Skip to main content
. 2018 May 3;1:36. doi: 10.1038/s42003-018-0042-6

Table 5.

Polygenic risk score for risk-taking propensity (created using summary statistics from UKB) related to diet and eating behaviours in the Fenland study

Variable Total (N) Effect (95% CI) r 2 P value
All participants—nutrient intake
 Energy (kcal/day) 8981 803.5 (140.1, 1466.8) 0.042 0.02*
 Total fat (g/day)a 8981 0.52 (0.12, 0.92) 0.042 0.01*
 Fruit and vegetables (g/day)a 8844 0.46 (−0.07, 0.99) 0.044 0.09
 Protein (g/day)a 8981 0.36 (0.06, 0.66) 0.010 0.02*
 Fibre (g/day)a 8981 0.28 (−0.10, 0.66) 0.005 0.15
 Carbohydrates (g/day)a 8981 0.25 (−0.10, 0.60) 0.028 0.16
Men only—eating behaviours
 Emotional eating 1646 94.6 (35.7, 153.6) 0.007 0.002**
 Cognitive restraint 1646 −2.62 (−48.0, 42.7) 0.005 0.91
 Uncontrolled eating 1646 32.0 (−9.3, 73.3) 0.019 0.13
Women only—eating behaviours
 Emotional eating 1869 −21.2 (−82.7, 40.6) 0.002 0.50
 Cognitive restraint 1869 −21.2 (−63.3, 20.8) 0.005 0.32
 Uncontrolled eating 1869 14.8 (−24.0, 53.6) 0.013 0.45
All participants—food-related behaviours OR (95% CI)
 TV snackingb 4414 1.03 (0.99, 1.06) 0.46
 Skipping breakfastb 11,441 1.05 (1.02, 1.07) 0.03*
 Home-cooked foodb 11,439 0.99 (0.97, 1.01) 0.59

All models were linear or logistic regressions of the PRS for risk-taking to the variable, adjusted for age and sex. Sex-stratified models were only adjusted for age

TV snacking was coded: 0 never/rarely; 1 occasionally /usually/always; skipping breakfast was coded: 0 < 2 times a week; 1 ≥ 2 times a week; home-cooked food was coded: 0; > 5 meals a week home-cooked, 1; < 5 meals a week are home-cooked

*Nominally significant (P < 0.05)

**Bonferroni significant after adjustment for 15 tests (P < 0.003)

a Log-transformed

b Logistic regression