Table 3.

Contributions to the Explainable Variability and Model Fit for Predictors of log Total Modified Sharp Score

Predictor	Total* Association	Independent* Association	AIC**	Difference†	p
Adiponectin	14.0%	6.1%	547.65	10.29	0.0008
Age	17.0%	2.6%	541.58	4.22	0.016
RA duration	62.3%	68.0%	589.37	52.01	<0.0001
RF seropositivity	13.6%	6.3%	548.52	11.16	0.0004
Presence of any SE alleles	8.2%	2.0%	545.12	7.76	0.014
PAD4 antibodies	23.6%	4.1%	545.81	8.45	0.005
log CRP	12.6%	3.8%	543.36	6.00	0.006
Use of biologic DMARDs	2.4%	6.8%	552.00	14.64	0.0013

^*Total association is the proportion of the total explainable variability (coefficient of determination: R²) contributed by the predictor in the univariate model expressed as a proportion of the total variability explained by all the predictors from the final model (R²=0.501). Independent association is the proportion of the total explainable variability contributed by the predictor in the final adjusted model. The sum of the total associations is greater than 100% due to overlapping associations of the predictors in the unadjusted models. The sum of the independent associations is equal to 100%.

^**Values indicate the change in Akaike’s Information Criterion (AIC) for the nested model that includes all of the predictors in the table excluding the indicated covariate. AIC for the full model = 537.36. Higher AIC values for the model when the predictor is excluded indicate better fit for the model that includes the predictor.

^†Values are the difference in the AIC between the simpler model excluding the predictor of interest and the full model including the predictor of interest. Higher values for the difference indicate a greater independent contribution to the full model