Fig. 7.1
Scatter plot of the Montreal Cognitive Assessment (MoCA) and the Mini-mental State Examination (MMSE) scores for normal controls (NC) and subjects with Mild Cognitive Impairment (MCI) and mild Alzheimer’s disease (AD) (Reproduced with permission [1])
The test-retest reliability was 0.92. The internal consistency of the MoCA was good with a Cronbach alpha on the standardized items of 0.83 [1]. In addition, the positive and negative predictive values for the MoCA were excellent for MCI (89 % and 91 %, respectively) and mild AD (89 % and 100 %, respectively).
7.3.2 Recommendations
The Third Canadian Consensus Conference on the Diagnosis and Treatment of Dementia (CCCDTD3) recommended administering the MoCA to subjects suspected to be cognitively impaired who perform in the normal range on the MMSE [123]. Immediate and Delayed recall, Orientation, and letter F fluency subtests of the MoCA have been proposed by the National Institute for Neurological Disorders and Stroke (NINDS) and the Canadian Stroke Network (CSN) to be a 5-min Vascular Cognitive Impairment screening test administrable by telephone [124]. The MoCA has also been recommended for MCI or Dementia screening in review articles [125–127].
7.3.3 Practical Approach
It is important to emphasize that MoCA is a cognitive screening instrument and not a diagnostic tool, hence clinical judgment, based on thorough clinical evaluation, is important in interpreting MoCA test results and correctly diagnosing patients who present with cognitive complaints. Figure 7.2 illustrates a practical approach to evaluate patients with cognitive complaints. Patients presenting with cognitive complaints and no functional impairment in their activities of daily living (ADL) would be better assessed by the MoCA as first cognitive screening test. Subjects presenting with cognitive complaints and ADL impairment would probably be better assessed by the MMSE first, then the MoCA if the MMSE is in the normal range.
Fig. 7.2
Practical approach to evaluate patients who present with cognitive complaints (Adapted from Nasreddine et al. [1]). ADL activities of daily living, NPV negative predictive value, PPV positive predictive value, MCI mild cognitive impairment
7.4 Demographic Effect on MoCA Performance
7.4.1 Age and Gender Effect
The MoCA has been shown to be age [80, 128–132] and gender independent [80, 128–130, 132–135]. However, in some studies, age negatively correlated with MoCA scores [133, 134, 136]. Upon further analysis, age was a significant factor in MoCA scores mostly for less educated subjects [133] which could be explained by low cognitive reserve among less educated individuals which may result in lessened ability to recruit neuronal network and compensatory age-related cognitive changes. Moreover, lower educated subjects are known to have more vascular risk factors that could also impair their cognition [137]. Comparing to the MMSE, the MoCA provided better ability to detect of age-related cognitive decline in healthy adults and elderly [138].
7.4.2 Education and Literacy Effect
A recent study analyzed how education affects cognitive performance on the MoCA. In cognitively healthy elderly with the clinical dementia rating (CDR) of 0, subjects were divided into three groups: illiterate (education years = 1.06), literate-low educated (education years = 4) and literate-high educated (education years = 14.21) [139]. Orientation item, which is the test of basic information required in daily living, was not affected by either literacy status or education level.
The tasks assessing working memory/attention (digit spans and vigilance), mental calculation (serial-7 subtraction) and 2-dimension processing semantic knowledge (animal naming) were affected by literacy status, not education level.
Education level affected the performance in the following tasks: structural interpretation of complex sentences (repetition), conceptual formation & constructional skill (clock drawing test and abstraction), 3-dimension processing skill (cube copy), planning and inhibition (trail making B), coordination of lexico-phonological knowledge (letter fluency) and encoding and retrieval strategy (verbal memory). Figure 7.3 demonstrates the literacy and education effect on the MoCA sub-items.
Fig. 7.3
The literacy and education effect on the MoCA sub-items among cognitively intact elderly [139]
Originally, the validation study for the MoCA recruited highly educated normal subjects, suggesting a correction of one added point for education of 12 years or less [1]. Subsequent studies locally in Montreal suggest that to better adjust the MoCA for lower educated subjects, 2 points should be added to the total MoCA score for subjects with 4–9 years of education, and 1 point for 10–12 years of education [140]. Education has been consistently reported around the world affecting total MoCA scores [1, 80, 128–131, 133, 134, 141, 142]. Trail making test and digit span of the Japanese version of the MoCA significantly correlate with years of schooling [143]. The cube copy, semantic fluency (substitution of letter F fluency), abstraction, serial-7 subtraction and naming in the Korean version of the MoCA positively correlated with education [81]. There are many cutoff scores reported according to the level of education of the studied population. In general, studies recruiting a higher proportion of low educated subjects recommend lower cutoff scores for the education correction.
7.5 Mild Cognitive Impairment (MCI) and Alzheimer’s Disease (AD)
The MoCA has been extensively studied as a screening tool for detection of MCI and Alzheimer disease (see Table 7.1). Sensitivity for MCI detection has been on average 85 % (range 67–96 %). Sensitivity to detect AD has been on average 94 % (range 88–100 %). Specificity defined as correctly identifying Normal Controls, was on average 76 % (range 19–98 %). Table 7.1 summarizes the MoCA validation in MCI and AD in diverse populations and languages. Variability in sensitivity and specificity is explainable by differences in selection criteria for normal controls, diagnostic criteria for MCI and AD, community or memory clinic setting, confirmation with neuropsychological battery, age and education levels, and possibly linguistic and cultural factors.
Table 7.1
MoCA studies in MCI and AD
Author (year) | Language | Subjects (n) | Education (years) | Condition to be screened | Cutoff point | Sn | Sp | PPV | NPV |
|---|---|---|---|---|---|---|---|---|---|
Nasreddine et al. (2005) [1] | English & French | 277 NC 90, aMCI 94, AD 93 | 11.86 | aMCI vs NC | 25/26a | 0.90 | 0.87 | 0.89 | 0.91 |
AD vs NC | 25/26a | 1.00 | 0.89 | 1.00 | |||||
Smith et al. (2007) [144] | English | 67 MCC 12, MCI 23, Dem (AD 18, VaD 13, PDD 1) | 12.1 | MCI vs MCC | 25/26a | 0.83 | 0.50 | – | – |
Dem vs MCC | 0.94 | 0.50 | |||||||
Ng Hoi Yee (2008) [142] | Cantonese- Hong Kong | 158 NC 74, aMCI 54, AD 30 | 5.37 | aMCI vs NC | 23/24a | 0.79 | 0.75 | 0.70 | 0.83 |
Lee et al. (2008) [81] | Korean | 196 NC 115, MCI 37, AD 44 | 8.03 | MCI vs NC | 22/23a | 0.89 | 0.84 | 0.65 | 0.96 |
AD vs NC | 22/23a | 0.98 | 0.84 | 0.70 | 0.99 | ||||
Luis et al. (2009) [128] | English | 118 NC 74, aMCI 24, AD 20 | 14.00 | aMCI vs NC | 23/24a | 0.96 | 0.95 | – | – |
Rahman et al. (2009) [145] | Arabic | 184 NC 90, MCI 94 | High school (49 %) | MCI vs NC | 25/26a | 0.92 | 0.86 | – | – |
Tangwong-chai et al. (2009) [146] | Thai | 120 NC 40, MCI 40, AD 40 | 10.59 | MCI vs NC | 24/25b | 0.80 | 0.80 | 0.80 | 0.80 |
AD vs NC | 21/22b | 1.00 | 0.98 | 0.98 | 1.00 | ||||
Duro et al. (2010) [147] | Portuguese | 212 MCI 82, AD 70, ODD 60 | ≤4 (n = 117) | MCI | 25/26a | Correctly identified 84.1 % f | |||
Dementia | 25/26a | Correctly identified 100 % f | |||||||
Fujiwara et al. (2010) [143] | Japanese | 96 NC 36, aMCI 30, AD 30 | 11.98 | aMCI vs NC | 25/26a | 0.93 | 0.89 | 0.88 | 0.94 |
AD vs NC | 25/26a | 1.00 | 0.89 | 0.88 | 1.00 | ||||
Selekler et al. (2010) [148] | Turkish | 205 NC 165, MCI 20, AD 20 | 11.59 | MCI/AD vs NC | 21/22a | 0.81 | 0.78 | 0.46 | 0.95 |
Larner (2012) [149] | English | 150 NC 85, MCI 29, Dem 36 | – | MCI/Dem vs NC | 25/26a | 0.97 | 0.60 | 0.65 | 0.96 |
Zhao et al. (2011) [141] | Chinese | 300 NC 150, aMCI 150 | 5–12 years (97 %) | aMCI vs NC | 23/24a | 0.77 | 0.90 | – | – |
Karunaratne et al. (2011) [150] | Sinhala | 98 NC 49, AD 49 | 10.34 | AD vs NC | 23/24a | 0.98 | 0.80 | – | – |
Damian et al. (2011) [151] | English | 135 Cognitively normal 89, Cognitively impaired 46 | 15.30 | Normal vs impaired | 23/24a | 0.87 | 0.75 | 0.38–0.54 | 0.95–0.97 |
Freitas et al. (2012) [152] | Portuguese | 360 NC 180, MCI 90, AD 90 | 6.38 | MCI vs NC | 21/22a | 0.81 | 0.77 | 0.78 | 0.80 |
AD vs NC | 16/17a | 0.88 | 0.98 | 0.98 | 0.89 | ||||
Chang et al. (2012) [153] | Chinese-Taiwan | 235 NC 97, very mild Dem 52, mild Dem 48, moderate Dem 38 | 7.90 | Very mild Dem vs NC | 22/23d | 0.83 | 0.88 | 0.81 | 0.89 |
Dong et al. (2012) [154] | Chinese-Singapore | 230 NC 33, MCI 61 (mdMCI 36, sMCI 25), Dem 136 | – | mdMCI vs NC/sMCI | 19/20 | 0.83 | 0.86 | 0.79 | 0.89 |
Tsai et al. (2012) [155] | Taiwan | 207 NC 38, MCI 71, AD 98 | – | MCI vs NC | 23/24 | 0.92 | 0.78 | – | – |
Yu et al. (2012) [156] | Chinese-Beijing | 1001 NC 865, MCI 115, Dem 21 | 10.10 | MCI vs NC | 21/22 | 0.69 | 0.64 | – | – |
Magierska et al. (2012) [157] | Polish | 114 NC 37, MCI 42, AD 35 | 9.59 | MCI vs NC | 24/25a | 0.81 | 0.54 | – | – |
AD vs MCI/NC | 19/20a | 0.86 | 0.82 | – | – | ||||
Hu et al. (2013) [135] | Chinese-Beijing | 302 NC 146, MCI 84, AD 72 | 9.40 | MCI vs NC | 26/27c | 0.92 | 0.85 | – | – |
AD vs NC | 25/26c | 0.92 | 0.96 | – | – | ||||
Memória et al. (2013) [158] | Brazilian | 82 NC 28, MCI 30, AD 24 | 11.41 | MCI vs NC | 24/25a | 0.81 | 0.77 | – | – |
AD vs NC | 21/22a | 0.91 | 1.00 | – | – | ||||
Ng et al. (2013) [159] | English | 212 NC 103, aMCI 49, AD 60 | 10.36 | Education > 10 | |||||
aMCI vs NC | 26/27a | 0.94 | 0.19 | – | – | ||||
aMCI vs AD | 24/25a | 0.90 | 0.70 | – | – | ||||
Education ≤ 10 | |||||||||
aMCI vs NC | 25/26a | 0.96 | 0.30 | – | – | ||||
aMCI vs AD | 23/24a | 0.85 | 0.81 | – | – | ||||
Zhou et al. (2014) [160] | Chinese | 172 NC 148, aMCI 24 | 0–6 | aMCI vs NC | 18/19 | 0.67 | 0.49 | – | – |
7–12 | aMCI vs NC | 22/23 | 0.89 | 0.64 | – | – | |||
0–12 | aMCI vs NC | 20/21 | 0.75 | 0.62 | – | – | |||
Goldstein et al. (2014) [161] | English | 81 NC 16, MCI 38, Dem 27 | – | MCI vs NC | 24/25a | 0.95 | 0.63 | – | – |
Dem vs NC | 22/23a | 0.96 | 0.88 | – | – | ||||
Yeung et al. (2014) [131] | Cantonese-Hong Kong | 272 NC 49, MCI 93, Dem 130 | 4.21 | MCI/Dem vs NC | 21/22b | 0.93 | 0.74 | – | – |
MCI vs NC | 21/22b | 0.83 | 0.74 | – | – | ||||
Dem vs NC | 18/19b | 0.92 | 0.92 | – | – | ||||
Chu et al. (2015) [132] | Cantonese-Chinese | 266 NC 115, aMCI 87, AD 64 | 5.62 | aMCI vs NC | 22/23e | 0.78 | 0.73 | – | – |
AD vs NC | 19/20e | 0.94 | 0.92 | – | – | ||||
Trzepacz et al. (2014) [162] | English | 618 NC 219, MCI 299, AD 100 | 16.19 | MCI vs AD | 16/17a | 0.92 | 0.58 | – | – |
MCI vs AD | 19/20a | 0.82 | 0.88 | – | – | ||||
Gil et al. (2015) [163] | Spanish | 193 NC 84, MCI 26, Dem 83 | 12.20 | MCI/Dem vs NC | 22/23a | 0.89 | 0.80 | 0.85 | 0.85 |
Lifshitz et al. (2012) [164] | Hebrew | 154 NC 80, MCI 74 | – | NC vs MCI | 25/26 | 0.95 | 0.76 | – | – |
7.6 The MoCA and the Memory Index Score (MIS)
In addition to the cognitive screening utility, the MoCA also provides the ability to predict AD conversion among patients with MCI. We newly devised the memory index score (MIS) which was calculated by adding the number of words remembered in free delayed recall, category-cued recall, and multiple choice-cued recall multiplied by 3,2 and 1, respectively, with a score ranging from 0 to 15 [112]. Individual patients meeting the Petersen’s MCI criteria (n =165) were recruited from our memory clinic and tested with the MoCA at MCI diagnosis. Within the average follow-up period of 18 months, 114 patients progressed to AD and 51 did not. Using a cutoff of <20/30 for MoCA total score and <7/15 for MIS, the AD conversion rate was 90.5 % for participants with MCI who were below the cutoff on both measures and was 52.8 % for those who were above the cutoff on both measures. This yields an annualized conversion rate of 60.3 % for the high-risk group and 35.2 % for the low-risk group. The mean time for AD conversion (n =114) was 17.5 months. We recommended the algorithm in Fig. 7.4 to predict conversion from MCI to AD with the MoCA total score and the MIS.
Fig. 7.4
The algorithm to predict conversion from MCI to AD with the MoCA total score and the MIS (Adapted from Julayanont et al. (2014) [112]
7.7 Vascular Cognitive Impairment (VCI)
7.7.1 Asymptomatic Cerebrovascular Disease Patients with Vascular Risk Factors
The MoCA has been shown to detect cognitive decline in asymptomatic subjects with hypertension alone, or thickening of the carotid artery wall, or multiple vascular risk factors [165, 166]. Cognitive decline was also detected in subjects with TIA or first ever stroke if they had more than two vascular risk factors or low cerebral perfusion on transcranial Doppler ultrasound [165, 166]. MoCA also correlated with the Framingham coronary and stroke risk scores [167].
Advanced internal carotid artery stenosis (>70 % occlusion) is also negatively correlated with MoCA but not MMSE scores in asymptomatic subjects [168, 169].
Subtle cognitive impairment among subjects from cardiac and diabetic/endocrine outpatient clinics of a tertiary-referral hospital were detected using the MoCA with sensitivity of 83–100 %, but with lower specificity of 50–52 % [170].
7.7.2 Symptomatic Cerebrovascular Disease
7.7.2.1 Cognitive Impairment Post-Stroke or TIA
The MoCA has been shown to detect cognitive impairment in 65 % of subjects 3 months post-stroke [171]. Thirty to 58 % of subjects with TIA or stroke who were considered normal on the MMSE scored below the normal cut-off on the MoCA ranging from 14 days to up to 5 years after the event [172, 173]. Table 7.2 presents a summary for MoCA studies on vascular cognitive impairment. A shortened version of the MoCA (miniMoCA) also provided a good validity in detection of vascular cognitive impairment after acute cerebrovascular events [191, 192]. Some factors may limit its applicability including high disability according to the National Institute of Health and Stroke Scale (NIHSS), left sided lesions, low education level and worse pre-morbid functional status [193, 194].
Table 7.2
Studies of the MoCA in vascular cognitive impairment
Author (year) Language | Objective of study | Subject (n) | Results |
|---|---|---|---|
Martinić-Popović et al. (2006, [165]; 2007, [166]) Croatian | To assess subtle cognitive decline in patients with first ever cerebrovascular disease (CVD) and in subjects without CVD symptoms but with CVD risk factors (CV-RF) | CV-RF (45) | The MoCA provided superior sensitivity than the MMSE in detection of MCI in CVD and CV-RF patients. |
Wong et al. (2008) [174] Cantonese-Hong Kong | To screen for subjects with white matter lesions (WML) | NC (33) WML (33) | At cutoff 21/22a, the MoCA provided sensitivity of 0.82 and specificity of 0.73 in detection of subjects with WML. |
Wong et al. (2009) [80] Cantonese-Hong Kong | To screen for subjects with small vessel disease (SVD) | NC (40) SVD (40) | At cutoff 21/22a, the MoCA provided sensitivity of 0.73 and specificity of 0.75 in detection of subjects with SVD. |
Martinić-Popović et al. (2009 [168]; 2011 [169]) Croatian | To assess MCI in patients with asymptomatic advanced internal carotid artery stenosis (ICS) | The MoCA proved to be a more sensitive tool than the MMSE for assessment of MCI in stroke-free patients with advanced ICS whose decline was most pronounced in the visuospatial/executive, delayed recall and abstraction subtest of the MoCA. | |
Dong et al. (2010) [172] English, Chinese, Malay | To assess cognitive impairment in acute post-stroke patients (mean 4.2 ± 2.4 days post-stroke) | Stable post-stroke patients (100) | 32 % of the normal MMSE (>24) patients were defined as cognitively impaired patients by the MoCA (≤21). The visuospatial/executive function, attention and delayed recall subtest of the MoCA provided a good discriminative power. |
Pendlebury et al. (2010) [173] English | To assess cognitive impairment in 6-month and 5-year post-stroke patients | Stable TIA/stroke patients (413) | 57 % of patients with normal MMSE (≥27) had abnormal MoCA (<26) which were associated with deficits in the delayed recall, abstraction, visuospatial/executive function, and sustained attention subtest of the MoCA. |
Godefroy et al. (2011) [175] French | To screen cognitive impairment after stroke | Infarct (88) Hemorrhage (7) | At cutoff 20/21b, the MoCA provided sensitivity of 0.67 and specificity of 0.90 in detection of cognitive impairment after stroke. |
McLennan et al. (2011) [170] English | To screen for MCI in patients with cerebrovascular disease (CVD) and vascular risk factors | CVD and risk factors (110) | At cutoff 23/24c, the MoCA provided sensitivity of 0.83–1.00 and specificity of 0.50–0.52 in detection of MCI in patients with CVD and vascular risk factors. |
You et al. (2011) [129] Cantonese | To screen for patients with mild to moderate vascular dementia (VaD) | NC (61) Mild VaD (30) Moderate VaD (40) | At cutoff 21/22, the MoCA provided sensitivity of 0.87 and specificity of 0.93 in detection of mild to moderate VaD. |
Cumming et al. (2011) [171] English | To assess the feasibility of the MoCA as a global cognitive screening tool in stroke trials. | 3-month post- stroke patients (294) | Of those surviving to 3 months, the MoCA was completed by 80 % of the patients. A majority of patients with stroke (65 %) were considered as cognitive impairment according to the MoCA cutoff scores <26. |
Harkness et al. (2011) [176] English | To assess MCI in patients with heart failure (HF) aged 65 years of more | HF (44) | More than 70 % of patients scored <26 on the MoCA, suggesting MCI, had significant deficits in the delayed recall, visuospatial/executive function, and language compared with the patients who scored ≥26. |
Athilingam et al. (2011) [177] English | To assess MCI in patients with heart failure (HF) aged 50 years of more | HF (90) | 54 % of participants scored ≤26 on the MoCA, whereas, only 2.2 % scored < 24 on the MMSE. Delayed recall, visuospatial/executive function and language subtest of the MoCA were impaired in more than 60 % of patients. |
Kasai et al. (2012) [178] Japanese | To screen for MCI (CDR 0.5) in patients with very mild small vessel disease (SVD) | NC (164) Very mild SVD (37) | At cutoff 18/19c, the MoCA provided sensitivity of 0.78 and specificity of 0.74 in detection of MCI in patients with very mild SVD. |
Wong et al. (2012) [179] Chinese-Hong Kong | To assess the cognitive impairment at 3 months after aneurysmal subarachnoid hemorrhage (aSAH) | aSAH (90) | Cognitive impairment (MoCA <26) was determined in 73 % of patients at 3 months. The MoCA correlated with functional outcome at 3 months. |
Schweizer et al. (2012) [180] English | To assess how the MoCA relates to cognitive impairment and return to work after aSAH | aSAH (32) | The MoCA was more sensitive than the MMSE in detection of cognitive impairment after aSAH. Naming and abstraction of the MoCA were associated with return to work. |
Wu et al. (2013) [181] Chinese | To screen for patients with vascular cognitive impairment (VCI) without dementia | NC (111) VCI without dementia (95) | At cutoff 22/23c, the MoCA provided sensitivity of 0.65 and specificity of 0.79 in detection of VCI without dementia. |
Wong et al. (2013) [182] Cantonese-Hong Kong | To screen for patients with traumatic intracranial hemorrhage (tICH) | NC (40) tICH (48) | At cutoff 25/26a, the MoCA provided sensitivity of 0.75 and specificity of 0.48 in detection of patients with tICH. |
Wong et al. (2013) [183] Cantonese-Hong Kong | To screen for patients with cognitive impairment after aneurysmal subarachnoid hemorrhage (aSAH) | aSAH (74, at 2–4 weeks) aSAH (80, at 1 year) | At cutoff 17/18a, the MoCA provided sensitivity of 0.75 and specificity of 0.95 in detection of patients with aSAH at 2–4 weeks. At cutoff 21/22a, the MoCA provided sensitivity of 1.00 and specificity of 0.75 in detection of patients with aSAH at 1 year. |
Tu et al. (2013) [184] Chinese- Changsha | To screen for vascular cognitive impairment (VCI) after ischemic stroke | NC (132) VCI (207) | At cutoff 23/24a, the MoCA provided sensitivity of 0.75 and specificity of 0.99 in detection of patients with VCI after ischemic stroke. |
Pendlebury et al. (2013) [185] English | To screen for MCI at 1 year after CVA | CVA (91) | At cutoff <17/22, the MoCA provided sensitivity of 0.83 and specificity of 0.70 in detection of patients with MCI at 1 year after stroke. |
Ihara et al. (2013) Japanese [186] | To assess the suitability of the MoCA in detecting VCI in patients with extensive leukoaraiosis on MRI | Extensive leukoaraiosis on MRI (12) | The MoCA was more sensitive than the MMSE in detecting VCI in patients with extensive leukoaraiosis on MRI. |
Cumming et al. (2013) [187] Swedish | To screen for vascular cognitive impairment (VCI) at 3 months after CVA | CVA (60) | At cutoff 23/24c, the MoCA provided sensitivity of 0.92 and specificity of 0.67 in detection of patients with VCI at 3 month after stroke. |
Ihara et al. (2013) Japanese [188] | To correlate the MoCA with daily physical activity in patients with subcortical leukoariaosis | Extensive leukoaraiosis on MRI (10) | The MoCA total score and its visuospatial/executive subscores correlated with the physical activity parameters. |
Webb et al. (2014) English [189] | To determine relationships between the MoCA and hypertension/hypertensive arteriopathy | TIA or minor stroke (492) | The MoCA provided stronger relationship to the hypertensive arteriopathy than the MMSE. The MoCA was more sensitive to detect cognitive impairment than the MMSE. |
Pasi et al. (2015) Italian [190] | To assess the association between white matter microstructural damage measured by diffusion tensor imaging and the MoCA score. | leukoaraiosis on MRI with MCI (76) | In patients with VCI secondary to small vessel disease, the MoCA performance more related to microstructural damage measured by diffusion tensor imaging than the MMSE. |
7.7.2.2 Heart Failure
Fifty-four to seventy percent of non-demented community-dwelling adults with heart failure (HF) (ejection fraction 37–40 %) had low cognitive scores on the MoCA (<26) [149, 150]. In acute setting during hospitalization, 41 % of patients scored lower than 26 points in the MoCA [195]. Reduction in ejection fraction and various associated vascular risk factors such as hypertension, dyslipidemia or diabetes mellitus may contribute to chronic reduction of cerebral blood flow in HF patients [196–198].
7.7.2.3 Chronic Atrial Fibrillation
The MoCA identified MCI in 65 % of older hospitalized patients with chronic atrial fibrillation. Executive, visuospatial and memory function were the most notable cognitive deficits. The predictors of MCI in these patients included low education level, high CHA2DS2-VASc score and prescribed digoxin [199].
7.7.2.4 Sub-optimal Self-Care and Functional Dependency
MoCA identified MCI in patients with heart failure that had suboptimal self-care behaviors [200]. HF patients with the MoCA score <26 had lower score on the self-care management than the patients with the MoCA ≥ 26 [201].
Using the MoCA as a cognitive assessment instrument, the self-rated version of the instrumental activities of daily living (IADL) scale was administered to evaluate functional dependence among 219 non-demented patients with cardiovascular diseases and risk factors [202]. MCI was diagnosed when MoCA was less than 23/30. Less dependence was associated with higher MoCA scores, and a person who scored in the MCI range was 7.7 times more likely to report need for assistance with one or more activity of daily living. This study indicated that subtle cognitive impairment was an independent predictor of functional status in patients with cardiovascular disease [202].
7.7.2.5 Subcortical Ischemic Vascular Dementia (SIVD)
7.7.2.6 Monitoring of Treatment
Cognitive outcomes after undergoing carotid endarterectomy (CEA) in severe unilateral internal carotid artery stenosis were studied using MoCA and MMSE as primary outcome measures. Symptomatic carotid stenosis (SCS) and asymptomatic severe carotid stenosis ≥60 % (ACS) patients were compared with age- and sex-matched control subjects who underwent laparoscopic cholecystectomy (LC). At baseline, the SCS group, but not the ACS, was significantly more impaired on the MoCA and MMSE total scores compared with the LC group. Postoperatively, only the SCS patients had significant improvement on both tests when comparing pre-operative and 12-month post-operative performance [207].
7.8 Parkinson’s Disease (PD)
The prevalence of dementia in PD is between 20 and 40 % [208]. The early cognitive changes are mediated by fronto-striatal disconnection, such as executive function and attention [209]. Single domain impairment is found more frequently than multiple domain deficits in early stages [209, 210]. Progression of PD affects other cognitive domains such as memory [208, 211]. The association between cognitive impairment and cholinergic denervation and frontostriatal dopaminergic deficits among PD and PD with dementia (PDD) has been demonstrated by neuroimaging studies [212, 213]. Detection of cognitive impairment in PD is clinically useful as it predicts the conversion to PDD [211], contributes to caregiver’s distress [214], and guides timing to initiate cognitive enhancing treatment [215].
The MoCA has an adequate sensitivity as a screening tool for detection of PD-MCI or PDD in a clinical setting (see Table 7.3), based on diagnostic criteria and neuropsychological test batteries [219, 220]. Half of PD patients with normal age and education-adjusted MMSE scores were cognitively impaired according to the recommended MoCA cutoff (25/26) [218, 229] as it lacks a ceiling [216, 217, 219]. Sensitivity and specificity for PDD were 70–82 % and 75–95 % respectively. Sensitivity and specificity for PD-MCI are 83–93 % and 53–75 % respectively [219, 220].
Table 7.3
MoCA in Parkinson’s disease (PD)
First author (year) Language | Objective of study | Subject (n) | Results |
|---|---|---|---|
Measurement | |||
Gill et al. (2008) [216] English | To establish the cognitive screening characteristics of the MoCA in PD patients | PD (n = 38) | There was no ceiling effect of the MoCA. The test–retest intraclass correlation coefficient was 0.79. The inter-rater intraclass correlation coefficient was 0.81. The correlation coefficient between the MoCA and a neuropsychological battery was 0.72. |
MoCA & MMSE | |||
Zadikoff et al. (2008) [217] English | To establish the MoCA and MMSE scores characteristics in PD | PD (n = 88) | The MoCA showed less prone to ceiling effect and identify more MCI in PD patients than the MMSE. |
MoCA & MMSE | |||
Nazem et al. (2009) [218] English | To examine the MoCA performance in PD patients with normal global cognition according to the MMSE score | PD (n = 100) | 52 % of subjects with normal MMSE scores had cognitive impairment according to their MoCA scores (<26). The impaired patients scored worse than unimpaired patients on visuospatial/executive, naming, attention, language and delayed recall subtest of the MoCA. |
MoCA & MMSE | |||
Hoops et al. (2009) [219] English | To assess the validity of the MoCA in detection of MCI and dementia among PD patients | PD-N (n = 92), PD-MCI (n = 23), PDD (n = 17) | At cutoff 26/27a, the MoCA provided sensitivity of 0.83 and specificity of 0.53 in detection of PD-MCI. At cutoff 24/25a, the MoCA provided sensitivity of 0.82 and specificity of 0.75 in detection PDD At cutoff 26/27a, the MoCA provided sensitivity of 0.90 and specificity of 0.53 in detection of PD with cognitive impairment (PD-MCI & PDD) |
MoCA | |||
Dalrymple-Alford et al. (2010) [220] English | To assess the validity of the MoCA in detection of MCI and dementia among PD patients | PD-N (n = 72), PD-MCI (n = 21), PDD (n = 21) | At cutoff 20/21a, the MoCA provided sensitivity of 0.81 and specificity of 0.95 in detection of PDD from PD-MCI/PD-N. At cutoff 25/26a, the MoCA provided sensitivity of 0.90 and specificity of 0.75 in detection PD-MCI |
MoCA | |||
Luo et al. (2010) [221] Chinese | To define and compare the cognitive profiles and clinical features of PD patients with slow or rapid cognitive deterioration rate (CDR),with normal controls (NC) | PD(n = 73) NC (n = 41) | The total scores and subscores for visuospatial abilities, verbal fluency and delayed recall of the MoCA were significantly lower in the PD than NC. The rapid CDR group (MoCA decline >1 point/year) was older, later age at onset, faster movement deteriorated and more impaired in CDT, attention, verbal fluency and abstraction subtest than the slow CDR group. |
MoCA | |||
Robben et al. (2010) [222] Dutch | To pilot a three-step cognitive diagnostic model for patients with PD dementia (PDD) | PDD (n = 15) PD no dementia (n = 26) | It is efficient and feasible to use the three consecutive diagnostic steps for PDD as the following: Screening questionnaire → if + → the MoCA or FAB or ACE-R as screening tools → if + → a detailed NPE as diagnostic tools. |
Screening questionnaire; MoCA/FAB/ACE-R; Detailed NPE | |||
Ling et al. (2013) [223] Chinese | To assess the validity of the MoCA Chinese in detection of dementia among PD patients | PD-N (n = 381) PDD (n = 235) | At cutoff 22/23, the MoCA provided sensitivity of 0.70 and specificity of 0.77 in detection PDD |
MoCA-Chinese | |||
Kandiah et al. (2014) [224] English | To assess the validity of the MoCA in detection of PD-MCI and prediction of cognitive decline | PD-N (n = 61) PD-MCI (n = 34) | At cutoff 26/27, the MoCA provided sensitivity of 0.93 in the diagnosis of PD-MCI. The score ≤ 26 increases the risk of cognitive decline in 2 years |
MoCA | |||
Ozdilek et al. (2014) [225] Turkish | To assess the validity of the MoCA-Turkish in screening for cognitive impairment in PD | PD (n = 50) NC (n = 50) | At cutoff 20/21, the MoCA provided sensitivity of 0.59 and specificity of 0.89 in detection cognitive impairment in PD |
MoCA-Turkish | |||
Van Steenoven et al. (2014) [226] English | To provide the conversion algorithm between the MoCA and MMSE in PD patients | PD (n = 360) | The score conversion between the MoCA, MMSE and DRS-2 were proposed. |
MoCA, MMSE & DRS-2 | |||
Krishnan et al. (2015) [227] Malayalam | To assess the validity of the MoCA-Malayalam in screening for cognitive impairment in PD | PD (n = 70) NC (n = 60) | The MoCA Malayalam had good internal consistency and test-retest reliability in patients with PD. The scores correlated with MMSE and ACE. |
MoCA-Malayalam, MMSE & ACE | |||
Chung et al. (2015) [228] Korean | To compare the MoCA performance in PD with and without visual hallucinations | PD-VH (n = 26) PD-NH (n = 32) | The language domain of MoCA-K was sensitive to cognitive deficit in PD-VH patients. |
MoCA-Korean |
Baseline MoCA scores predicted the rate of cognitive deterioration among PD patients. The group of rapid decliners had lower scores on total MoCA score, clock drawing, attention, verbal fluency and abstraction subtest when compared with slow decliners [221].
MoCA was shown to have good reliability in this population. The test–retest correlation coefficient is 0.79, and the inter-rater correlation coefficient is 0.81 [216]. The superiority of the MoCA compared to the MMSE is probably explained by its more sensitive testing of executive, visuospatial, and attention domains which are frequently impaired in PD. Some of MoCA’s limitations are that there are no studies yet regarding its sensitivity to detect of cognitive change over time or after treatment [230] and MoCA contains items that require fine motor movement such as trail making test, cube copy and clock drawing (5/30 points), which can impact on the results when administering the test to patients with severe motor symptoms.
7.9 Huntington’s Disease
Subtle cognitive impairment has been shown to precede motor manifestations of Huntington’s disease (HD) [231–234]. While global cognitive function is relatively preserved in asymptomatic carriers (AC) of HD mutation, attention, psychomotor speed, working memory, verbal memory and executive function are often impaired early [232–234]. These impaired functions are caused by abnormal fronto-striatal circuitry as shown in morphological and functional studies [235, 236].
Two studies compared the ability of the MoCA and the MMSE in detection of cognitive impairment in HD patients with mild to moderate motor symptom. Compared with the MMSE, the MoCA achieved higher sensitivity (MoCA 97.4 %; MMSE 84.6 %), however, comparable but not impressive specificity (MoCA 30.1 %; MMSE 31.5 %) in discriminating HD from normal subjects [237, 238]. The limitation for interpreting these results is that the available studies did not use standardized neuropsychological evaluation as a gold standard for classifying cognitive function in HD. A subsequent study reported even better results for the MoCA in detection of cognitive dysfunction in HD patients at the cut off <26 points with sensitivity of 94 % and specificity of 84 % [239]. The MoCA is a useful instrument to detect cognitive changes from mild to severe stages of HD patients [240].
The superiority of the MoCA compared to the MMSE in this population is explained by more emphasis in the MoCA on cognitive domains frequently impaired in early HD. Clock drawing, trail making, cube copy, abstraction, and letter F fluency in the MoCA increase its ability to detect executive and visuo-spatial dysfunction. Five word delayed recall, digit span, letter tapping/vigilance test in the MoCA provide a better assessment of memory and attention.
7.10 Brain Tumors
MoCA detected cognitive impairment among patients with brain metastases in 70 % of patients who performed the MMSE in the normal range (≥26/30). Patients had abnormal delayed recall (90 %) or language (90 %) followed by deficits in visuospatial/executive function (60 %) and the other sub-domains [241].
Detection of MCI among patients with primary and metastatic brain tumors using a standardized neuropsychological assessment as a gold standard has also shown the superiority of the MoCA compared to the MMSE in sensitivity but at the expense of lower specificity. MoCA sensitivities and specificities were 62 % and 56 % respectively, whereas MMSE sensitivities and specificities were 19 % and 94 % respectively. Visuospatial/executive function items of the MoCA correlated with patients’ perceived quality of life (ability to work, sleep, enjoy life, enjoy regular activities and accept their illness) [242].
Cognitive function is one of the survival prognostic factors and correlates with tumor volume in metastatic brain cancer [243, 244]. The survival prognostic value of the MoCA was studied among patients with brain metastases [245]. After dichotomizing MoCA scores into two groups based on average scores (≥22 and <22), below-average MoCA scores were predictive of worse median overall survival (OS) compared with above-average group (6.3 versus 50.0 weeks). Stratified MoCA scores were also predictive of median OS, as the median OS of patients who performed the MoCA with scores in the range of >26, 22–26, and <22, were 61.7, 30.9 and 6.3 weeks, respectively. MoCA scores were superior to the MMSE scores as a prognostic marker. Although, the MoCA scores correlated with the median OS, it is essential to clarify that cognitive impairment does not directly result in decreased survival. Lower MoCA scores may represent other unmeasured confounders such as the extent of disease, location of tumor or previous treatment [245].
7.11 Systemic Lupus Erythematosus (SLE)
Cognitive dysfunction is a common symptom of SLE-associated neuropsychiatric manifestation. It can occur independently of clinically overt neuropsychiatric SLE [246–252]. Magnetic resonance spectroscopy reveals the association between metabolic change in white matter of non-neuropsychiatric SLE (non-NSLE) patients and cognitive impairment [247, 253]. Early cognitive impairments in non-NSLE patients are verbal fluency, digit symbol substitution and attention [252, 254]. Some investigators suggested that the pattern of cognitive decline in non-NSLE is mostly classified as subcortical brain disease since the psychomotor and mental tracking impairment are observed early [255]. The domains which are subsequently impaired in patients who develop neuropsychiatric SLE (NSLE) symptoms are memory, psychomotor speed, reasoning and complex attention [254, 256].
The MoCA was validated among SLE patients in hospital-based recruitment, using the Automated Neuropsychologic Assessment Metrics (ANAM) as a gold standard. At the standard cutoff score <26/30, the MoCA provided good sensitivity (83 %), specificity (73 %) and overall accuracy (75 %) in detection of cognitive impairment [257].
7.12 Substance Use Disorders
The validity of the MoCA to detect cognitive impairment in subjects with non-nicotine substance dependence disorders according to the DSM-IV criteria was established by using the Neuropsychological Assessment Battery-Screening Module (NAB-SM) as a gold standard to define cognitively impaired participants. The NAB-SM is composed of 5 domains: attention, language, memory, visuospatial, and executive function. The participants were composed of alcohol dependence (65 %; n =39), dependence on opioids (32 %; n =19), cocaine (17 %; n =10), cannabis (12 %; n =7), benzodiazepine (10 %; n =6), and amphetamine (8 %; n =5). At the optimal cutoff point of 25/26, the MoCA provided acceptable sensitivity and specificity of 83 % and 73 %, respectively, with good patient acceptability [258].
7.13 Idiopathic Rapid Eye Movement Sleep Behavior Disorder (Idiopathic RBD)
RBD is characterized by the intermittent loss of REM sleep electromyographic atonia resulting in motor activity associated with dream mentation. Approximately 60 % of cases are idiopathic [259]. MCI is found in 50 % of idiopathic RBD and most of them are single domain MCI with executive dysfunction and attention impairment [260]. Visuospatial construction and visuospatial learning may be impaired in neuropsychologically asymptomatic idiopathic RBD patients who have normal brain MRI [261]. Subtle cognitive changes in idiopathic RBD may reflect the early stage of neurodegenerative diseases [261] as some studies reported an association between idiopathic RBD and subsequent development of Parkinson’s disease (PD), Lewy body dementia (LBD) and multiple system atrophy [262–264]. Moreover, cognitive changes in idiopathic RBD are similar (visuoconstructional and visuospatial dysfunction) to LBD [265] and to early PD (executive dysfunction) [209].
The MCI screening property of the MoCA was validated among 38 idiopathic RBD patients, based on neuropsychological assessment as a gold standard. At the original cutoff point of 25/26, the MoCA had sensitivity for cognitive impairment of 76 % and specificity of 85 % with an accuracy of 79 %. However, for screening purposes, the higher cutoff (26/27) may be applied as it increases sensitivity to 88 %, at the expense of reduced specificity (61 %). The demanding visuospatial/executive function subtests of the MoCA makes it sensitive for detection of mild cognitive impairment in idiopathic RBD patients who are impaired early in these domains [266].
7.14 Chronic Obstructive Pulmonary Disease (COPD)
Cognitive impairment is a frequent feature of COPD. MCI was reported in 36–63 % of patients with COPD [267, 268]. At the cutoff <26/30, the MoCA provided 81 % sensitivity and 72 % specificity in detecting cognitive impairment among patients with moderate to severe COPD [267]. Patients with COPD with acute exacerbation had significantly lower MoCA scores than patients with stable COPD and normal controls [269]. In patients with acute COPD exacerbation who were hospitalized, cognitive impairment was identified in 57 % which related to worse health status and longer length of stay [270].
7.15 Obstructive Sleep Apnea (OSA)
In a recent study by Chen et al. [271], the MoCA was administered to 394 obstructive sleep apnea (OSA) patients categorized into four groups according to OSA severity based on the total number of apnea and hypopnea per hour of sleep (AHI), measured by polysomnography. The groups were composed of primary snoring (AHI <5 events/h), mild OSA (AHI 5–20 events/h), moderate OSA (AHI 21–40 events/h) and severe OSA (AHI >40 events/h). The total MoCA scores progressively decreased as the severity of OSA increased. The scores of moderate-to-severe OSA groups were significantly lower than the scores of the primary snoring and mild OSA groups. Furthermore, defining MCI with a cutoff of 25/26, the moderate-to-severe OSA groups were more classified as MCI than the other groups. Domains that were significantly impaired in the severe OSA group, compared to the primary snoring group, were delayed recall, visuospatial/executive function, and attention/concentration. Even though the mild OSA group performed similarly to the primary snoring group on total MoCA scores, impairment in the visuospatial/executive function and delayed recall domains was more prominent. Moreover, MoCA scores correlated with oxygen saturation levels [271]. A subsequent study reported that at the cut off <26 point, the sensitivity and specificity to differentiate between normal subjects and non-normal subjects were 54 % and 70 % respectively [272].
7.16 Risk of Falls
Liu-Ambrose and colleagues used the MoCA to classify 158 community-dwelling women as MCI or cognitively intact by the cutoff point of 25/26 [273]. The short form of Physiologic Profile Assessment (PPA) was used to assess the fall risk profile. In the PPA, the postural sway, quadriceps femoris muscle strength, hand reaction time, proprioception and edge contrast sensitivity are evaluated. Participants with MCI had higher global physiological risk of falling and greater postural sway compared with the counterparts. However, the other four PPA components were not significantly different between the two groups. This study suggested that screening for MCI using the MoCA is valuable in preventing falls in the elderly.
In another study, forty-seven patients were classified into faller and non-faller groups. The non-faller group performed significantly better than the faller group in physical activities (timed Up-and-Go, the 10 min walk test and the 6 min walked test) and cognitive functions measured by the MoCA. The study suggested that in order to decrease the risk of falls, physical activity and cognitive evaluation are recommended in community-dwelling stroke patients [274].
7.17 Rehabilitation Outcome
The MoCA has been shown to be more sensitive than the MMSE for detection of MCI in an inpatient rehabilitation setting [275]. The association between cognitive status measured by the MoCA and rehabilitation outcomes was studied among 47 patients admitted to a geriatric rehabilitation inpatient service [276]. Patients had an orthopedic injury (62 %), neurological condition (19 %), medically complex condition (11 %) and cardiac diseases (4 %). MoCA had good sensitivity (80 %), but poor specificity (30 %), at the cutoff scores 25/26 to predict successful rehabilitation outcome. The patients who reached the successful rehabilitation criteria tended to have higher MoCA scores at admission than the patients who did not achieve the rehabilitation goal. Many studies have reported the negative effect of cognitive impairment on the rehabilitation outcomes [276–279].
In a short term rehabilitation program in post-stroke patients (median time post-stroke 8.5 days) who had MCI, the MoCA had a significant association with discharge functional status. The discharge functional status was measured by the motor subscale of Functional Independence Measures (mFIM) and motor relative functional efficacy taking the individual’s potential for improvement into account [280]. The visuospatial/executive domain of the MoCA was the strongest predictor of functional status and improvement. This domain was previously shown as an independent predictor of post-stroke long term functional outcome [281].
7.18 MoCA in Epilepsy
A cross-sectional study examined the MoCA performance in cryptogenic epileptic patients aged more than 15 years with normal global cognition according to the Mini-Mental State Examination (MMSE) score. The mean MoCA score was 22.44 (±4.32). In spite of a normal MMSE score, which was an inclusion criterion, cognitive impairment was detected in 60 % of patients based on the MoCA score. The variable that correlated with a higher risk of cognitive impairment was the number of antiepileptic drugs (polytherapy: OR 2.71; 95 % CI 1.03–7.15). No neuropsychological batteries were used for comparison [282].
7.19 Human Immunodeficiency Virus (HIV) Infection
Cognitive impairment in HIV patients may result in medication compliance problems. The ability of the MoCA to detect cognitive impairment in patients with HIV infection has been studied. At the cut off <26 points, the sensitivity and specificity were 51–85 % and 40–77 % respectively [283–285]. There was global cognitive decline in HIV patients, in particular visuospatial, executive, attention and language functions were impaired [286]. Current CD4+ level and depression severity is a strong predictor of the MoCA score among HIV patients [287]. Because of its low specificity the MoCA may be useful as a first screening tool for identifying HIV patients who may need further formal neuropsychological testing.
7.20 Miscellaneous Conditions
The MoCA has been studied in many other conditions including frontotemporal dementia, multiple sclerosis, traumatic brain injury, diabetes, Korsakoff syndrome, chronic hemodialysis, schizophrenia, macular degeneration, severe mental illness, ALS, psychiatry inpatients, and driving, studies which are summarized in the Table 7.4.
Table 7.4
The MoCA in other conditions
First author (year) Language | Objective of study | Subject (n) | Results |
|---|---|---|---|
Measurement | |||
Freitas et al. (2012) [288] | To assess the validity of the MoCA in behavioral-variant frontotemporal dementia (bv-FTD) | bv-FTD (50) NC (50) | At cutoff <17, the MoCA provided sensitivity of 78 % and specificity of 98 % in detection of bv-FTD from NC which is better than the MMSE (sensitivity 58 %, specificity 88 %). |
MoCA and MMSE | |||
Kaur et al. (2013) [289] | To assess the validity of the short MoCA in multiple sclerosis (MS) | MS (50), NC (50) | At cutoff 10/11 from total 12 points, the short MoCA provided sensitivity of 97 % and specificity of 90 % in detection of cognitive impairment in MS patients.
Related posts: The General Practitioner Assessment of Cognition (GPCOG)
The IQCODE: Using Informant Reports to Assess Cognitive Change in the Clinic and in Older Individuals Living in the Community
 Six-Item Cognitive Impairment Test (6CIT)
 DemTect
 Assessment of the Utility of Cognitive Screening Instruments
 The Quick Mild Cognitive Impairment Screen (Qmci)
 Stay updated, free articles. Join our Telegram channel
Full access? Get Clinical Tree
Get Clinical Tree app for offline access
|