Can the prediction model using regression with optimal scale improve the power to predict the Parkinson's dementia?

Table 5 Results of regression with optimal scale

Test	b	SE by boost 1	df	F	P value
K-MMSE	-0.522	0.168	2	9.684	< 0.001
KMoCA	-0.206	0.238	3	0.750	0.527
CDR (Global CDR score)	0.127	0.269	1	0.222	0.639
CDR (sum of boxes)	-0.271	0.412	3	0.431	0.732
K-IADL	0.237	0.224	2	1.119	0.334
UPDRS (Total UPDRS score)	0.433	0.444	3	0.949	0.423
UPDRS (Motor UPDRS score)	-0.338	0.330	3	1.045	0.380
H&Y staging	0.440	0.197	3	5.008	0.004
Schwab and England ADL	0.353	0.333	2	1.123	0.333

The regression model was adjusted for demographic factors, family disease history, health behaviors, disease history, Parkinson's disease-related motor signs and neuropsychological test. 1 SE by boost: Standard error by bootstrap (with n = 1000); K-MoCA: Korean version of montreal cognitive assessment; K-MMSE: Korean version of mini mental state examination; CDR: Clinical dementia rating; UPDRS: Untitled parkinson disease rating; H&Y staging: Hoehn and Yahr staging; K-IADL: Korean version of instrumental activities of daily living; Schwab and England ADL: Schwab and England avtivities of daily living scale.