Skip to main content

Table 4 Impact of the intervention on student’s mean percentage of ‘Everyday’ online lunch items: subgroup analysisa

From: The efficacy of a multi-strategy choice architecture intervention on improving the nutritional quality of high school students’ lunch purchases from online canteens (Click & Crunch High Schools): a cluster randomized controlled trial

Variable

Intervention

Control

Intervention vs Controla

Baseline

Mean (SD)

n = 656

students

Follow up

Mean (SD)

n = 463

students

Baseline

Mean (SD)

n = 675

students

Follow up

Mean (SD)

n = 536

students

Group by time differential

Effect

(95% CI)

P

Group by time by Subgroup differential effect

(95% CI)

P

Student grade

 Grade 7–9

44.0 (33.8)

47.2 (32.1)

42.8 (36.1)

41.6 (36.0)

3.9 (0.2, 7.7)

0.04

 Grade 10–11

45.4 (35.9)

51.2 (32.8)

44.6 (37.3)

37.1 (31.2)

11.7 (4.2, 19.1)

0.002

7.7 (−0.6, 16.1)

0.07

Frequency of use

 High users (≥1 order / week on average)

43.9 (29.7)

49.0 (32.4)

43.8 (30.9)

40.7 (32.3)

7.7 (2.5, 12.9)

0.004

3.88 (−2.92, 10.69)

0.26

 Low users (< 1 order / week on average)

44.6 (36.7)

47.4 (32.2)

42.7 (39.2)

40.8 (37.2)

3.85 (−0.5, 8.2)

0.08

School sector

 Government

43.9 (36.5)

45.9 (33.1)

29.9 (30.1)

27.3 (27.8)

5.1 (−1.4, 11.7)

0.12

 Non-Government

44.6 (33.2)

49.4 (31.7)

46.6 (37.0)

44.1 (36.0)

6.0 (2.0, 10.0)

0.003

0.9 (−6.8, 8.5)

0.82

  1. aData were analyzed using linear mixed models and included a group by time interaction fixed effect, a random intercept for school, a nested random intercept and random time effect for students (to account for repeated measurements between time points), and fixed effects for SEIFA and school sector