Table 4.

Association between MoCA score, CDR score and number of reported falls and balance confidence in regression models.

	Fallers (OR, 95% CI), p value	Balance confidence (ABC-6)
MoCA score
Unadjusted model	0.87 (0.80–0.95), p = 0.001	beta = 14.63, p < 0.001
Adjusted model 1e	0.86 (0.78–0.95), p = 0.004a	beta = 9.57, p = 0.005c
Adjusted model 2f	0.87 (0.80–0.95), p = 0.001b	beta = 10.15, p = 0.003d
Unadjusted modelg	0.92 (0.82–1.02), p = 0.10	beta = 14.98, p < 0.001
CDR score
Unadjusted model	OR = 3.96 (1.69–9.28), p = 0.002	beta = −138.97, p < 0.001
Adjusted model 1e	OR = 3.95 (1.50–10.45), p = 0.006a	beta = −112.38, p = 0.001c
Adjusted model 2f	OR = 3.96 (1.69–9.28), p = 0.002b	beta = −109.07, p = 0.001d
Unadjusted modelg	OR = 3.43 (0.78–15.06), p = 0.10	beta = −194.01, p = 0.001

^aAdjusted for age, total tremor score, and number of medications that could affect cognitive performance.

^bNo covariates included in the model.

^cAdjusted for age, total tremor score, education, and number of medications that could affect cognitive performance.

^dAdjusted for age and total tremor score.

^eLess restrictive model: model 1 used less restrictive criteria for confounding [an association (p < 0.05) between the confounding variable and either the independent or dependent variable].

^fMore restrictive model: model 2 used more restrictive criteria for confounding [an association (p < 0.05) between the confounding variable and both the independent and dependent variables].

^gUnadjusted model after having removed 13 cases with dementia.

Bolded values are significant.

ABC-6, Transformed Activities of Balance Confidence Scale score; CDR, Clinical Dementia Rating; CI, confidence interval; MoCA, Montreal Cognitive Assessment; OR, odds ratio.