Introduction 1 2 5 6 We examined the association between lifetime breast-feeding and the incidence of type 2 diabetes mellitus in a large population-based cohort study of middle-aged women, the Shanghai Women’s Health Study (SWHS). Methods Study population 7 n Outcome ascertainment Breast-feeding duration assessment Measurement of potential confounders 8 9 10 11 12 Statistical analysis p Results n p 1 p Table 1 Age-standardised characteristics of parous women in the Shanghai Women’s Health Study by duration of breast-feeding   Duration of breast-feeding (months) 0 >0 to 6 >6 to 11 >11 to 35 ≥36 p a n 11,395 10,463 14,958 18,144 7,135  n 225 163 253 585 335  n 53,038 48,770 69,741 83,870 32,535  <10 years since last pregnancy (%) 26.9 21.9 28.9 18.6 3.7  Age, median (Q25–Q75) 45 (42–49) 45 (42–49) 45 (42–49) 54 (49–61) 64 (62–67) <0.001 Energy intake, mean kJ/day (SE) 6,875 (14) 6,927 (15) 6,920 (13) 6,985 (11) 6,738 (21) <0.001 Live births, median (Q25–Q75) 1 (1–1) 1 (1–1) 1 (1–1) 2 (1–2) 4 (3–5) <0.001 2 23.4 (0.03) 23.2 (0.03) 23.6 (0.03) 24.2 (0.02) 24.7 (0.04) <0.001 WHR, mean (SE) 0.80 (0.0005) 0.8 (0.0005) 0.80 (0.0005) 0.81 (0.004) 0.82 (0.007) <0.001 Weight gain, mean kg/year (SE) 0.34 (0.003) 0.31 (0.003) 0.33 (0.003) 0.36 (0.003) 0.37 (0.005) <0.001 Weight at age 20 years, median kg (Q25–Q75) 48 (45–53) 49 (45–53) 50 (45–54) 50 (45–54) 50 (45–55) <0.001 Smoker (%) 1.7 (1.5–1.9) 1.4 (1.2–1.9) 1.5 (1.3–1.7) 2.4 (2.3–2.5) 3.6 (3.2–4.0) <0.001 Ever drinker (%) 1.9 (1.7–2.1) 2.1 (1.9–2.3) 2.1 (1.9–2.3) 2.4 (2.3–2.5) 1.9 (1.7–2.1) <0.001 Exercise (%) 30.5 (29.5–31.5) 32.9 (31.9–33.9) 34.3 (33.3–35.3) 33.7 (32.7–34.7) 31.4 (30.4–32.4) 0.18 b 22.8 (22.0–23.6) 22.1 (21.3–22.9) 24.4 (21.6–25.2) 27.3 (26.9–27.7) 41.2 (40.2–42.2) <0.001 Education (%) None 11.0 (12.4–11.6) 8.5 (7.9–9.1) 9.6 (9.0–10.2) 17.8 (17.4–18.2) 41.2 (40.2–42.2) <0.001 Elementary 38.2 (37.2–39.2) 34.4 (33.4–35.4) 39.2 (38.2–40.2) 44.3 (43.8–44.8) 46.9 (47.9–48.9)  Middle/high school 32.5 (31.5–33.5) 34.3 (33.3–35.3) 32.4 (31.4–33.4) 27.1 (26.7–27.5) 8.6 (8.0–9.2)  College 18.3 (17.5–19.1) 22.7 (21.9–23.5) 18.8 (18.0–19.6) 10.8 (10.5–11.1) 3.3 (2.9–3.7)  Household income (%) <10,000 yuan 14.0 (13.4–14.6) 11.6 (11.0–12.2) 12.6 (12.0–13.2) 15.6 (15.3–15.9) 22.1 (21.3–22.9) <0.001 10,000–19,999 yuan 38.8 (37.8–39.8) 35.8 (34.8–36.8) 37.1 (36.1–38.1) 38.0 (37.5–38.5) 38.7 (37.7–39.7)  20,000–29,999 yuan 29.4 (28.5–30.3) 31.3 (30.3–32.3) 30.4 (29.4–31.4) 28.6 (28.2–29.0) 27.8 (27.0–28.6)  >30,000 yuan 17.8 (17.0–18.6) 21.4 (20.6–22.4) 19.8 (19.0–20.6) 17.8 (17.4–18.2) 11.4 (10.8–12.0)  Occupation (%) Professional 21.6 (20.8–22.4) 25.9 (25.1–26.7) 22.2 (21.4–23.0) 16.4 (15.6–17.2) 6.8 (6.2–7.4) <0.001 Clerical 12.8 (12.2–13.4) 12.4 (11.6–12.8) 13.2 (12.6–13.8) 13.7 (13.4–14.0) 19.5 (18.7–20.3)  Manual labour/others 21.8 (21.0–22.6) 21.0 (20.2–21.8) 23.1 (22.3–23.9) 23.6 (23.2–24.0) 20.0 (19.2–20.8)  Housewife/retired 43.8 (42.8–44.8) 40.7 (39.7–41.7) 41.6 (40.6–42.6) 46.3 (45.8–46.8) 53.6 (52.6–54.6)  Hypertension (%) 18.4 (17.6–19.2) 18.3 (17.5–19.1) 18.8 (18.0–19.6) 19.8 (19.4–20.2) 16.6 (15.8–17.4) 0.12 n Means of energy intake, BMI, WHR and standard weight are adjusted by age Weight at 20 years and number of live births were not normally distributed and could not be adjusted for age a p b p 2 p Table 2 Associations between type 2 diabetes mellitus and duration of breast-feeding for all participants, Shanghai Women’s Health Study   % Cases per person-years RR 95% CI p Breast-feeding No 18.35 225/53,038 1.00   Yes 81.65 1,336/234,916 0.88 0.76–1.02  Duration of breast-feeding  Years        0 18.35 225/53,038 1.00  0.01   >0–0.99 47.99 416/118,511 0.88 0.75–1.04    >0.99–1.99 17.31 343/57,418 0.89 0.75–1.06    >1.99–2.99 7.57 242/26,451 0.88 0.71–1.07    >2.99–3.99 4.15 148/14,929 0.75 0.59–0.96    ≥4 4.63 187/17,606 0.68 0.52–0.90   Years per child        0 18.35 225/53,038 1.00  0.11   >0–0.49 15.10 200/43,615 0.91 0.75–1.10    >0.49–0.99 44.20 659/127,453 0.87 0.74–1.02    ≥1 22.35 477/63,848 0.87 0.78–1.03  Values are adjusted for age, daily energy intake, BMI, WHR, number of live births, smoking, alcohol consumption, physical activity, education, income, occupation and hypertension 2 p 3 r p 4 Table 3 Incidence of type 2 diabetes by duration of breast-feeding, Shanghai Women’s Health Study   0 >0 to 6 months >6 to 11 months >11 to 35 months ≥36 months   RR 95% CI RR 95% CI RR 95% CI RR 95% CI RR 95% CI p Model 1 1.00  0.81 0.66–0.99 0.88 0.74–1.07 1.08 0.92–1.27 1.03 0.84–1.25 0.20 Model 2 1.00  0.84 0.69–1.03 0.89 0.74–1.07 0.99 0.85–1.18 0.88 0.71–1.08 0.75 Model 3 1.00  0.84 0.69–1.03 0.88 0.73–1.05 0.90 0.76–1.05 0.76 0.62–0.93 0.06 Model 4 1.00  0.87 0.71–1.06 0.91 0.76–1.09 0.98 0.83–1.15 0.84 0.68–1.03 0.39 Model 5 1.00  0.89 0.71–1.06 0.89 0.74–1.07 0.89 0.76–1.05 0.75 0.61–0.92 0.04 Model 6 1.00  0.87 0.71–1.06 0.89 0.75–1.07 0.89 0.75–1.05 0.73 0.58–0.91 0.05 Model 1 age-adjusted only; Model 2: age plus other confounders, but not BMI, WHR or number of live births; Model 3: model 2 plus BMI; Model 4: model 2 plus WHR; Model 5: model 2 plus BMI and WHR; Model 6: model 5 plus number of live births Table 4 Associations between type 2 diabetes mellitus and duration of breast-feeding, Shanghai Women’s Health Study (confirmed cases of diabetes only)   % Cases per person-years RR 95% CI p Breast-feeding No 18.40 128/52,848 1.00  0.17 Yes 81.60 741/233,836 0.87 0.72–1.06  Duration of breast-feeding  Months        0 18.40 128/52,848 1.00  0.17   >0–6 16.91 86/48,617 0.80 0.61–1.05    >6–11 24.21 159/69,558 0.98 0.77–1.24    >11–35 29.11 314/83,398 0.84 0.68–1.05    >36 11.37 182/32,263 0.74 0.55–1.01   Years        0 18.40 128/52,848 1.00  0.01   >0–0.99 48.20 297/11,817 0.91 0.73–1.13    >0.99–1.99 17.25 200/57,156 0.90 0.72–1.13    >1.99–2.99 7.48 99/26,242 0.72 0.54–0.95    >2.99–3.99 4.10 65/14,809 0.72 0.52–1.01    >4 4.57 78/17,454 0.67 0.46–0.96   Years per child        0 18.40 128/52,848 1.00  0.23   >0–0.49 15.13 116/43,443 0.94 0.73–1.21    >0.49–0.99 44.21 359/126,916 0.84 0.69–1.04    >1 22.25 266/63,447 0.88 0.70–1.11  Values are adjusted for age, daily energy intake, BMI, WHR, number of live births, smoking, alcohol consumption, physical activity, education, income, occupation and hypertension 5 p p p p Table 5 Duration of breast-feeding and risk of type 2 diabetes mellitus, Shanghai Women’s Health Study   Time since last pregnancy ≤10 years since last pregnancy >10 years since last pregnancy RR 95% CI p RR 95% CI p All participants  Months         0 1.00  0.05 1.00  0.20   >0–6 0.90 0.57–1.42  0.88 0.70–1.10    >6–11 0.83 0.55–1.25  0.91 0.74–1.12    >11–35 0.63 0.40–0.98  0.94 0.79–1.13    ≥36 0.66 0.30–1.48  0.76 0.60–0.97   Years         0 1.00  0.04 1.00  0.06   >0–0.99 0.87 0.60–1.24  0.90 0.75–1.08    >0.99–1.99 0.68 0.43–1.09  0.94 0.77–1.13    >1.99–2.99 0.44 0.21–0.92  0.94 0.76–1.17    >2.99–3.99 0.69 0.29–1.62  0.79 0.61–1.02    ≥4 0.49 0.18–1.35  0.72 0.54–0.96   Years per child         0 1.00  0.03 1.00  0.45   >0 –0.49 1.00 0.64–1.59  0.91 0.74–1.13    >0.49–0.99 0.74 0.51–1.07  0.90 0.76–1.07    ≥1 0.65 0.41–1.05  0.92 0.76–1.10  Confirmed diabetes only  Months         0 1.00  0.26 1.00  0.34   >0–6 0.80 0.41–1.53  0.81 0.60–1.09    >6–11 0.96 0.55–1.65  0.98 0.75–1.27    >11–35 0.69 0.39–1.22  0.87 0.69–1.11    >36 0.60 0.23–1.57  0.78 0.56–1.08   Years         0 1.00  0.10 1.00  0.06   >0–0.99 0.90 0.55–1.48  0.91 0.72–1.16    >0.99–1.99 0.80 0.44–1.44  0.93 0.72–1.19    >1.99–2.99 0.44 0.18–1.09  0.76 0.56–1.03    >2.99–3.99 0.53 0.17–1.59  0.77 0.54–1.09    >4 0.53 0.16–1.73  0.71 0.48–1.04   Years per child         0 1.00  0.25 1.00  0.44   >0–0.49 0.97 0.52–1.82  0.94 0.71–1.25    >0.49–0.99 0.81 0.49–1.33  0.86 0.68–1.08    >1 0.73 0.40–1.34  0.92 0.71–1.17  Values are adjusted for age, daily energy intake, BMI, WHR, number of live births, smoking, alcohol consumption, physical activity, education, income, occupation and hypertension Discussion 13 6 p p 6 6 p p 14 20 p 21 2 3 4 5 22 23 Our study has several strengths. Our population is representative of urban Shanghai. The high follow-up rates minimised the possibility of selection bias. In addition, the prevalence of breast-feeding in this population was very high (81.62%), and there was a high number of live births (of the total number of women in our study at baseline, 74,942 or 96.4% were parous). In addition, the extensive information on potential confounders and the large study size allowed us to examine the effect of duration of breast-feeding on the development of type 2 diabetes mellitus in detail. 13 24 p p p r r In summary, we found that breast-feeding duration was inversely associated with the risk of type 2 diabetes mellitus in this population. Together with results from two other large US cohorts and some clinical evidence of improved glucose homeostasis in breast-feeding women, these data suggest that breast-feeding may reduce the risk of type 2 diabetes mellitus in middle-aged women.