 50, % | 1.0 | 1.1 | 0.6 | 0.7 |
| 2+ and age 50+, % | 1.1 | 1.1 | 1.3 | 1.3 |
| Outcomes | ||||
| Breast cancer in 5-year follow-up, % | 1.7 | 1.7 | 1.7 | 1.8 |
| Breast cancer in 10-year follow-up, % | 3.1 | 3.2 | 3.1 | 3.2 |
| Breast cancer death in 5-year follow-up, % | 0.1 | 0.1 | 0.2 | 0.2 |
| Breast cancer death in 10-year follow-up, % | 0.3 | 0.2 | 0.4 | 0.6 |
| Non-breast cancer death 5-year follow-up, % | 7.7 | 8.0 | 4.5 | 8.1 |
| Non-breast cancer death 10-year follow-up, % | 20.1 | 20.6 | 11.1 | 19.7 |
| Not included in model | ||||
| Race/ethnicity | ||||
| Non-Hispanic White, % | 96.2 | 96.3 | 0.0 | 0.0 |
| Non-Hispanic Black, % | 1.8 | 1.6 | 99.1 | 99.9 |
| Hispanic,% | 0.9 | 0.9 | 0.9 | 0.1 |
| Asian, Pacific Islander % | 0.8 | 0.9 | 0.0 | 0.0 |
| Native American, % | 0.2 | 0.2 | 0.0 | 0.0 |
| Education (years)i | ||||
| < 12 | – | – | 2.7 | 4.0 |
| 12 | – | – | 16.4 | 19.1 |
| 13–15 | – | – | 27.8 | 26.2 |
| 16 | 91.7 | 91.7 | 19.5 | 17.1 |
| 17+ | 8.4 | 8.4 | 33.6 | 33.5 |
| Unknown | 0.0 | 0.0 | 0.1 | 0.1 |
| Predicted mammographic densityj | n = 35,073 | n = 17,482 | ||
| Higher than median (≥ 24.7% dense) | 50.0 | 50.1 | – | – |
| Lower than median (< 24.7% dense) | 50.0 | 49.9 | – | – |
NHS (Nurses’ Health Study) included participants that completed the 2004 questionnaire. BWHS (Black Women’s Health Study) included participants that completed the 2009 questionnaire
aThere were no significant differences between the NHS development and validation cohorts, except for limited bathing/dressing (p = 0.04) and diabetes (p = 0.04). All comparisons between NHS development cohort and the BWHS age-standardized cohort were statistically significant except for emphysema/asthma, 5- and 10-year breast cancer incidence, and 5- and 10-year non-breast cancer death incidence. To calculate the age-standardized values for BWHS, we weighted each of six 5-year age groups, where the weight corresponded to the proportion of that age group in the whole NHS sample, and then summed the weighted age-specific values to get the overall adjusted prevalence
bA standard drink is any drink that contains about 14 g of pure alcohol (12 oz. of beer, 5 oz. of wine or 1.5 oz. of liquor)
cBWHS does not ask about participants’ about walking several blocks, limitations in bathing or dressing oneself, or inability to walk
dHealth conditions were self-reported
eParticipants with hysterectomy and age at menopause unknown in BWHS were placed in the 45–49 years category for all analyses; sensitivity analyses were performed showing no significant difference when placed in any other age category
fParticipants missing data on mammography use in NHS had completed a short version of the 2004 questionnaire
gParticipants who were parous with an unknown number of children were categorized as having 1–2 children
hRelatives with an unknown age at diagnosis were categorized as having been diagnosed at 50 + years of age
iParticipants in NHS who had a master’s or doctorate degree were placed in the 17 + years category. All others were placed in the 16 years category, as registered nurses, (22.8% of these specifically reported a Bachelor’s degree)
jPredicted mammographic density was defined using a 9-item validated index. Prevalence is reported for those with complete data for the prediction model [18]