The relationship between synovitis quantified by an ultrasound 7-joint inflammation score and physical disability in rheumatoid arthritis – a cohort study

Table 5 Time-lag model: multivariate analyses

Analysis number	Predictor	β (95% CI)	P value	Percent variability explained by the model (R ²)
1	Previous HAQ	0.762 (0.698; 0.827)	<0.001	64.8
1	Previous PDsynSS	-0.017 (-0.027; -0.007)	0.001	64.8
2	Previous PDsynSS	-0.025 (-0.039; -0.011)	<0.001	32.8
2	Previous DAS28	0.156 (0.109; 0.203)	<0.001	32.8
3	Previous HAQ	0.703 (0.620; 0.786)	<0.001	65.4
	Previous PDsynSS	-0.022 (-0.034; -0.011)	<0.001
	Previous DAS28	0.050 (0.005; 0.096)	0.029
4	Previous HAQ	0.754 (0.690; 0.819)	<0.001	64.2
4	Previous GSsynSS	-0.009 (-0.017; -0.001)	0.024	64.2
5	Previous GSsynSS	-0.012 (-0.023; -0.001)	0.036	32.5
5	Previous DAS28	0.135 (0.088; 0.183)	<0.001	32.5
6	Previous HAQ	0.707 (0.623; 0.792)	<0.001	64.5
	Previous GSsynSS	-0.013 (-0.022; -0.004)	0.006
	Previous DAS28	0.039 (-0.007; 0.085)	0.096

Predicted variable is current health assessment questionnaire (HAQ) score. Predictors used in the six analyses are either previous HAQ score or disease activity score in 28 joints (DAS28) or both, and previous power Doppler (PD) synovitis (SS) or greyscale (GS) syn sum-score (SS) (previous = measured 12 months ago. ES = erosions score. P-values higlighted in italics are significant at p < 0.05

ISSN: 1478-6362