The effect of changing movement and posture using motion-sensor biofeedback, versus guidelines-based care, on the clinical outcomes of people with sub-acute or chronic low back pain-a multicentre, cluster-randomised, placebo-controlled, pilot trial

Table 4 Results for secondary outcome measures

	Analysis n=	Movement Biofeedback	Movement Biofeedback	End of treatment period
		Group effect ^#	Group by time effect ^#	Mean (SD)
		Beta coefficient (95 % CI)	Group by time effect ^#
		Beta coefficient (95 % CI)	Beta coefficient (per 10 days in the treatment period) (95 % CI)
Daily pain score	Sites = 8	−0.62 (−1.25 to 0.01)	−0.051 (−0.075 to −0.026)	Movement Biofeedback Group
(0 to 10 scale)	Clinicians = 8
	Participants = 98	p = 0.053	p < 0.001
	Assessments = 6,036			4.26 (3.44 to 4.99)
	Assessments = 6,036			Guidelines-based Care Group4.54 (3.88 to 5.19)
LBP recurrence (difference in proportions of days with reported recurrence)	Sites = 8	−0.018 (−0.129 to 0.093)	0.003 (−0.002 to 0.008)	Movement Biofeedback Group
	Clinicians = 8
	Participants = 100	p = 0.752	p = 0.263
	Assessments = 5,999			0.230 (0.098 to 0.362)
				Guidelines-based Care Group
				0.173 (0.070 to 0.276)
Analgesic use	Sites = 8	0.056 (−0.099 to 0.211)	−0.007 (−0.013 to −0.002)	Movement Biofeedback Group
(difference in proportion of days with reported taking of analgesics)	Clinicians = 8			Movement Biofeedback Group
	Participants = 98	p = 0.483	p = 0.008	0.288 (0.137 to 0.440)
	Assessments = 5,815			0.288 (0.137 to 0.440)
				Guidelines-based Care Group
				0.360 (0.109 to 0.612)
Number of pain and	Sites = 8	0.054 (0.003 to 0.107)	0.004 (0.002 to 0.007)	Movement Biofeedback Group
medication free days	Clinicians = 8			0.064 (−0.034 to 0.163)
				Guidelines-based Care Group
				0.036 (−0.006 to 0.077)

^#The main effect of group indicates the average difference between the groups across treatment and outcome time points. The time-by-group interaction effect indicates the average difference between the groups in the rate of change over time. Both calculated by use of multilevel mixed-effects linear regression adjusted for baseline value of the outcome measure, age, gender, and duration of back pain episode (fixed effects) and treatment site, clinician and individual patient (random effects)
FABQpa = Fear Avoidance Beliefs Questionnaire (physical activity subscale) where low scores are better
Statistically significant p-values are bolded

ISSN: 1471-2474