Table 3 Mean baseline-adjusted change in intervention and control groups, effect size and significance of treatment effect in completers and intention-to-treat population

From: Improving diet, activity and wellness in adults at risk of diabetes: randomized controlled trial

Variable (per week)

Change in completers at 6 months (95% CI)

  
 

Alive-PD n=100a

Control n=147a

Effect size b

Pc

Pd

Aerobic activity (days per week)

1.21 (0.94, 1.47)

0.42 (0.20, 0.64)

0.49

<0.001

<0.001

Red meat

−0.91 (−1.31, −0.51)

−0.93 (−0.26, −0.60)

0.07

0.95

0.52

Bread

−2.39 (−2.91, −1.87)

−1.29, (−1.72, −0.86)

0.19

0.002

0.046

Pasta, white rice

−1.40 (−1.32, −0.78)

−0.69 (−0.83, −0.33)

0.36

0.001

0.010

Sum of bread, pasta, white rice

−3.77 (−4.44, −3.10)

−1.99 (−2.55, −1.44)

0.34

<0.001

0.006

Sweets

−2.26 (−2.69, −1.82)

−1.02 (−1.38, −0.67)

0.40

<0.001

0.011

Fruits

2.03 (1.43, 2.62)

0.09 (−0.41, 0.58)

0.58

<0.001

<0.001

Vegetables

1.75 (1.14, 2.35)

0.05 (−0.45, 0.55)

0.43

<0.001

<0.001

Sum of fruits and vegetables

3.71 (2.73, 4.70)

0.16 (−0.65, 0.98)

0.62

<0.001

<0.001

  1. Abbreviation: CI, confidence interval.
  2. a6-month value and 95% confidence limits from least squares means in models of following form: 6-month value=baseline+treatment group.
  3. bHedges g. Similar to Cohen’s d, Hedges g is the difference between the two mean changes divided by the pooled s.d., and represents the difference in s.d.’s.
  4. cSignificance of the difference in change in intervention vs control groups over 6 months, from model described in ‘a’.
  5. dSignificance of the difference in change in intervention vs control groups over 6 months, from model described in ‘a’: intention-to-treat analysis, all initially enrolled subjects, missing data imputed using Last Observation Carried Forward imputation approach.
Back to article page