Table 2

Effect (odds ratios with 95% CI) of legume intake and selected factors on the risk of diabetes among men (n=56,742), India, 2005-06
Predictors and confounders Model 1 OR [95% CI] Model 2 OR [95% CI] Model 3 OR [95% CI] Model 4 OR [95% CI] p value for trend
Legumes intake <0.0001
 Daily 0.57[0.32–0.99] 0.68[0.38–1.22] 0.62[0.35–1.11] 0.70[0.39–1.26]
 Weekly 0.39[0.22–0.69] 0.56[0.31–1.00] 0.51[0.28–0.91] 0.54[0.30–0.98]
 Occasionally 0.42[0.23–0.79] 0.54[0.29–1.03] 0.52[0.28–0.99] 0.56[0.30–1.07]
 NeverR 1.00[Reference] 1.00[Reference] 1.00[Reference] 1.00[Reference]
Consumption of other food items
Milk or curd <0.0001
 Daily 0.84[0.63–1.12] 0.79[0.60–1.06]
 Weekly 0.71[0.52–0.98] 0.64[0.47–0.88]
 Occasionally 0.64[0.48–0.87] 0.60[0.44–0.81]
 NeverR 1.00[Reference] 1.00[Reference]
Green leafy vegetables 0.002
 Daily 1.04[0.72–1.49] 1.00[0.69–1.43]
 Weekly 1.17[0.81–1.69] 1.15[0.79–1.66]
 Never/Occasionally 1.00[Reference] 1.00[Reference]
Fruits <0.0001
 Daily 0.34[0.23–0.51] 0.35[0.22–0.50]
 Weekly 0.35[0.24–0.51] 0.38[0.23–0.49]
 Occasionally 0.39[0.28–0.56] 0.43[0.28–0.56]
 Never R 1.00[Reference] 1.00[Reference]
Eggs 0.017
 Daily 1.59[1.05–2.39] 1.28[0.84–1.94]
 Weekly 1.34[0.97–1.83] 1.18[0.86–1.62]
 Occasionally 1.09[0.80–1.49] 1.07[0.79–1.47]
 NeverR 1.00[Reference] 1.00[Reference]
Fish 0.109
 Daily 1.52[1.03–2.25] 1.55[1.04–2.33]
 Weekly 1.32[0.94–1.87] 1.32[0.92–1.89]
 Occasionally 1.11[0.82–1.50] 1.14[0.81–1.60]
 NeverR 1.00[Reference] 1.00[Reference]
Chicken or meat 0.568
 Daily 0.35[0.14–0.86] 0.31[0.12–0.82]
 Weekly 0.93[0.63–1.35] 0.88[0.65–1.42]
 Occasionally 0.83[0.58–1.19] 0.72[0.55–1.15]
 NeverR 1.00[Reference] 1.00[Reference]
Body mass index and lifestyle factors
Body mass index (kg/m2) <0.0001
 ≤18.5 (Underweight) 0.93[0.73–1.18] 0.92[0.72–1.16]
 18.5–22.9 (Normal)R 1.00[Reference] 1.00[Reference]
 23.0–24.9 (Overweight) 1.67[1.34–2.09] 1.61[1.29–2.00]
 ≥25.0 (Obese) 1.84[1.49–2.27] 1.83[1.49–2.27]
Current tobacco smoking <0.0001
 NoR 1.00[Reference] 1.00[Reference]
 Yes 0.93[0.79–1.10] 0.92[0.78–1.09]
Alcohol consumption 0.016
 NeverR 1.00[Reference] 1.00[Reference]
 Occasionally 1.15[0.95–1.40] 1.15[0.94–1.39]
 Once a week 0.96[0.73–1.26] 0.94[0.71–1.23]
 Almost everyday 0.97[0.66–1.44] 0.97[0.65–1.44]
Frequency of watching TV <0.0001
 Not at allR 1.00[Reference] 1.00[Reference]
 Less than once a week 0.86[0.65–1.14] 0.90[0.67–1.19]
 At least once a week 1.23[0.92–1.63] 1.24[0.93–1.65]
 Almost everyday 0.84[0.64–1.10] 0.92[0.70–1.22]
Background factors
Age <0.0001
 20–29R 1.00[Reference]
 30–39 2.12[1.62–2.76]
 40–49 6.96[5.46–8.87]
Education <0.0001
IlliterateR 1.00[Reference]
Literate, <middle school 0.78[0.60–1.01]
Middle school completed 0.75[0.57–0.94]
High school complete and above 0.72[0.53–0.99]
Religion 0.092
 HinduR 1.00[Reference]
 Muslim 1.22[0.94–1.58]
 Christian 0.67[0.41–1.09]
 Sikhs 1.63[0.92–2.89]
 Others 0.51[0.20–1.28]
Caste/tribe 0.001
 Scheduled casteR 1.00[Reference]
 Scheduled tribes 0.49[0.29–0.70]
 Other backward class 0.71[0.62–0.97]
 Others 0.78[0.62–0.99]
 Missing caste 1.15[0.84–1.94]
Wealth index <0.0001
 LowestR 1.00[Reference]
 Second 1.36[0.99–1.88]
 Middle 0.92[0.64–1.32]
 Fourth 1.74[1.21–2.49]
 Highest 3.48[2.33–5.19]
Place of residence <0.0001
 Urban 1.07[0.89–1.29]
 RuralR 1.00[Reference]
Geographic regions <0.0001
 NorthR 1.00[Reference]
 Northeast 2.02[1.14–3.58]
 Central 1.44[0.98–2.13]
 East 3.06[2.09–4.48]
 West 1.49[1.01–2.20]
 South 3.27[2.27–4.71]
Number of cases 56,729 56,665 52,746 52,693

For variable definition see Table 1; RReference category; Model 1 unadjusted; Model 2 adjusted for consumption of other food items and background factors; Model 3 adjusted for BMI and other lifestyle indicators and background factors; Model 4 adjusted for all.

p-Likelihood ratio test for no difference between the three groups for age and standard of living index ignoring the correlated data. As the urban group was expected to have the highest and the rural group the lowest levels of risk factors and disease,.

trend tests were carried out scoring the groups 1 to 3 and using likelihood ratio tests.

Agrawal and Ebrahim

Agrawal and Ebrahim BMC Public Health 2013 13:706   doi:10.1186/1471-2458-13-706

Open Data