Findings from a cluster randomised feasibility study of a school-based physical activity role model intervention (CHARMING) for 9–10-year-old girls

Table 7 Primary and secondary outcomes at baseline and follow-up for the intervention group and a control group

	Baseline				12 months follow-up
	Intervention (n = 72)		Control (n = 23)		Intervention (n = 61)		Control (n = 17)		Adjusted mean difference^a (95% CI)	ICC (95% CI)^b	n
	% / M (SD)	N	% / M (SD)	n	% / M (SD)	n	% / M (SD)	N
Primary outcome: Average daily minutes spent in MVPA^c	48.9 (15.7)	47	46.6 (18.3)	11	64.4 (24.5)	35	44.5 (21.1)	6
Missing		25		12		26		11
Complete case analysis (baseline and follow-up data)	50.8 (19.8)	24	49.6 (23.0)	6	64.4 (25.5)	24	44.5 (21.1)	6	19.0 (1.35 to 36.70)	0.014	30^d
Secondary outcomes
PAQ-C scale^e	3.0 (0.7)	44	3.1 (0.6)	21	2.9 (0.8)	52	2.8 (0.5)	14	0.19 (−0.26 to 0.64)	0.084 (0.002 to 0.77)	47
Missing		28		2		9		3
LCQ Autonomy scale^f	25.9 (6.4)	48	24.7 (7.1)	17	26.5 (6.9)	51	29.1 (5.0)	16	− 2.17 (− 7.12 to 2.78)	0.178 (0.01 to 0.80)	47
Missing		24		6		10		1
LCQ Competence scale^f	23.7 (6.1)	47	24.4 (6.3)	16	24.5 (6.4)	54	28.0 (5.3)	15	− 3.94 (− 8.23 to 0.35)	0.028	49
Missing		25		7		7		2
LCQ Relatedness scale^f	24.6 (7.4)	47	24.7 (6.7)	17	25.1 (7.4)	48	27.4 (6.4)	15	− 1.68 (− 6.83 to 3.48)	0.027	44
Missing		25		6		13		2
Average daily minutes spent sedentary	518.1 (68.8)	47	508.2 (65.4)	11	469.9 (71.2)	34	567.9 (89.9)	6	− 87.0 (− 200.7 to 26.6)	0.291 (0.004 to 0.98)	30^d
Missing		25		12		27		11
									Adjusted odds ratio^g (95% CI)
School sport club/PA at least 3/week	10.5	57	9.5	21	31.6	57	35.3	17	1.16 (0.29 to 4.63)	0.018	63
Missing		15		2		4		0
Non-school sport club/PA at least 3/week	27.9	61	35.0	20	42.1	57	35.3	17	1.07 (0.27 to 4.17)	0.006	66
Missing		11		3		4		0

SD standard deviation, CI confidence interval, ICC intra-cluster coefficient, M mean, SD standard deviation, MVPA moderate to vigorous physical activity, PAQ-C self-reported physical activity, LCQ Learning Climate Questionnaire, PA physical activity
^aAdjusted for baseline outcome and school. Mean differences are intervention minus control. A positive difference favours intervention; a negative difference favours Control
^b95% confidence interval for some ICCs could not be estimated; assented to accelerometer at follow-up n = 77 (intervention group n = 62; control group n = 15)
^dLimited to those who had valid data at baseline and follow-up
^eScore range 1 = less to 5 = more physical activity
^fScore range 6 = less, e.g. autonomy to 36 more autonomy
^gAdjusted for baseline outcome and school. Odds ratio compares intervention to control as a ratio: OR > 1 favours intervention; < 1 favours Control
^hIntervention group only at 12 months follow-up

ISSN: 2055-5784