Propensity score to detect baseline imbalance in cluster randomized trials: the role of the c-statistic

Table 2 Patient baseline characteristics per group in the study on management of osteoarthritis with a standardized evaluation tool (first motivating example)

Characteristics	Control	Standardized tool	p	SDiff (%)
	n ₀=1495	n ₁=1462
	mean (standard deviation)	mean (standard deviation)
Duration of symptoms (months)	68.0 (69.8)	70.9 (75.0)	0.2737	4.03
Age (years)	67.2 (9.7)	66.4 (10.0)	0.0344	7.81
BMI (kg.m ⁻²)	27.8 (4.9)	27.7 (4.7)	0.3824	3.14
Patient global assessment (0-100)	61.1 (18.2)	56.7 (17.4)	< 0.0001	24.44
Pain evaluation VAS (0-100)	59.4 (16.0)	55.3 (15.1)	< 0.0001	26.32
WOMAC function score (0-100)	45.5 (16.3)	43.8 (16.0)	0.0050	10.32
	n (%)	n (%)
Osteoarthritis in other joints	1366 (91.4)	1311 (89.7)	0.1297	5.81
Prior treatment
IA treatment	434 (29.0)	432 (29.6)	0.7877	1.14
NSAIDs	955 (63.9)	958 (65.5)	0.3689	3.45
SYSADOA	617 (41.3)	605 (41.4)	0.9810	0.22
Male	424 (28.4)	459 (31.4)	0.0782	6.65
Kellgren and Lawrence grade			0.2594
III	724 (48.4)	677 (46.3)		4.25
IV	516 (34.5)	547 (37.4)		6.02

BMI: Body Mass Index; VAS: Visual Analogue Scale; WOMAC: Western Ontario and McMaster Universities Arthritis Index; IA: intra-articular; NSAID: non-steroidal anti-inflammatory drug; SYSADOA: systematic slow acting drug for osteoarthritis. SDiff: standardized difference; p: p-value for univariate tests (adjusted t test for quantitative variables, adjusted chi-square test for qualitative variables to take the clustering into account). Bold values are significant tests at a 5 % significance level

ISSN: 1471-2288