A Retrospective Analysis of the Sentence Writing Component of the Mini Mental State Examination: Cognitive and Affective Aspects

A Retrospective Analysis of the Sentence Writing Component of the Mini Mental State Examination: Cognitive and Affective Aspects
of 7
All materials on our website are shared by users. If you have any questions about copyright issues, please report us to resolve them. We are always happy to assist you.
Related Documents
  Fax +41 61 306 12 34E-Mail  Original Research Article Dement Geriatr Cogn Disord 2012;33:125–131 DOI: 10.1159/000337843 A Retrospective Analysis of the Sentence Writing Component of the Mini Mental State Examination: Cognitive and Affective Aspects Yan Press a–e  Natalia Velikiy a–c  Alex Berzak a–e  Howard Tandeter a–c  Roni Peleg a–c  Tamar Freud a–c  Boris Punchik a–e  Tzvi Dwolatzky b, d–f    a  Department of Family Medicine, b  Faculty of Health Sciences, and c  Siaal Research Center for Family Medicine and Primary Care, Faculty of Health Sciences, Ben-Gurion University of the Negev, d  Comprehensive Geriatric Assessment Unit, e  Community Geriatric Service, Southern District, Clalit Health Services, and f   Geriatric Department and Memory Clinic, Mental Health Center, Beer-Sheva , Israel   Introduction The written word and language serves as a mirror to many aspects of human function and behavior. Hand-writing is closely related to the functioning of the mind and is therefore linked with how people think, feel and behave [1] . Agraphia can be de fi ned as the partial or total loss of the ability to produce written language associated with brain pathology [2] . Alzheimer’s disease (AD) is a neurodegenerative condition resulting in progressive cognitive and functional decline, and agraphia or dys-graphia are common features of this disease. As AD pro-gresses there is a deterioration in spelling, poor narrative organization, content word errors and perseverations, re-duced syntactic complexity, and shorter sentences [3, 4] . In a study of patients with AD, the ability to formulate a complete sentence was related to dementia severity [5] . When asked to write a story about a picture, patients with AD of mild to moderate severity wrote shorter descrip-tive paragraphs than normal subjects matched for age and gender (28.4 vs. 79.9 words) [6] . The Mini Mental State Examination (MMSE) [7] is widely used as a screening instrument for cognitive im-pairment. It assesses a number of cognitive domains, namely orientation, registration, concentration, short-term memory, language and visual spatial function. The MMSE includes a request to ‘write a sentence’, with 1 Key Words Mini Mental State Examination   Screening   Cognitive impairment   Dementia   Depression   Dysgraphia   Language   Emotion   Aging Abstract Background:  One of the components of the Mini Mental State Examination (MMSE) is the request to write a sentence. We investigated the relationship between the characteristics of the written sentence of the MMSE and the cognitive and affective status of elderly patients. Methods:  The character-istics of the sentence were compared to the total MMSE score, sociodemographic characteristics, tests evaluating cognition and affective status, and diagnoses. Results:  The number of words was significantly associated with the de-gree of cognitive impairment, whereas the emotional polar-ity of sentences and concerns about health were associated with depression. Conclusions:  Characteristics of the MMSE sentence may provide important additional information re-garding both cognition and affect when assessing older people. Copyright © 2012 S. Karger AG, Basel Accepted: February 29, 2012 Published online: April 21, 2012 Yan Press, MD Comprehensive Geriatric Assessment Unit, Clalit Health Services, Yassky Clinic 24 King David Street Beer-Sheva 84539 (Israel) Tel. +972 8640 7743, E-Mail yanp @ © 2012 S. Karger AG, Basel1420–8008/12/0333–0125$38.00/0 Accessible online   Press /Velikiy /Berzak /Tandater /Peleg /Freud /Punchik /Dwolatzky Dement Geriatr Cogn Disord 2012;33:125–131 126 point awarded if the sentence has a subject and a verb and makes sense regardless of spelling, grammar or content. Depending on the setting the MMSE has been shown to have reasonable sensitivity and good specificity [7–12] . Not all components of the MMSE are equal as indicators of cognitive function [13–15] , and writing a sentence is one of the components that contributes to the variance in global MMSE scores. In a retrospective review by McCarthy et al. [16] the sentences written by 280 patients from a geriatric day hospital performing the MMSE were analyzed. Signifi-cant correlation was detected between the overall MMSE score and the number of words. The mean MMSE score of subjects writing ‘positive’ sentences was marginally but significantly higher than that of those writing a ‘neutral’ or ‘negative’ sentence. In another study [17] , the sentences written as a task of the MMSE were compared between cognitively healthy older subjects and those with dementia. Sentences with inadequate emotional content seemed to indicate a de-menting illness. In this study we investigated the relationship between the characteristics of the written sentence of the MMSE and the cognitive and affective status of patients attending the Comprehensive Geriatric Assessment Unit (CGAU) of Clalit Health Services in the Southern District of Israel. Materials and Methods Participants and Instruments The CGAU was established in the Southern District of Israel by the Clalit Health Services in 2005. The staff of the CGAU in-cludes geriatricians, a geriatric nurse, a social worker, an occupa-tional therapist and a secretary. The average annual turnover of the CGAU is approximately 150 patients. Patients are referred to the CGAU for cognitive, functional and affective symptoms, as well as problems with mobility or falls, and social difficulties. As part of the multidisciplinary assessment, cognitive status is evalu-ated by means of the MMSE, and in patients with mild to moder-ate cognitive impairment further evaluation includes the Mon-treal Cognitive Assessment instrument [18] , as well as computer-ized neuropsychological testing using the Mindstreams battery (Neurotrax, Newark, N.J., USA) [19] . Affective status is evaluated using the 15-item Geriatric De-pression Scale (GDS-15) [20] . The diagnoses of dementia and de-pression are based on clinical findings according to the DSM-IV criteria [21] , and that of mild cognitive impairment (MCI) is based on the criteria of the Expert Conference on Mild Cognitive Im-pairment [22] . Functional assessment includes the Barthel Index [23] for basic activities of daily living as well as the Older Americans Resources and Services scale for instrumental activities of daily living [24] . For measuring comorbidity we use two instruments. The first is the Cumulative Illness Rating Scale-Geriatrics [25] , which permits an estimate of illness burden and diversity on the basis of a fi  ve-point physician rating scale (scores of 0–4, the higher score indicat-ing a greater degree of burden) re fl ecting the severity of pathology in each of 14 categories (maximum score 56). The second instru-ment used for assessing comorbidity is the Charlson’s Comorbid-ity Index [26] . Because this index was designed primarily as a pre-dictor of mortality it presents a list of 19 conditions with fi xed de-grees of severity according to the relative risk of death. For risk calculated at 6  1.2 but !  1.5, the disease receives a score of 1; for relative risk 6  1.5 but !  2.5, a score of 2, and for relative risk 6  2.5 but !  3, a score of 3; both a second metastatic solid tumor and HIV-AIDS receive a score of 6. A total score is calculated. The age score represents an extra point for each decade above age 50 and is used for adjusting the Charlson’s Comorbidity Index for age. The total combined score is derived by adding the age score to the total score.  Procedures and Characteristics of the Written Sentence Three independent investigators (N.V., Y.P. and A.B.), blinded to the results of geriatric assessment and all tests performed, were asked to assess the following characteristics of the written sen-tence of the MMSE: number of words in the sentence, emotional polarity (positive/neutral/negative) and concerns about health (yes/no). Initially, the investigators analyzed 30 MMSE written sentences to determine interrater reliability. The concordance be-tween raters was an acceptable   100% for characteristic one, 92% for characteristic two and 96% for characteristic three. Subse-quently, all analyses were performed by one investigator who speaks fluent Hebrew and Russian (N.V.). The Ethics Committee of the Meir Hospital approved the study. Statistical Analysis The characteristics of the sentence were compared to the total MMSE score, the language of administration, age, gender, educa-tion, tests evaluating cognition and affective status, and clinical diagnoses. Continuous variables are shown as means and stan-dard deviations. Categorical variables are described as frequen-cies. t tests and    2  tests were used to analyze statistically signifi-cant differences of continuous and categorical variables, respec-tively. Two-tailed p !  0.05 were considered statistically significant. Results Data of 344 patients evaluated in the CGAU during the years 2007–2009 were assessed. The basic characteristics of the study population are presented in table   1 .  Associations between MMSE Sentence Characteristics and the Language of Assessment A total of 209 (60.8%) were Russian speakers who un-derwent assessment in Russian; 135 (39.2%) were fluent in Hebrew and were assessed in Hebrew, although He-brew was the native language of only 6 (1.7%). No differ-ences were found between the number of words in the MMSE sentences of Hebrew and Russian speakers (3.9 8  1.3 vs. 3.9 8  1.2, p = 0.85)   Sentence Writing Component of the MMSE Dement Geriatr Cogn Disord 2012;33:125–131 127  Impact of Age, Gender and Education on the MMSE Sentence Characteristics (   table   2   )   Of the female patients, 41 (18.1%) expressed concern about their health in the MMSE sentence versus 11 (10.1%) of the male patients, p value = 0.04. No other associations between the characteristics of the MMSE sentence were found regarding gender, age and education.   Associations between MMSE Sentence Characteristics and Cognitive and Affective Tests The association between the number of words in the MMSE sentences and the overall MMSE score did not reach statistical significance (Pearson’s correlation 0.103, p = 0.057). A statistically significant association was found between the number of words and GDS-15, i.e. pa-tients with higher GDS-15 scores wrote shorter sentences (Pearson’s correlation –0.329, p = 0.03). With regard to the emotional polarity of MMSE sen-tences, patients writing ‘negative’ or ‘neutral’ sentences had higher GDS-15 scores (10.6 8  4.1 and 10.1 8  3.4, re-spectively) than those writing ‘positive’ (7.5 8  3.8) sen-tences (p = 0.03). No associations were found between the emotional polarity of MMSE sentences and overall MMSE scores (MMSE score 8  SD of the ‘negative’ sen-tences were 23.5 8  4.1, ‘neutral’ sentences 24.3 8  4.7 and ‘positive’ sentences 24.2 8  4.7, p = 0.64). With regard to overall MMSE scores, there was no as-sociation between patients who expressed concern about their health and those not expressing this concern (mean MMSE 23.2 8  4.3 vs. 24.1 8  4.8, p = 0.18). Similarly, no association was found with regard to GDS-15 (10.5 8  3.6  vs. 9.6 8  3.6, p = 0.36)   Associations between MMSE Sentence Characteristics and Clinical Diagnoses (   table   3 )   Of the 344 patients, there were 142 (41.3%) with a di-agnosis of dementia, 84 (24.4%) with MCI and 118 (34.3%) without cognitive impairment. During the geriatric as-sessment a total of 128/344 patients (37.2%) were diag-nosed with depression.  Number of Words  . The number of words in MMSE sentences had a statistically significant association with the degree of cognitive impairment. While cognitively normal patients wrote longer (4.18 8  1.41) sentences, sentences were shorter in those with MCI (3.82 8  1.14) and dementia (3.74 8  1.22, p value = 0.02). No associa-tions were found between the number of words in the sentence and a diagnosis of depression.   Emotional Polarity of Sentence. Overall 323 sentences of the MMSE were classified according to their emotional polarity (in 21 instances MMSE sentences were ‘bizarre’ or without any information, such as ‘I am here’, ‘My sister’, etc.). Sentences were determined to be ‘negative’ in 32 (9.9%) patients (e.g. ‘My life is very hard’, ‘I am very pes-simistic about the future’, etc.), ‘neutral’ in 215 (66.6%) and ‘positive’ in 76 (23.5%) (e.g. ‘I am always very satisfied with life’, ‘Very beautiful and interesting life’, etc.). The emo-tional polarity of sentences had no association with cogni-tive diagnoses but had a statistically significant association with the diagnosis of depression, with 21 patients (17.5%) of those writing a ‘negative’ sentence being depressed com-pared to 11 (5.4%) patients without a diagnosis of depres-sion writing a ‘negative’ sentence (p value !  0.0001). Concerns about Health  . Concerns about health ex-pressed in the sentence were not associated with the cog-nitive diagnosis. However, while 30 of the 128 patients (23.4%) with depression wrote a sentence expressing this concern, only 23 of the 216 patients (10.6%) without de-pression expressed concerns about their health (p value !  0.0001).  Impact of Age, Gender and Education on Cognitive and Affective Status ( table   4 )   Age was inversely proportional to MMSE scores (Pear-son’s correlation –0.233, p = 0.000). However, there was Table 1.  C haracteristics of study population (n = 344)Characteristics (patients with available data)Mean 8  SDRangePatientsn (%)Age (n = 334), years78.2 8 5.465–92Gender, female (n = 335)226 (67.5)Marital status, married (n = 219)109 (49.8)Education (n = 178), years10.9 8 4.20–20Language of assessment(n = 344)Hebrew135 (39.2)Russian209 (60.8)Comorbidity scoresTotal CIRS-G Score(n = 125)13.4 8 4.92–23TCS of CCI5.6 8 1.83–11Barthel Index (n = 186)81.8 8 15.735–100OARS (n = 182)7.3 8 3.11–14MMSE (n = 344)23.9 8 4.78–30GDS-15 (n = 79)9.75 8 3.641–15T otal CIRS-G = Total Cumulative Illness Rating Scale Geriat-ric Score; TCS of CCI = total combined score of the Charlson Co-morbidity Index; OARS = Older Americans Resources and Ser- vices Score.   Press /Velikiy /Berzak /Tandater /Peleg /Freud /Punchik /Dwolatzky Dement Geriatr Cogn Disord 2012;33:125–131 128 no significant difference in age between cognitively im-paired patients and those without impairment. No cor-relation was found between GDS-15 scores and age, nor was age correlated with the diagnosis of depression. No statistically significant correlations were found be-tween education on the one hand, and MMSE, GDS-15 scores and cognitive or affective diagnoses on the other. Although there was no association between the gender of patients and GDS-15, more women were diagnosed as de-pressed than men [98 women (43.4%) vs. 30 men (27.5%), p = 0.03]. Table 2.  I mpact of age, gender and education on MMSE sentence characteristics Number of words in sentenceEmotional polarity of sentenceConcerns about health in sentencePearson’s correlationmean 8  SDrangepatients negativepositiveneutralpatients yesno patients  Age, years Pearson’s correlation–0.032Mean 8  SD77.5 8 6.178.5 8 5.478.2 8 5.379.2 8 5.677.9 8 5.3Range 68–8865–9166–9268–9165–92Patients334317620831552282334p value0.560.710.13 Gender  Female3.9 8 1.32–1022624 (11.2)52 (24.3)138 (64.5)21441 (18.1)185 (81.9)226Male3.9 8 1.31–91097 (6.9)24 (23.5)71 (69.6)10211 (10.1)98 (89.9)109Patients335317620931652283335p value0.610.440.04 Education ≤ 12 years3.7 8 1.22–911515 (14.6)17 (16.5)71 (68.9)10325 (21.7)90 (78.3)115 ≥ 13 years4.0 8 1.22–9635 (8.1)16 (25.8)41 (66.1)627 (11.1)56 (88.9)63Patients178203311216532146178p value0.160.220.06 Figures are numbers (%) unless otherwise specified. Table 3.  A ssociations between MMSE sentence characteristics and diagnosis of MCI, dementia and depressionCognitive diagnosisD epression dementia (n = 142)MCI (n = 84)normal (n = 118)totalp valueyes (n = 128 )no (n = 216)totalp value Number of words Mean 8  SD3.74 8 1.223.82 8 1.144.18 8 1.413.91 8 8 1.193.96 8 1.333.91 8 1.23<0.36Range0–200–200–170–202–91–101–10Patients, n14284118344128216344 Emotional polarity of sentence Negative12 (9.5)9 (11.3)11 (9.4)32 (9.9)0.9421 (17.5)11 (5.4)32 (9.9)<0.001Positive28 (22.2)21 (26.3)27 (23.1)76 (23.5)24 (20.0)52 (25.6)76 (23.5)Neutral86 (68.3)50 (62.5)79 (67.5)215 (66.6)75 (62.5)140 (69.0)215 (66.6)Patients, n12680117323101203323 Concerns about health Yes25 (17.6)12 (14.3)16 (13.6)53 (15.4)0.6330 (23.4)23 (10.6)53 (15.4)<0.001No117 (82.4)72 (85.7)102 (86.4)291 (84.6)98 (76.6)193 (89.4)291(84.6)Patients, n14284118344128216344Figures are numbers (%) unless otherwise specified.   Sentence Writing Component of the MMSE Dement Geriatr Cogn Disord 2012;33:125–131 129  Discussion In this retrospective study we demonstrated a clear re-lationship between the characteristics of the sentence written as part of the MMSE test and the cognitive and affective features of older patients evaluated by the CGAU. The characteristics of the patients seen at the CGAU were indicated by low functional performance, high comor-bidity indexes, and a high prevalence of cognitive impair-ment and depression ( table   1 ). Apart from a gender predisposition, with more wom-en expressing concern about their health in the MMSE sentence than men, no other associations were found be-tween the characteristics of the MMSE sentence and age, gender or education. The greater degree of health con-cerns expressed by the women may reflect the fact that more women were diagnosed as being depressed. The finding that age does not affect the sentences is not consistent with that of a previous study [27] , which found that with increasing age individuals use more pos-itive and fewer negative words, use fewer self-references, use more future-tense and fewer past-tense verbs, and demonstrate a general pattern of increasing cognitive complexity. The lack of correlation with age in our study may be explained at least partially by the fact that our study population had a relatively high degree of cognitive and affective impairments. An important finding of our study is that there was no association between the number of words in the MMSE sentence and the overall MMSE score. This result is sim-ilar to that reported by Shenkin et al. [28] but is in con-trast with the finding of McCarthy et al. [16] . We found a significant association between the number of words per sentence and the cognitive diagnosis. The shortest sen- Table 4.  I mpact of age, gender and education on the cognitive and affective testsMMSEDementiaMCINormalGDS-15DepressionNormal  Age Pearson’s correlation–0.2330.0Mean 8  SD79.2 8 5.177.3 8 5.277.6 8 5.678.1 8 5.278.2 8 5.5Range 67–9265–8767–9269–8865–92Patients 3341398411176128206p value 0.0000.120.950.92 Education Mean 8  SD10.9 8 3.810.4 8 4.911.6 8 3.810.9 8 4.411.0 8 3.9Range 0–200–200–170–200–17Patients8452429385p value0.330.95 ≤ 12 years Mean 8  SD 23.2 8 4.610.4 8 3.9Range 8–301–15Patients 11526 ≥ 13 years Mean 8  SD 24.6 8 5.49.1 8 3.8Range 8–301–15Patients 638p value 0.080.42 Gender  MaleMean 8  SD 24.3 8 4.79.0 8 3.5Range 8–303–14Patients, n (%) 10937 (34.0)32 (29.4)40 (36.7)2030 (27.5)79 (72.5)FemaleMean 8  SD 23.8 8 4.710.0 8 3.8Range 8–301–15Patients, n (%) 226102 (45.2)52 (23.0)72 (31.9)5698 (43.4)128 (56.6)p value0.360.140.320.03
Similar documents
View more...
Related Search
We Need Your Support
Thank you for visiting our website and your interest in our free products and services. We are nonprofit website to share and download documents. To the running of this website, we need your help to support us.

Thanks to everyone for your continued support.

No, Thanks