The Montreal Cognitive Assessment: Is It Suitable for Identifying Mild Cognitive Impairment in Parkinson's Disease?

ABSTRACT

Background

Administering an abbreviated global cognitive test, such as the Montreal Cognitive Assessment (MoCA), is necessary for the recommended first‐level diagnostic criteria for mild cognitive impairment (MCI) in Parkinson's disease (PD). Level II requires administering cognitive functioning neuropsychological tests. The MoCA's suitability for identifying PD‐MCI is questionable and, despite the importance of cognitive deficits reflected through daily functioning in identifying PD‐MCI, knowledge about it is scarce.

Objectives

To explore neuropsychological test scores of patients with PD who were categorized based on their MoCA scores and to analyze correlations between this categorization and patients’ self‐reports about daily functional‐related cognitive abilities.

Methods

A total of 78 patients aged 42 to 78 years participated: 46 with low MoCA scores (22–25) and 32 with high MoCA scores (26–30). Medical assessments and level II neuropsychological assessment tools were administered along with standardized self‐report questionnaires about daily functioning that reflects patients’ cognitive abilities.

Results

A high percentage of the low MoCA group obtained neuropsychological test scores within the normal range; a notable number in the high MoCA group were identified with MCI‐level scores on various neuropsychological tests. Suspected PD‐MCI according to the level I criteria did not correspond well with the level II criteria. Positive correlations were found among the 3 self‐report questionnaires.

Conclusions

These results support the ongoing discussion of the complexity of capturing PD‐MCI. Considering the neuropsychological tests results, assessments that reflect cognitive encounters in real life daily confrontations are warranted among people diagnosed with PD who are at risk for cognitive decline.

Parkinson's disease (PD) is the second most common progressive neurodegenerative disorder, affecting more than 1% of the population aged older than 60 years. Impairment in cognitive functioning without dementia is present in about 25% of newly diagnosed patients with PD and usually worsens as the disease progresses. ¹ Understanding the characteristics of mild cognitive impairment (MCI) in PD (PD‐MCI) and identifying it in its earliest phase may be important for early intervention and the preservation of patients’ physical and emotional health and well‐being. ² The question then arises as to how we can practically accomplish this in a real‐world clinical setting while considering limitations of time and other resources. ³

In 2012, the Movement Disorder Society (MDS) proposed the following diagnostic criteria for PD‐MCI: ⁴ (1) PD diagnosis based on the UK PD Brain Bank criteria ⁵ ; (2) gradual decline in cognitive ability in the context of established PD, as reported by the patient or an informant or observed by the clinician; (3) cognitive deficits on formal neuropsychological tests or scales of global cognitive abilities; and (4) cognitive deficits that do not significantly interfere with functional independence, although subtle difficulties on complex functional tasks may be present.

Consequently, the guidelines for clinical cognitive assessment of PD‐MCI consist of a 2‐level testing scheme. The first level includes an abbreviated global cognitive performance test, such as the Mini‐Mental State Examination ⁶ or the Montreal Cognitive Assessment (MoCA), ⁷ with preference in the literature for using the MoCA for a PD‐MCI diagnosis. ⁸ , ⁹ Evidence for using scales besides the Mini‐Mental State Examination for this purpose indeed exists in the literature (eg, ref. 10). These include the Dementia Rating Scale, ¹¹ the Parkinson's Disease–Cognitive Rating Scale, ¹² and the Scales for Outcomes in PD–Cognition. ¹³ However, because of its practicality—given that it takes only 10 minutes to administer—the MoCA is widely used in research and in clinical settings as a first screening tool for PD‐MCI (eg, refs. 13, 14, 15). Furthermore, the MoCA has a valid Hebrew translation that was implemented in the current study. ¹⁵ , ¹⁶

However, concerns have been raised regarding the MoCA's sensitivity and specificity for level I identification of PD‐MCI related to the significant age effect, gender, and education variables and cutoff scores. ¹⁵ , ¹⁷ , ¹⁸ , ¹⁹ As such, accomplishing the initial PD‐MCI identification level is not a trivial task. Level II of the PD‐MCI evaluation ⁴ includes a comprehensive neuropsychological assessment. That assessment can provide diagnosis subtyping and guidelines for defining MCI when impairment is found in at least 2 neuropsychological tests based on established cutoff scores. ³ , ²⁰ , ²² Many centers have implemented and analyzed the discriminant validity of the neuropsychological battery. ¹⁴ However, the fourth MDS diagnostic criteria for PD‐MCI ⁴ (ie, functional difficulties while performing complex daily tasks) remains ambiguous and is still debated, and standardized tools for addressing this criteria are scarce. ¹⁴ , ¹⁵ , ¹⁶ , ¹⁷ , ¹⁸ , ¹⁹ , ²¹ , ²³ , ²⁴ , ²⁵ , ²⁷ , ²⁸ , ³¹ , ³² , ³³ , ³⁴ , ³⁷ , ³⁸ , ³⁹ , ⁴⁰ , ⁴¹ , ⁴² , ⁴³ , ⁴⁴ , ⁴⁶ , ⁴⁷ Moreover, little is known about the day‐to‐day functional limitations of patients with PD resulting from their MCI. ²³ Thus, it is unclear whether the MoCA is indeed the most suitable tool to evaluate global cognitive performance indicating mild cognitive decline and, if so, what should be the cutoff score differentiating between normal cognitive state and PD‐MCI.

To support the use of the MoCA for PD‐MCI, we would expect a high correlation among level I (MoCA) scores, level II (neuropsychological tests) scores, and self‐reported cognitive functional performance (MDS criterion 4). To explore these issues, we focused on nondemented early to mid‐phase, fully independent, patients with PD. We assessed the relationship between the categorization of normal versus MCI based on MoCA cutoff scores and the neuropsychological battery. We also explored the correlation between the MoCA scores and 3 additional self‐report cognitive functional scales: The Cognitive Failures Questionnaire (CFQ), ²⁴ the Parkinson's Disease Cognitive Functional Rating Scale (PD‐CFRS), ²² and a newly composed MDS‐UPDRS–derived cognitive functional items measure (UPDRS‐CF).

Methods

Participants

Patients with PD were recruited at the Sheba Movement Disorders Institute. Inclusion criteria were age 40 to 80 years; living in Israel for at least the past 10 years; fluent in Hebrew speaking, reading, and writing; and normal or corrected normal vision and hearing. Patients met PD clinical diagnostic criteria of the MDS ²⁵ and were Hoehn and Yahr stage 1 or 2, ²⁶ functionally independent, and lived in their private homes or in assisted‐living facilities. Furthermore, a movement disorders neurologist assessed PD‐related symptoms and signs using the MDS‐UPDRS. ²⁷ Patients with active signs of depression or a diagnosis of dementia were not included. This exclusion was based on an interview and a score of 3 or more on 2 MDS‐UPDRS items: the depressed mood question (number 1.3) and the cognitive impairment question (number 1.1). In addition, the participants were to score 18 or below on the Beck Depression Inventory ²⁸ and 22 or higher on the MoCA validated Hebrew version for inclusion. ⁷ , ⁹ , ²⁹ , ³⁰

The participants were further divided into 2 groups based on their MoCA scores: 22 to 25, the low MoCA group (suggesting PD‐MCI), ³¹ and 26 to 30, the high MoCA group. Exclusion criteria were significant systemic or other diseases, such as cancer or congestive heart failure. Furthermore, patients with any other neurological diseases, such as epilepsy, previous stroke or transient ischemic attack, brain tumor, brain surgery or brain trauma in the past, or premorbid psychiatric disease, or who were being treated with antipsychotic or anticholinergic medications were excluded.

Measures

Neuropsychological Cognitive Assessments

Attention was assessed with the Trail Making Test, Part A (TMT‐A) ³² and the Wechsler Adult Intelligence Scale (WAIS)–III digit span forward. ³³

Executive function/working memory was tested with the Trail Making Test, Part B (TMT‐B) ³² and the WAIS‐III digit span backward. ³³

Language ability was tested with the WAIS‐III vocabulary test. ³³

Visual spatial functioning was measured with the Rey Osterrieth Complex Figure Test (RCFT) copying ²⁸ , ²⁹ and the WAIS‐III block design. ²⁷

Memory was assessed with the RCFT recall ³⁴ , ³⁵ and the Wechsler Memory Scale ³⁶ logical memory, including immediate and delayed verbal recall of stories.

Self‐Report Cognitive Functional Questionnaires

Patient‐reported dysfunction in nonmotor and motor daily experiences that are associated with cognitive procedure were retrieved from Parts I and II of the MDS‐UPDRS. A total of 7 items were chosen for the current study to create the UPDRS‐CF ³⁷ score. These included the following: 1.1 cognitive impairment, 2.1 speech, 2.4 eating, 2.5 dressing, 2.6 hygiene, 2.7 handwriting, and 2.8 doing hobbies. The UPDRS‐CF was calculated as the mean score of these 7 MDS‐UPDRS items.

The frequency of everyday slips and errors was assessed with the CFQ, ²⁴ a 25‐item standardized instrument developed to assess the frequency of everyday slips and errors (eg, “Do you find you forget whether you've turned off a light or a fire or locked the door?”). Participants are asked to report the frequency of failure in each item on a 5‐point Likert scale from 1 (never) to 5 (always). The Hebrew version of the CFQ used in this study was translated with the permission of the original authors. Reliability of the current version was similar to the original (Cronbach's α = 0.94 for the entire current sample and α = 0.91 in the original study).

Functional aspects of PD were assessed with the Hebrew version (translated by the authors and the translation validated with permission of the original authors) of the PD‐CFRS, ²² a 12‐item self‐report questionnaire specifically designed to explore problems performing functions to detect PD‐MCI. The items explore whether the person has had trouble performing a daily activity (eg, 0 = none, 1 = some of the time, 2 = most of the time, 8 = never done the activity in the past). The initial PD‐CFRS validation study provided evidence for reliability and discriminant validity across stages of cognitive impairment. The Hebrew version of the PD‐CFRS used in this study was translated and the translation validated with permission of the original authors. Reliability of the current version was similar to the original (Cronbach's α = 0.74 in the entire current sample and α = 0.79 in the original study).

Procedure

The patients were invited for 1 or 2 study visits (maximum 2 weeks apart) at the Sheba Movement Disorders Institute. The visit included an evaluation by a movement disorders neurologist using the MDS‐UPDRS ³⁶ and the Hoehn and Yahr scale ²⁶ in the on medication (on) state. The on state was defined as the time after patients had taken their dopaminergic medication and, in those exhibiting motor fluctuations, when the dopaminergic medications had an effect. Demographic and PD‐related clinical data were collected from patient files along with documentation of medical comorbidities and PD medication. The total daily levodopa equivalent dose was calculated for each patient. ³⁸ In addition, a certified occupational therapist (N.G.) administered a neuropsychological battery that encompassed the aforementioned 5 level II cognitive domains ⁴ along with self‐report questionnaires focusing on participant's daily functional abilities (CFQ and CFRS). Evaluations were performed between June 2017 and February 2019.

Statistical Analysis

Analyses were performed using IBM SPSS Statistics software (version 23; IBM Corp., Armonk, NY). Statistical assumptions for all analyses were tested to ensure the accurate interpretation of results, with the level of statistical significance set at α = 0.05. The t tests for continuous variables and chi‐square tests for categorical variables were conducted to assess for significant differences between groups. To compare the groups' neuropsychological assessment z scores, the Bonferroni α adjusted level of 0.005 was calculated. For the self‐report questionnaire scores, the Bonferroni α adjusted level of 0.016 was calculated and used to adjust for the number of comparisons to reduce the probability of type I error.

Results

A total of 78 patients with PD (64.7% men), aged 42 to 78 years (mean ± standard deviation, 63.6 ± 8.5 years) were recruited to the study. Disease duration since diagnosis was 4.7 ± 5.0 years and ranged from 0.5 to 13 years. Approximately half of the participants received levodopa treatment, whereas the others received dopamine agonists (n = 5), monoamine oxidase type B inhibitors (n = 55), or amantadine (n = 24) alone or in combination.

The MDS‐UPDRS Part III motor scores of the entire sample ranged from 6 to 71 (26.2 ± 11.7) when on medication (and in patients experiencing motor fluctuations, when in the on condition). Total scores ranged from 14 to 110 (43.4 ± 18.6).

The low MoCA group consisted of 46 patients (MoCA, 23.9 ± 1.1), and the high MoCA group consisted of 32 patients (MoCA, 27.3 ± 1.3). No significant group differences were found regarding Beck Depression Inventory scores, demographic characteristics, or PD‐related medical features (Table 1). The results of the neuropsychological tests are presented in Table 2 below.

TABLE 1.

Demographic characteristics and medical status of patients with PD according to MoCA group

Variable	Low MoCA 22–25, n = 46; Mean ± SD or %	High MoCA 26–30, n = 32; Mean ± SD or %	t/χ2	P Value
Age, y	65.9 ± 7.1	62.1 ± 8.8	t 76 = 2.12	0.04
Gender, male	67.4	71.9	χ2 = 1.78	0.67
Education, y	15.2 ± 3.7	15.5 ± 3.7	t 76 = −0.36	0.72
PD duration since diagnosis, y	4.4 ± 3.4	3.8 ± 3.1	t 76 = 0.74	0.46
Hand dominance, right	89.1	90.6	χ2 = 1.89	0.39
Country of birth
Israel	73.9	84.4	χ2 = 2.367	0.67
Other	26.0	12.5
Hoehn and Yahr stage on	1.80 ± 0.49	1.77 ± 0.46	t 76 = 0.353	0.73
MDS‐UPDRS Part I	5.7 ± 4.5	7.2 ± 5.5	t 76 = −1.326	0.19
MDS‐UPDRS Part II	8.26 ± 4.9	10.4 ± 6.5	t 76 = −1.667	0.10
MDS‐UPDRS Part III	26.2 ± 10.9	26.2 ± 12.9	t 76 = 0.014	0.99
MDS‐UPDRS Part IV	1.5 ± 0.87	1.8 ± 3.7	t 76 = −0.363	0.72
MDS‐UPDRS total	41.7 ± 16.2	45.56 ± 21.614	t 76 = −0.897	0.46
LED	405.0 ± 367.1	337.3 ± 309.8	t 76 = 0.853	0.39
Patients treated with levodopa	47.8	50.0	χ2 = 0.037	0.85

Abbreviations: PD, Parkinson's disease; MoCA, Montreal Cognitive Assessment; SD, standard deviation; MDS‐UPDRS P‐I, Movement Disorder Society–Unified Parkinson's Disease Rating Scale Part I, nonmotor experiences of daily living; MDS‐UPDRS P‐II, Movement Disorder Society–Unified Parkinson's Disease Rating Scale Part II, motor experiences of daily living; MDS‐UPDRS P‐III, Movement Disorder Society–Unified Parkinson's Disease Rating Scale Part III, motor examination; MDS‐UPDRS P‐IV, Movement Disorder Society–Unified Parkinson's Disease Rating Scale Part IV, motor complications; LED, the total daily levodopa equivalent dose.

TABLE 2.

Comparison of the neuropsychological assessment z or scaled scores in patients with PD according to MoCA group

Domain	Variable	Low MoCA 22–25, n = 46; Mean ± SD	High MoCA ≥26, n = 32; Mean ± SD	t	P Value	d
Neuro tests
Attention	TMT‐A	2.80 ± 2.34	1.51 ± 2.43	2.26	0.03	0.54
	Digit Span Forward	0.08 ± 0.93	0.31 ± 0.74	−1.20	0.23	0.27
Executive functions
	TMT‐B	2.47 ± 2.56	1.30 ± 2.79	1.89	0.06	0.43
	Digit Span Backward	0.12 ± 0.73	0.24 ± 0.84	−0.67	0.50	0.15
Language	WAIS Voc	10.93 ± 2.17	11.78 ± 1.90	−1.78	0.08	0.42
Memory	RCFT delayed copy	8.64 ± 3.38	9.28 ± 3.54	−0.806	0.42	0.18
	WMS‐III logical memory‐stories immediate recall	10.69 ± 2.75	10.48 ± 2.79	0.324	0.74	0.08
	WMS‐III logical memory‐stories delayed recall	11.09 ± 2.29	11.19 ± 2.93	−0.175	0.86	0.04
Visuo‐spatial	RCFT copy	7.48 ± 2.48	9.53 ± 3.72	−0.273	0.01	1.10
	WAIS block design	9.55 ± 1.95	10.34 ± 2.72	−1.417	0.16	3.60
Self‐report questionnaires
CFQ mean score		0.98 ± 0.63	1.11 ± 0.47	−0.99	0.32	0.23
PD‐CFRS total score		2.16 ± 2.74	2.64 ± 2.38	−0.80	0.43	0.19
UPDRS‐CF mean score		0.69 ± 0.48	0.88 ± 0.61	−1.56	0.12	0.35

Abbreviations: PD, Parkinson's disease; MoCA, Montreal Cognitive Assessment; SD, standard deviation; TMT‐A, Trail Making Test–Part A; Digit Span Forward, Wechsler Adult Intelligence Scale (WAIS)‐III digit span forward; TMT‐B, Trail Making Test–Part B; Digit Span Backward, Wechsler Adult Intelligence Scale–III digit span backward; WAIS Voc, Wechsler Adult Intelligence Scale (WAIS)–III vocabulary test; RCFT delayed copy, Rey Osterrieth Complex Figure Test recall; WMS‐III immediate recall, Wechsler Memory Scale immediate verbal recall of stories; WMS‐III delayed recall, Wechsler Memory Scale delayed verbal recall of stories; RCFT copy, Rey Osterrieth Complex Figure Test copying; WAIS block design, Wechsler Adult Intelligence Scale (WAIS)–III block design; CFQ, Cognitive Failures Questionnaire; PD‐CFRS, Parkinson's Disease Cognitive Functional Rating Scale; UPDRS‐CF, Unified Parkinson's Disease Rating Scale cognitive functional variable.

Considering the Bonferroni α adjusted level, no significant group differences were found. However, a trend for signification was found in the attention (TMT‐A) and the copying task of the RCFT. ³⁴ , ³⁵

The internal reliability of all 3 self‐reported scales was found as acceptable to high in the current study (N = 78): the MDS‐UPDRS‐CF ²⁷ (α = 0.71), the CFQ ²⁴ (α = 0.93), and the PD‐CFRS ²² (α = 0.75). No significant group differences were found on any self‐report scales.

A high percentage of the low MoCA group participants obtained neuropsychological test scores within the normal range. In addition, a notable number of participants in the high MoCA group were identified with MCI‐level scores on various neuropsychological tests (eg, TMT‐A, 43.8%; TMT‐B, 21.9%; Rey complex figure delayed copy, 25%; and Rey complex figure copy, 25). It is worth noting that only the TMT‐B test differentiated between the low and high MoCA groups. In line with the MDS diagnostic criteria to define PD‐MCI, ⁴ Figure 1 presents the percentage of participants in each MoCA group who obtained impaired scores in 0 to all 5 neuropsychological assessments. A comparison of the identified PD participants with neuropsychological deficits based on the neuropsychological assessments are shown in Table 3 below.

Figure 1.

Number and percentage of PD participants with pathological neuropsychological scores in each group according to high/low MoCA scores. PD, Parkinson's disease; MoCA, Montreal Cognitive Assessment.

TABLE 3.

Comparison of frequency of identified PD participants with neuropsychological deficits based on neuropsychological assessment cutoff scores

			Participants With Pathology According to Original Cutoff
Domain	Raw Variable	Cutoff	Low MoCA 22–25, n = 46; n (%)	High MoCA ≥26, n = 32; n (%)	χ2	P Value
Attention	TMT‐A	z ≥ 1.5	30 (65.2)	14 (43.8)	3.54	0.06
	Digit Span Forward	z ≤ −1.5	2 (4.3)	1 (3.1)	0.76	0.78
Executive functions	MoCA clock drawing test	z ≤ −1.5	0	0
	TMT‐B	z ≥ 1.5	23 (50.0)	7 (21.9)	6.31	0.01
	Digit Span Backward		0	0
Language	Meaning of words	SS ≤6	1 (2.2)	0	0.71	0.40
	MoCA naming	z ≤ −1.5	6 (13.0)	1 (3.1)	2.27	0.13
	MoCA language fluency	z ≤ −1.5	3 (6.5)	1 (3.1)	0.45	0.50
Memory	REY figure delayed copy	SS ≤6	12 (26.1)	8 (25.0)	0.01	0.91
	Logical memory‐stories immediate recall	SS ≤6	2 (4.3)	2 (6.3)	0.14	0.71
	Logical memory‐stories delayed recall	SS ≤6	2 (4.3)	2 (6.3)	0.14	0.71
Visuo‐spatial	REY figure copy	SS ≤6	14 (30.4)	8 (25.0)	0.28	0.60
	WAIS block design	SS ≤6	1 (2.2)	2 (6.3)	0.85	0.36

Abbreviations: PD, Parkinson's disease; MoCA, Montreal Cognitive Assessment; TMT‐A, Trail Making Test–Part A; Digit Span Forward, Wechsler Adult Intelligence Scale (WAIS)–III digit span forward; TMT‐B, Trail Making Test–Part B; Digit Span Backward, Wechsler Adult Intelligence Scale (WAIS)–III digit span backward; MoCA naming, MoCA naming norms according to normal controls norm scores; WAIS block design, Wechsler Adult Intelligence Scale (WAIS)–III block design; SS, Scaled Score.

As presented in Figure 1, the diagnosis of suspected PD‐MCI according to the level I criteria (MoCA ≤25) did not correspond well with the level II criteria. On one hand, 10.9% of the participants with low MoCA (≤25) exhibited normal performance, and 23.9% exhibited pathology on just 1 neuropsychological assessment. On the other hand, 45% of the high MoCA group (≥26) exhibited cognitive deficits on 2 or more neuropsychological tests.

To further analyze the relationship between MoCA scores and the neuropsychological tests scores, we then divided the entire PD sample into 2 groups based on the patient's z scores or cutoff scores obtained in the neuropsychological battery. ³ , ²⁰ , ²² Those with scores below the predefined cutoff on 1 or more neuropsychological test were categorized as group 1 (n = 34); those with scores that indicated a pathology on 2 or more neuropsychological tests were categorized as group 2 (n = 44). Although no significant group differences were found in gender, age, education, or years with PD, group 1 achieved significantly higher MoCA scores compared with group 2 (respectively, 26.09 ± 2.15, 24.70 ± 1.83, t ₇₆ = 3.06, P = 0.003). However, when examining the self‐report measures—the CFQ, PD‐CFRS, and UPDRS‐CF scores—no significant group differences were found.

When analyzing the relationship between the MoCA scores and the self‐report scales of the entire sample, no significant correlations were found between the MoCA scores and the CFQ mean score (r = 0.10, P = 0.37), the PD‐CFRS score (r = −0.04, P = 0.72), or the UPDRS‐CF score (r = 0.11, P = 0.32). However, significant positive correlations were found between the UPDRS‐CF and both the CFQ score (r = 0.38, P = 0.001) and the PD‐CFRS score (r = 0.35, P = 0.002) in the entire sample. A significant medium correlation was found in the entire sample between the CFQ and the PD‐CFRS scores (r = 0.50, P < 0.001).

Following the surprisingly incongruous results between the MoCA groups when analyzing the neuropsychological tests and the self‐report questionnaire results, we divided the patients with PD into 2 groups based on the UPDRS‐CF score (which is the mean score of 7 items): group A (UPDRS‐CF score ≥ 1), which has cognitive functional difficulties, and group B (<1), which has no difficulties. As presented in Table 4, significant UPDRS‐CF group differences were found for the MoCA, CFQ, and PD‐CFRS scores.

TABLE 4.

Comparison of MoCA score, CFQ, and PD‐CFRS self‐report scores between groups according to UPDRS cognitive functional score

Test	Group A, UPDRS‐CF ≥1 (n = 25), With Cognitive Functional Difficulties; Mean ± SD	Group B, UPDRS‐CF < 1 (n = 53), Without Cognitive Functional Difficulties; Mean ± SD	t	P Value
MoCA	26.00 ± 2.04	24.98 ± 2.04	t 76 = 2.06	0.04
CFQ	1.28 ± 0.65	0.92 ± 0.49	t 75 = 2.70	0.01
CFRS	0.28 ± 0.23	0.16 ± 0.19	t 66 = 2.37	0.02

Abbreviations: MoCA, Montreal Cognitive Assessment; CFQ, Cognitive Failures Questionnaire; PD‐CFRS, Parkinson's Disease Cognitive Functional Rating Scale; UPDRS‐CF, Unified Parkinson's Disease Rating Scale cognitive functional variable; SD, standard deviation.

We further examined the concurrence between MoCA and the UPDRS‐CF scores. In the low MoCA group, only 11 patients (24%) reported cognitive functional difficulties (MDS‐UPDRS‐CF ≥1); in the high MoCA group, 14 people (44%) reported cognitive functional difficulties.

Discussion

This interdisciplinary study was conducted by researchers from 2 disciplines—the biomedical field (movement disorders neurologists) and occupational therapy—in an attempt to analyze the PD‐MCI syndrome from both the medical and the daily functional viewpoints within a holistic patient‐centered approach.

Participants in this study were men and women with an average of 15 education years and diagnosed with PD for a mean of 4 years, but still conducting independent and active lives. Identification of PD‐MCI as suggested by the established clinical criteria guidelines for assessing and defining PD‐MCI ⁴ was not found to be conclusive according to MoCA. Such delineation was not supported by the neuropsychological testing, the MDS‐UPDRS measure, or the functional cognitive score. ³ In fact, attempting to distinguish between the groups based on the MoCA scores showed the definition to be weak and ambiguous.

Examination of the neuropsychological assessment revealed that, although not addressing the significant levels, 2 tests presented a tendency for differentiation between the 2 MoCA groups. Specifically, the TMT‐A and RCFT copy tasks represented skills that spread across 2 cognitive domains (attention and visual spatial), whereas praxis and visuo‐spatial previously had been found as predictors for conversion from PD‐MCI to PD with dementia. ³⁹ However, considering the Bonferroni α adjusted level, we did not find significant differences between the groups. Therefore, determining definite cognitive decline according to the suggested criteria ⁴ is complex and ambiguous (eg, ref. 40).

To further understand the complexity of identifying PD‐MCI, we compared the pathological scores found in the neuropsychological tests in each MoCA group. Although we found that more participants in the low MoCA group had difficulty with attention (TMT‐A) in comparison with the high MoCA group, no significant group difference was found. This unanticipated result also was evident in the more demanding TMT‐B score, which assesses executive functions. Possibly, the sensitivity of the TMT to impairment in multiple cognitive domains caused the pathological scores. Therefore, the decline may be attributed to attention deficits or other cognitive deficits, such as cognitive flexibility, in either group. ⁴¹ Still, these results did not sufficiently clarify the diagnosis criteria of PD‐MCI.

Although scores indicating cognitive impairment were not expected in the MoCA ≥26 group, high percentages of scores indicating cognitive impairment were found in a range of 1 to 4 tests among participants in both MoCA groups. This finding makes differentiating between those with and without cognitive challenges even more complex and may jeopardize the ability to understand their actual daily challenges. In fact, in our sample, some participants defined with cognitive decline according to the MoCA obtained scores within normal range in other neuropsychological tests. Others, defined without cognitive decline according to the MoCA, obtained deficit scores in other cognitive tests. Such ambiguity in results concerning measures of cognitive decline can have critical clinical implications in determining further assessment and treatment plans.

Further to the MoCA and neuropsychological tests, we assessed the value of information received from the patients’ self‐reports. Despite the limitations of self‐reports, ⁴² including subjectivity, this type of data enables people to report their own feelings and the perceptions of their daily functional cognitive abilities. Indeed, adding self‐reported data in this study revealed additional information by using a non‐disease‐specific questionnaire (CFQ) ²⁴ and a PD‐specific questionnaire (PD‐CFRS) ²² and by creating the new UPDRS‐CF to identify MCI reflected in functional cognition. ³⁷ , ⁴³ The positive correlations found among these 3 questionnaires indicate that participants reported consistently about their cognitive functional abilities. This reinforces the importance of listening to each person's perspective and considering his or her subtle daily functioning challenges. Possibly, subtle changes that a person experiences but is not always asked about are the real‐world evidence that some sort of cognitive change is occurring.

The percentage of participants who reported cognitive functional difficulties in the UPDRS‐CF groups again affirms how the self‐report results do not correspond with the MoCA scores that define the cutoff for MCI. The important information derived from the self‐reports cannot be ignored, especially considering the sufficient internal reliability found in all 3 measures. This is critical when considering the fourth functional criterion for PD‐MCI ⁴ —that cognitive deficits do not significantly interfere with functional independence, although subtle difficulties on complex functional tasks may be present. Ultimately, 24% of the participants with normal MoCA scores reported that they experienced functional cognitive difficulties, whereas 76% of participants with low MoCA scores reported none.

Overlooking MCI in the initial cognitive screening process because of normal MoCA scores can prevent people from being referred for additional assessment and consequently being provided a suitable intervention. Missing this opportunity to improve the daily functional deficits of patients with PD (which relate not only to motor but also in some patients to an additional cognitive component) could lead to further functional deterioration and could increase their frustration and emotional reactions. ⁴⁴ , ⁴⁵

This raises the question of what type of assessment we can and should rely on. What is the most effective, accurate, and standardized method to identify PD‐MCI? Because questionnaires are probably insufficient, lacking objectivity and perspective, we could consider an assessment process that includes a set of real‐world tasks that the patient performs. An attractable alternative option is to collect objective performance measures of real‐life tasks that reflect the patient's actual functional cognition status ⁴⁶ via cellular technology, such as a smartphone in a patient's home environment. These could include routine daily tasks, such as preparing coffee or arranging a travel bag (to be published separately).

Some limitations of the current study cannot be ignored. First, the sample in this study had a high education level. Although a significant correlation between cognitive decline and education level was not found, the MoCA may be less sensitive to cognitive decline in this population group. ⁴⁷ In addition, this study's limited sample and use of the neuropsychological tests’ z scores need to be considered when generalizing the results. Future studies may indicate whether implementing the composite score Bezdicek and colleagues ¹⁴ developed for the neuropsychological battery would enhance concordance with the MoCA scores toward improving PD‐MCI detection. Another limitation is the inclusion of a single‐language test in the neuropsychological battery. Future studies with larger and more heterogeneous worldwide samples are recommended to explore the issue of DP‐MCI and its diagnosis, combining real‐world performance assessments.

Author Roles

(1) Research Project: A. Conception, B. Organization, C. Recruitment, D. Execution; (2) Statistical Analysis: A. Design, B. Execution, C. Review and Critique; (3) Manuscript Preparation: A. Writing of the First Draft, B. Review and Critique.

S.R.: 1A, 1B, 2A, 2B, 2C, 3A, 3B

S.M.: 1B, 1D, 2A, 2B, 2C, 3A, 3B

N.G.: 1B, 1D

L.M.: 1B, 1D

A.R.: 3B

S.I.K.: 1C, 2C, 3B

V.L.: 1C, 1D

T.F.K.: 1C, 1D

T.N.: 1C, 1D

G.Y.: 1C, 1D

S.H.: 1B, 1C, 1D, 2A, 2B, 2C, 3A, 3B

Disclosures

Ethical Compliance Statement

This study was approved by the ethics committee of the Sheba Medical Center. Written declaration of patient consent was obtained from each participant and documented. We confirm that we have read the Journal's position on issues involved in ethical publication and affirm that this work is consistent with those guidelines.

Funding Sources and Conflict of Interest

This study was funded by a grant provided by the Israeli Ministry of Science and Technology (3‐13719). The authors declare that there are no conflicts of interest relevant to this work.

Financial Disclosures for Previous 12 Months

S.R. reports a National Insurance grant and an Insurance Companies Association grant. S.I.K. reports Medison. G.Y. reports AbbVie. S.H. reports a National Insurance grant and an Insurance Companies Association grant and AbbVie, Teva, and Medison. S.M., N.G., L.M., A.R., V.L., T.F.K., and T.N. declare that there are no additional disclosures to report.