Cognitive Screening Instruments to Identify Vascular Cognitive Impairment: A Systematic Review
International Journal of Geriatric Psychiatry
Ghafar, M. Z. A. A., Miptah, H. N., et al. (2019).
International Journal of Geriatric Psychiatry, 34(8), 1114-1127.
This systematic review investigates the psychometric properties of cognitive screening instruments for individuals with vascular cognitive impairment.
No funding received
From database inception to May 31, 2018
Original studies
15
To differentiate vascular dementia from normal cognition, the MoCA:<br />
<ul>
<li>had good internal consistency with Cronbach's α between 0.83 and 0.88, strong interrater reliability (K=0.88) and test-retest reliability (r = 0.97)</li>
<li>had excellent accuracy (AUC > 0.90) with cut-off values between 16.5-21.5</li>
<li>had sensitivity of 77% using a cut-off of <17 and sensitivity of 92.7% when using a cut-off of 16.5</li>
<li>had high specificity over 90%</li>
</ul>
To differentiate vascular mild cognitive impairment from normal cognition, the MoCA:<br />
<ul>
<li>had good internal consistency Cronbach's α = 0.88, moderate IRR (K = 0.74) and strong TRT (r = 0.97)</li>
<li>had good-to-excellent accuracy AUC = 0.87 - 0.95 with cut-off score of 26</li>
<li>had high sensitivity >80%</li>
<li>had specificity <80%</li>
</ul>
To differentiate vascular dementia from normal cognition, the MMSE:<br />
<ul>
<li>had acceptable internal consistency α = 0.77</li>
<li>had good-to-excellent accuracy (AUC = 0.86-0.99) with a cut-off score of 26-26.5</li>
<li>specificity was high at >90%</li>
<li>sensitivity ranged from 62% with a cut-off of <26 to 91.8% with a cut-off of < 26.5</li>
</ul>
To differentiate vascular MCI from normal cognition, the MMSE had sensitivity of 93.7% when combined with the CDT (an improvement from 65% when used as a stand-alone test).
To differentiate vascular dementia from normal cognition, the clock drawing test (CDT):<br />
<ul>
<li>had internal consistency of α = 0.88 when scored using Babin's method and α = 0.50 when using Cahn and Roleau scores</li>
<li>had fair-to-good accuracy (AUC = 0.73-0.83)</li>
<li>had moderate sensitivity of 69% for the Roleau scoring system and high sensitivity of 91.2% for the CLOX1</li>
<li>had low specificity of 58% with the Babins method and 66% for the Roleau method</li>
</ul>
To differentiate vascular MCI from normal cognition, the CDT:<br />
<ul>
<li>had fair-to-good accuracy AUC = 0.66 - 0.81</li>
<li>had low sensitivity (65% - 68.7%) when used as a stand-alone test </li>
<li>had improved sensitivity of 93.7% when combined with the MMSE</li>
</ul>
To differentiate vascular dementia from Alzheimer's Disease, the Combination Clock Drawing Executive Test 2 had the highest sensitivity at 93.3%, low specificity at 60%, and reasonable accuracy (AUC = 0.74).
To differentiate vascular dementia from Alzheimer's Disease (AD), the Brief Memory and Executive Test (BMET) had acceptable to good internal consistency between α = 0.71 and α = 0.85 and strong test-retest reliability (r = 0.80).<br /><br />To differentiate vascular mild cognitive impairment from normal cognition, the BMET had excellent accuracy AUC = 0.94 with a cut-off value of 13.<br /><br />To differentiate vascular MCI from Alzheimer's Disease, the executive functioning and memory ratio test of the BMET had good accuracy with AUC = 0.82 with sensitivity of 63% and specificity of 88%.
To differentiate vascular dementia from Alzheimer's Disease, the most accurate was the Recognition Rey Auditory Verbal Learning Test and Controlled Word Association Test (AUC = 0.89) which had high sensitivity 81% and specificity 84%. <br /><br />To differentiate vascular MCI from Alzheimer's Disease, the Frontal Assessment Battery's 1-minute orthographical fluency subtest had sensitivity of 85% and specificity of 62% with a cut-off of 0.5.
