Table 2.
Geometric means (95% CI) of glycemic and inflammatory parameters by categories of coffee consumptiona
| Never or rarely | < 1 cup/day | 1-2 cup(s)/day | ≥ 3 cups/day | P trendb | ||
|---|---|---|---|---|---|---|
| Fasting plasma glucose (mmol/L) | ||||||
| Model-1c | 4.83 (4.79 - 4.86) | 4.81 (4.75 - 4.86) | 4.86 (4.83 - 4.89) | 4.82 (4.76 - 4.88) | 0.554 | |
| Model-2 | 4.76 (4.69 - 4.82) | 4.73 (4.66 - 4.81) | 4.78 (4.72 - 4.84) | 4.74 (4.66 - 4.82) | 0.916 | |
| HOMA-IR | ||||||
| Model-1 | 1.58 (1.52 - 1.65) | 1.54 (1.44 - 1.64) | 1.56 (1.51 - 1.62) | 1.53 (1.42 - 1.64) | 0.449 | |
| Model-2 | 1.47 (1.38 - 1.57) | 1.41 (1.30 - 1.52) | 1.41 (1.32 - 1.50) | 1.34 (1.23 - 1.46) | 0.007 | |
| Model-3d | 1.43 (1.34 - 1.53) | 1.36 (1.26 - 1.47) | 1.36 (1.28 - 1.45) | 1.32 (1.22 - 1.44) | 0.021 | |
| HOMA-beta | ||||||
| Model-1 | 104.68 (101.2 - 108.28) | 105.57 (100.27 - 111.15) | 100.54 (97.81 - 103.35) | 106.42 (100.24 - 112.98) | 0.715 | |
| Model-2 | 106.24 (99.34 - 113.63) | 106.15 (98.43 - 114.47) | 101.9 (95.64 - 108.57) | 107.02 (98.29 - 116.51) | 0.591 | |
| HbA1c % | ||||||
| Model-1 | 5.79 (5.76 - 5.83) | 5.72 (5.67 - 5.78) | 5.78 (5.76 - 5.81) | 5.8 (5.74 - 5.86) | 0.567 | |
| Model-2 | 5.75 (5.69 - 5.82) | 5.69 (5.62 - 5.76) | 5.72 (5.66 - 5.78) | 5.72 (5.64 - 5.8) | 0.375 | |
| Adiponectin (high molecular weight) µg/ml | ||||||
| Model-1 | 1.03 (0.98 - 1.08) | 1.04 (0.97 - 1.12) | 1.02 (0.98 - 1.06) | 1.03 (0.95 - 1.12) | 0.953 | |
| Model-2 | 0.99 (0.91 - 1.08) | 1.02 (0.92 - 1.12) | 1.00 (0.92 - 1.08) | 1.04 (0.93 - 1.16) | 0.347 | |
| CRP (mg/L) | ||||||
| Model-1 | 1.42 (1.32 - 1.52) | 1.59 (1.43 - 1.77) | 1.49 (1.41 - 1.58) | 1.59 (1.41 - 1.81) | 0.139 | |
| Model-2 | 1.31 (1.16 - 1.49) | 1.43 (1.25 - 1.65) | 1.28 (1.14 - 1.44) | 1.23 (1.05 - 1.44) | 0.185 | |
a HOMA-IR, homeostatic model assessment-insulin resistance; HOMA-beta, homeostatic model assessment-beta cell function; CRP, C-reactive protein. Numbers of participants differ across outcomes because participants with outlier values (response values > 4 SD from mean) were excluded. These numbers across the 4 categories of coffee consumption are fasting plasma glucose: 1187,461,2084,342; HOMA-IR: 1195,468,2102,343; HOMA-beta: 1187, 468, 2097, 343; HbA1c: 967, 388, 1717, 281; HMW adiponectin: 1176, 452, 2056, 333 and CRP: 1168, 450, 2046, 331.
b P-values were obtained from multiple linear regression models with median cups of coffee per week (0, 4,10.5 and 28) as predictors and log transformed parameters as dependant variables.
c Model-1: adjusted for age (years), sex and ethnicity (Chinese, Malay, Indian).
Model-2: adjusted for Model-1 covariates (above) and BMI (kg/m2), physical activity level (kcal/week), education level (primary, secondary, polytechnic/diploma and university), alcohol level (non-drinkers, < 1 serving/day, and ≥ 1 serving/day), cigarette smoking (never-smokers, ex-smokers, current smokers < 10 cigarettes/d, and current smokers ≥ 10 cigarettes/d), history of dyslipidemia (yes/no), history of hypertension (yes/no), and dietary confounders i.e. energy intake (kcal), fiber (per 1000 kcal), cholesterol (per 1000 kcal), PUFA (% energy), MUFA (% energy), SFA (% energy), green, black, Oolong tea (never/< 1cup per week, 2-6 cups per week, ≥ 1 cup(s) per day). HOMA-beta models were further adjusted for HOMA-IR.
d Model-3 was further adjusted for CRP and adiponectin