Robust estimation of dementia prevalence from two-phase surveys with non-responders via propensity score stratification

Table 1 Baseline characteristics of the sub-population with and without diagnosis in phase II

	Sub-population with diagnosis (n = 3275)	Sub-population without diagnosis (n = 1170)	P-value	SMD
Age,	\(82.89\pm 3.72\)	\(83.08\pm 3.76\)	\(0.14\)	\(0.05\)
Year of Education	\(7.52\pm 4.58\)	\(7.55\pm 4.76\)	\(0.86\)	\(0.01\)
PADL score	\(12.25\pm 5.01\)	\(13.01\pm 6.00\)	\(<0.001\)	0.14
IADL score	\(15.28\pm 8.74\)	\(15.86\pm 9.29\)	\(0.06\)	0.06
MMSE score	\(24.91\pm 5.16\)	\(24.15\pm 5.02\)	\(<0.001\)	0.15
MoCA score	\(21.48\pm 5.56\)	\(20.00\pm 5.21\)	\(<0.001\)	0.27

Data are expressed as mean ± SD
Abbreviations: SMD standardized mean difference, MoCA Montreal Cognitive Assessment, MMSE Mini-Mental State Examination, PADL physical activities of daily living, IADL Instrumental activities of daily living. P-values were calculated using t-test or Wilcoxon signed-rank test, depending on the distribution of covariates. A two-tailed P-value < 0.05 was considered statistically significant. The PADL score, MMSE score, and MoCA scores of the two sub-populations were significantly different in the sense that the P-value was less than 0.05, a result that still holds after Bonferroni correction

ISSN: 1471-2288