The index ‘Treatment Duration Control’ for enabling randomized controlled trials with variation in duration of treatment of chronic pain patients

Table 5 TDC-values for different modes of addition and initial Contrast-values of reference items

Group	Occasion	Mode of addition			% difference relative to NA		Test	p-level
Group	Occasion	s-A	c-A	NA	s-A	c-A	Test	p-level
use of positive value relative to baseline as initial Contrast-value for added reference items
all	EM	-0.635	-0.653	-0.659	3.6	1.0	s-A vs. NA	***
patients		(0.300)	(0.283)	(0.270)			c-A vs. NA	ns
n = 118							s-A vs. c-A	***
all	LM	-0.541	-0.556	-0.585	7.5	4.9	s-A vs. NA	****
patients		(0.339)	(0.324)	(0.298)			c-A vs. NA	*
n = 118							s-A vs. c-A	**
patients with	EM	-0.420	-0.454	-0.514	18.2	11.8	s-A vs. NA	****
added items		(0.333)	(0.309)	(0.276)			c-A vs. NA	**
from assessor							s-A vs. c-A	**
n = 33
patients with	LM	-0.229	-0.268	-0.395	42.0	32.0	s-A vs. NA	****
added items		(0.281)	(0.271)	(0.254)			c-A vs. NA	***
from assessor							s-A vs. c-A	**
n = 33
use of zero as initial Contrast-value for added reference items
all	EM	-0.651	-0.665	-0.659	1.2	-1.0	s-A vs. NA	**
patients		(0.274)	(0.262)	(0.270)			c-A vs. NA	ns
n = 118							s-A vs. c-A	**
all	LM	-0.570	-0.583	-0.585	2.5	0.3	s-A vs. NA	***
patients		(0.303)	(0.293)	(0.298)			c-A vs. NA	ns
n = 118							s-A vs. c-A	*
patients with	EM	-0.485	-0.498	-0.514	5.6	3.1	s-A vs. NA	**
added items		(0.274)	(0.267)	(0.276)			c-A vs. NA	ns
from assessor							s-A vs. c-A	ns
n = 33
patients with	LM	-0.342	-0.357	-0.395	13.3	9.5	s-A vs. NA	***
added items		(0.230)	(0.225)	(0.254)			c-A vs. NA	*
from assessor							s-A vs. c-A	ns
n = 33

mean and (between brackets) SD values of TDC for two ways of attributing an initial Contrast-value of an added reference item at the visit of addition, i.e. (i) by calculating the Contrast between the large score value at addition with respect to the low value at baseline using equation (1) (see text), or (ii) by attributing the value zero as initial Contrast-value (cf. Discussion). For explanation of the abbreviations under ‘occasion’ and ‘mode of addition’, see Table 4. Test: Bonferroni’s multiple comparison test which could always be applied as all 1-way ANOVAs for repeated measures were significant (p < 0.0001-0.05). p-level: ****, p < 0.0001; ***, p < 0.001; **, p < 0.01; *, p < 0.05; ns, non-significant.

ISSN: 1471-2288