Skip to main content

Table 2 Changes in mean daily steps per day across any valid day/s at 6-month follow-up

From: The effectiveness of the Structured Health Intervention For Truckers (SHIFT): a cluster randomised controlled trial (RCT)

 

Number of clusters

Number of participants

Baseline

Mean (SD)

6-month follow-up

Mean (SD)

Mean change from baseline to 6 months (SD)

SHIFT vs. control at 6 months

Control

SHIFT

Control

SHIFT

Control

SHIFT

Control

SHIFT

Control

SHIFT

Adjusted mean

difference (95% CI)a

p-value

Primary analysis (complete case)b

13

12

119

90

8932 (2922)

9355 (3305)

8216 (2767)

9387 (3455)

− 716 (2109)

32 (2939)

1008 (145, 1871)

0.022

Intention to treatb

 Multiple imputation c

13

12

199

183

–

–

–

–

–

–

335 (− 471, 1141)

0.414

 Worst case scenario d

13

12

199

183

8788 (2843)

9394 (3134)

8244 (2270)

8846 (2527)

− 544 (2499)

− 548 (3056)

399 (− 129, 927)

0.139

 Best case scenario e

13

12

199

183

8788 (2843)

9464 (3127)

8244 (2270)

9344 (2466)

− 543 (2499)

− 120 (2985)

868 (398, 1338)

< 0.001

Sensitivity analyses—effect of number of valid activPAL days (complete-case):

 ≥ 2 days

13

12

118

88

8960 (2919)

9427 (3291)

8238 (2768)

9411 (3488)

− 722 (2117)

− 16 (2949)

981 (102, 1860)

0.029

 ≥ 3 days

13

12

116

87

8925 (2921)

9459 (3296)

8278 (2770)

9456 (3484)

− 647 (2015)

− 4 (2963)

906 (27, 1784)

0.043

 ≥ 4 days

13

12

113

79

8832 (2877)

9385 (3290)

8179 (2710)

9480.69 (3509)

− 653 (2020)

96 (3028)

973 (76, 1870)

0.034

  1. aAdjusted for steps per day at baseline, average waking wear time across baseline and 6 months, and cluster size category (small < 40; large ≥ 40) with a random effect for cluster (depot).
  2. b≥ 1 valid day at baseline and 6 months
  3. cMeans (SDs) cannot be calculated for intention to treat population because multiple imputation methodology was used
  4. dMissing outcome data in the final analysis model (at baseline and 6 months) was replaced using the mean for the standard care arm
  5. eMissing outcome data in the final analysis model (at baseline and 6 months) was replaced using the mean from the respective arm