Iris Kanera

5 123 Lifestyle-related effects of the Kanker Nazorg Wijzer Table 5.2 Results from multilevel analysis of the overall intervention effects on Physical Activity and Diet (Continued) Complete cases analysis 1 Intention-to-treat analysis 2 B SE [95% CI] p p fdr d [95% CI] f 2 B SE [95% CI] p p fdr Dietary behavior outcomes Vegetables (g/p/d) Crude 10.26 4.10 [2.23; 18.30] . 012 .096 11.16 4.19 [2.94; 19.38] .008 .064 Adjusted 9.15 4.10 [1.03; 17.27] . 027 .148 -.37 [-.57; -.17] -.013 9.57 4.21 [1.32; 17.82] .023 .160 Fruit (servings p/d) Crude .08 0.07 [-.06; .23] .242 .323 .07 0.08 [-.08; .22] .339 .452 Adjusted .08 0.07 [-.06; .23] .271 .433 -.15 [-.35; .05] .002 .07 0.08 [-.08; .22] .351 .468 Whole grain bread (p/d) Crude .15 0.12 [-.08; .38] .195 .312 .15 0.12 [-.08; .39] .189 .302 Adjusted .11 0.12 [-.12; .34] .354 .472 -.22 [-.42; -.02] .000 .11 0.12 [-.12; .34] .330 .468 Fish (servings p/w) Crude .37 0.23 [-.09; .82] .113 .264 .34 0.24 [-.13; .82] .157 .302 Adjusted .32 0.23 [-.14; .77] .173 .346 -.28 [-.47; -.08] .004 .32 0.24 [-.17; .80] .198 .396 Note: Multilevel analysis with two-level data structure: persons (level1) nested in hospitals (level2); B = Regression coefficient, d = Cohen’s d, f 2 = Cohen’s f 2 . PA = physical activity; p/w = per week, p/d = per day. Crude analysis: models includes intervention condition, outcome behavior at baseline, and hospital; adjusted analysis: adjusted for gender, age, marital status, education level, income level, employment, baseline BMI, cancer type, having had cancer before, treatment type, time since last treatment, participation in aftercare, comorbidities, baseline vegetable, fruit, bread, and fish consumption. 1 For physical activity outcomes N = 394; for diet outcomes N = 403 2 Imputed data: for physical activity outcomes N = 452; for diet outcomes N = 462