Table 2

Association between anthropometric measures and diabetes risk by race/ethnicity*

Overall (n = 136,112)NHW (n = 115,412)American Indian or Alaska Native (n = 524)Asian (n = 3,484)Black or African American (11,370)Hispanic/Latina (n = 5,322)P for interaction
BMI, kg/m2 (%)0.0002
 <25 (36.5)111111
 25 to <30 (35.2)1.66 (1.59–1.73)1.62 (1.55–1.70)1.45 (0.72–2.94)2.20 (1.79–2.71)1.65 (1.41–1.92)1.82 (1.47–2.25)
 ≥30 (28.3%)3.01 (2.89–3.13)3.02 (2.88–3.15)3.64 (1.86–7.15)3.38 (2.55–4.46)2.57 (2.22–2.97)3.41 (2.76–4.22)
 In continuous**1.43 (1.41–1.44)1.45 (1.43–1.47)1.56 (1.24–1.95)1.51 (1.39–1.63)1.29 (1.25–1.33)1.39 (1.32–1.46)<0.0001
WC, cm (%)<0.0001
 <78 (30.5)111111
 78–88 (32.1)1.59 (1.52–1.67)1.52 (1.44–1.60)1.06 (0.48–2.35)2.61 (2.08–3.27)1.78 (1.50–2.11)1.73 (1.38–2.17)
 >88 (37.4)3.24 (3.10–3.39)3.15 (3.01–3.31)2.55 (1.27–5.14)4.37 (3.39–5.64)3.13 (2.67–3.66)3.98 (3.22–4.93)
 In continuous**1.54 (1.52–1.55)1.55 (1.53–1.57)1.57 (1.29–1.91)1.93 (1.76–2.12)1.42 (1.37–1.47)1.58 (1.50–1.67)<0.0001
WHR (%)<0.0001
 <0.78 (37.2)111111
 0.78–0.85 (37.1)1.64 (1.57–1.70)1.61 (1.54–1.68)2.69 (1.25–5.76)2.31 (1.71–3.13)1.66 (1.48–1.85)1.62 (1.34–1.96)
 >0.85 (25.7)2.95 (2.84–3.07)2.95 (2.83–3.09)3.38 (1.62–7.08)4.90 (3.62–6.64)2.48 (2.22–2.77)3.34 (2.77–4.03)
 In continuous**1.22 (1.21–1.23)1.22 (1.21–1.23)1.25 (1.07–1.46)1.19 (1.15–1.24)1.25 (1.22–1.29)1.23 (1.19–1.27)0.17
  • Data are HR (95% CI).

  • *In the multivariable-adjusted models, we adjusted for age at enrollment (in continuous), race/ethnicity (for overall model only), education (high school or less, some college/technical training, college or some postcollege, and master’s or higher), family history of diabetes (no/yes), different study cohorts (participation in OS or CTs), smoking (never, former, and current), alcohol intake (nondrinker, past drinker, current and <7 drinks/week, and current and ≥7 drinks/week), physical activity (<5, 5 to <10, 10 to <20, 20 to <30, and 30+ METs/week), HEI-2005 score, high cholesterol requiring medicine (no/yes), and different treatment assignments for all three CTs.

  • **When anthropometric measures were analyzed as continuous variables, the corresponding HR presents increased risk of diabetes per 1 increased SD.