Table 3 Descriptive statistics of the sample according to MCI diagnostic consistency considering conventional or neuropsychological criteria.

From: Conventional and neuropsychological criteria for mild cognitive impairment show similar prognostic value for dementia across 12 years in a non-clinical setting

 

Conventional criteria consistency

Cognitively normal

Reverters

Consistent

Uncertain

p-value

n = 488

n = 192

n = 133

n = 145

Age at baseline

71.97 (1.25)c

72.18 (1.23)bc

72.31 (1.29)a

72.17 (1.33)ab

0.019

Female

238 (48.8%)

100 (52.1%)

75 (56.4%)

81 (55.9%)

0.277

Education

7.00 (3.17)

7.66 (3.76)

7.11 (3.03)

6.74 (3.22)

0.054

Incident dementia

47 (9.6%)a

20 (10.4%)a

43 (32.3%)b

38 (26.2%)b

 < 0.001

Time to dementia

90.09 (40.56)ab

108.40 (32.44)a

109.63 (40.21)a

77.92 (46.09)b

0.004

Number of Assessment

4.00 [2.00, 5.00]b

5.00 [2.00, 5.00]a

4.00 [2.00, 5.00]b

4.00 [2.00, 5.00]b

 < 0.001

Wave at first MCI diagnosis

/

2.00 [1.00, 4.00]b

2.00 [1.00, 4.00]b

4.00 [1.00, 5.00]a

 < 0.001

Number of MCI diagnosis

/

2.00 [1.00, 4.00]b

2.00 [2.00, 5.00]a

1.00 [1.00, 1.00]c

 < 0.001

 

Neuropsychological criteria consistency

Cognitively normal

Reverters

Consistent

Uncertain

p-value

n = 232

n = 246

n = 285

n = 195

Age at baseline

71.91 (1.19)c

71.99 (1.24)bc

72.28 (1.32)a

72.13 (1.28)ab

0.006

Female

130 (56.0%)

117 (47.6%)

139 (48.8%)

108 (55.4%)

0.141

Education

7.19 (3.29)

7.24 (3.40)

7.14 (3.09)

6.80 (3.46)

0.524

Incident dementia

15 (6.5%)a

9 (3.7%)a

73 (25.6%)b

51 (26.2%)b

 < 0.001

Time to dementia

119.07 (31.18)a

106.33 (38.49)ab

103.84 (38.49)a

73.61 (45.27)b

 < 0.001

Number of assessment

4.00 [2.00, 5.00]b

5.00 [2.00, 5.00]a

4.00 [2.00, 5.00]b

4.00 [2.00, 5.00]b

 < 0.001

Wave at first MCI diagnosis

/

1.00 [1.00, 4.00]b

1.00 [1.00, 4.00]b

3.00 [1.00, 5.00]a

 < 0.001

Number of MCI diagnosis

/

2.00 [1.00, 4.00]b

3.00 [2.00, 5.00]a

1.00 [1.00, 1.00]c

 < 0.001

  1. P value indicates significance using ANOVA for normally distributed continuous variables, Kruskal–Wallis test for ordinal variables, and Chi-squared test for categorical variables. Superscript letters denote significance at post-hoc analysis with a > b > c.