Table 7.

Construct validity: correlations between SAT and PRO measures (pooled dataset; N = 698)¹.

				Spearman correlations2
	SAT 1 (Pain relief)	SAT 2 (Activity level)	SAT 3 (Quality of life)	SAT 4 (Undergo treatment again)	SAT 5 (Compared to previous treatment)	3-Item SAT treatment effect subscale	2-Item SAT treatment satisfaction subscale
PGIC	0.90	0.68	0.77	0.44	0.62	0.89	0.61
CGIC3	0.85	0.63	0.73	0.48	0.64	0.83	0.64
NPRS change scores
Pain now	−0.60	−0.47	−0.51	−0.30	−0.37	−0.59	−0.38
Average 24 hour pain	−0.69	−0.53	−0.60	−0.35	−0.45	−0.68	−0.46
BPI change scores
Pain in last 24 hours—at worst	−0.64	−0.50	−0.53	−0.28	−0.40	−0.63	−0.39
Pain in last 24 hours—at least	−0.52	−0.38	−0.44	−0.27	−0.32	−0.50	−0.34
SF-MPQ present pain intensity change scores	−0.45	−0.41	−0.43	−0.20	−0.29	−0.47	−0.29
SF-36v2 change scores
Physical functioning	0.20	0.16	0.25	0.144	0.154	0.23	0.16
Bodily pain	0.43	0.40	0.40	0.27	0.29	0.46	0.33
Vitality	0.25	0.23	0.21	0.095	0.144	0.25	0.144

BPI: Brief Pain Inventory; CGIC: Clinician Global Impression of Change; NPRS: Numeric Pain Rating Scale; PGIC: Patient Global Impression of Change; SAT: Self-Assessment of Treatment.

¹Construct validity of SAT item and subscales were evaluated by examining relationships between SAT items and subscales with component scores of conceptually-related outcome measures using Spearman correlation coefficients. Pooled dataset included data from two clinical trials, Studies C116 (N = 349) and C117 (N = 349).

²Bivariate correlations were significant at P < 0.0001, except where noted otherwise.

³Study C117 only.

⁴ P < 0.001.

⁵ P < 0.05.