Reducing misclassification of mild cognitive impairment based on base rate information from the uniform data set

Abstract Insomnia is a common disorder that affects up to 40% of people age 65 and older. Untreated insomnia can decrease quality of life, increase healthcare use, and exacerbate cognitive problems. Individuals with cognitive impairment experience more sleep disorders than those without cognitive concerns, yet little is known about insomnia and mild cognitive impairment (MCI). Our objective was to examine predictors of insomnia in persons with MCI (PwMCI). Using data from the National Alzheimer’s Coordinating Center Uniform Data Set, a cross-sectional study of older PwMCI was conducted. Independent sample t-tests and contingency tables with chi-square tests of independence were used to examine differences between PwMCI with and without insomnia. Multivariate binary logistic modeling was performed. The total sample (N=1543) was comprised of 234 (15.1%) with clinician-reported insomnia and 1309 (84.9%) without insomnia. PwMCI and insomnia were more likely to be younger, take more medications, and smoke cigarettes (p.05). Three variables significantly predicted insomnia in PwMCI subjects in a multivariate model: active depression (OR 1.66, 95%CI 1.21, 2.27), active anxiety (OR 2.16, 95%CI 1.57, 2.99) and arthritis (OR 1.78, 95%CI 1.33, 2.39). Differences in predictors of insomnia in PwMCI highlight the need for geriatric and mental health specialists to provide specialized care to this population. Future studies should examine conversion of PwMCI with insomnia to dementia and the compounding effects of insomnia on cognition.

Download Full-text

Comparing measures of decline to dementia in amnestic MCI subjects in the National Alzheimer's Coordinating Center (NACC) Uniform Data Set

International Psychogeriatrics ◽

10.1017/s1041610212000452 ◽

2012 ◽

Vol 24 (10) ◽

pp. 1553-1560 ◽

Cited By ~ 8

Author(s):

Sarah E. Monsell ◽

Danping Liu ◽

Sandra Weintraub ◽

Walter A. Kukull

Keyword(s):

Cognitive Impairment ◽

Mild Cognitive Impairment ◽

Neuropsychological Tests ◽

Mini Mental State Examination ◽

Clinical Dementia Rating ◽

Data Set ◽

Global Test ◽

Allele Status ◽

Rate Of Decline ◽

State Examination

ABSTRACTBackground: Many studies have investigated factors associated with the rate of decline and evolution from mild cognitive impairment (MCI) to Alzheimer's disease (AD) dementia in elderly patients. In this analysis, we compared the rates of decline to dementia estimated from three common global measures of cognition: Mini-Mental State Examination (MMSE) score, Clinical Dementia Rating sum of boxes (CDR-SB) score, and a neuropsychological tests composite score (CS).Methods: A total of 2,899 subjects in the National Alzheimer's Coordinating Center Uniform Data Set aged 65+ years diagnosed with amnestic mild cognitive impairment (aMCI) were included in this analysis. Population-averaged decline to dementia rates was estimated and compared for standardized MMSE, CDR-SB, and CS using Generalized Estimating Equations (GEE). Associations between rate of decline and several potential correlates of decline were also calculated and compared across measures.Results: The CDR-SB had the steepest estimated slope, with a decline of 0.49 standard deviations (SD) per year, followed by the MMSE with 0.22 SD per year, and finally the CS with 0.07 SD per year. The rate of decline of the three measures differed significantly in a global test for differences (p < 0.0001). Age at visit, body mass index (BMI) at visit, Apolipoprotein E (APOE) ɛ4 allele status, and race (black vs. white) had significantly different relationships with rate of decline in a global test for difference among the three measures.Conclusions: These results suggest that both the rate of decline and the effects of AD risk factors on decline to dementia can vary depending on the evaluative measure used.

Download Full-text

Informant Reporting in Mild Cognitive Impairment: Sources of Discrepancy on the Functional Activities Questionnaire

Journal of the International Neuropsychological Society ◽

10.1017/s1355617719001449 ◽

2020 ◽

Vol 26 (5) ◽

pp. 503-514 ◽

Cited By ~ 3

Author(s):

Katherine Hackett ◽

Rachel Mis ◽

Deborah A.G. Drabick ◽

Tania Giovannetti

Keyword(s):

Cognitive Impairment ◽

Mild Cognitive Impairment ◽

Outcome Measurement ◽

Everyday Functioning ◽

Data Set ◽

Black African ◽

Functional Activities ◽

Visit Frequency ◽

Post Hoc ◽

Racial Ethnic

AbstractObjective:Relative to dementia, little is known about informant bias in mild cognitive impairment (MCI). We investigated the influence of informant demographic and relational characteristics on reports of everyday functioning using the Functional Activities Questionnaire (FAQ).Method:Four thousand two hundred eighty-four MCI participants and their informants from the National Alzheimer’s Coordinating Center Uniform Data Set were included. Informants were stratified according to cohabitation, relationship, visit frequency, race/ethnicity, education, and sex. Informant-rated Mean FAQ score was compared across these groups using univariate general linear model analyses and post hoc tests. Interactions were tested between informant variables. The predictive contribution of informant variables to FAQ score was explored using hierarchical linear regression. Analyses covaried for participant cognition using a cognitive composite score, and for participant age, sex, and depression.Results:After controlling for participant cognition, depression, age, and sex, informant-rated FAQ scores varied significantly across all informant variables (p’s < .005, ηp2’s ≤ .033) except sex and visit frequency. FAQ scores were higher (more impaired) among informants who cohabitate with the participant, among paid caregivers, spouses, and adult children, and among informants with higher levels of education. Scores were lowest (less impaired) among Black/African American informants as compared to all other racial/ethnic groups.Conclusions:Demographic and relational characteristics of informants influence the perception and reporting of instrumental activities of daily living in adults with MCI. As everyday functioning is crucial for differential diagnosis and treatment outcome measurement, it is important to be aware of sources of informant report discrepancies.

Download Full-text

The ecological validity of the Uniform Data Set 3.0 neuropsychological battery in individuals with mild cognitive impairment and dementia

The Clinical Neuropsychologist ◽

10.1080/13854046.2020.1837246 ◽

2020 ◽

pp. 1-18

Author(s):

Jared F. Benge ◽

Jonathan D. Artz ◽

Andrew M. Kiselica

Keyword(s):

Cognitive Impairment ◽

Mild Cognitive Impairment ◽

Ecological Validity ◽

Neuropsychological Battery ◽

Data Set

Download Full-text

The Association Between the Number of Neuropsychological Measures and the Base Rate of Low Scores

Assessment ◽

10.1177/1073191119864646 ◽

2019 ◽

pp. 107319111986464 ◽

Cited By ~ 4

Author(s):

Javier Oltra-Cucarella ◽

Miriam Sánchez-SanSegundo ◽

María Rubio-Aparicio ◽

Juan Carlos Arango-Lasprilla ◽

Rosario Ferrer-Cascales

Keyword(s):

Cognitive Impairment ◽

Mild Cognitive Impairment ◽

Base Rate ◽

Z Score ◽

Expected Number ◽

Cognitively Normal ◽

Neuropsychological Measures ◽

Normal Individuals

Obtaining one or more low scores, or scores indicative of impairment, is common in neuropsychological batteries that include several measures even among cognitively normal individuals. However, the expected number of low scores in batteries with differing number of tests is unknown. Using 10 neuropsychological measures from the National Alzheimer’s Coordinating Center database, 1,023 permutations were calculated from a sample of 5,046 cognitively normal individuals. The number of low scores (i.e., z score ≤−1.5) varied for the same number of measures and among different number of measures and did not increase linearly as the number of measures increased. According to the number of low scores shown by fewer than 10% of the sample, cognitive impairment should be suspected for 1 or more, 2 or more, and 3 or more in batteries with up to 2 measures, 3 to 9 measures, and 10 measures, respectively. These results may increase the identification of mild cognitive impairment.

Download Full-text

Machine-Learning Algorithms Based on Screening Tests for Mild Cognitive Impairment

American Journal of Alzheimer s Disease & Other Dementias® ◽

10.1177/1533317520927163 ◽

2020 ◽

Vol 35 ◽

pp. 153331752092716

Author(s):

Jin-Hyuck Park

Keyword(s):

Machine Learning ◽

Cognitive Impairment ◽

Mild Cognitive Impairment ◽

Test Data ◽

Learning Algorithms ◽

Test System ◽

Machine Learning Algorithms ◽

Screening Tools ◽

Screening Tests ◽

Data Set

Background: The mobile screening test system for mild cognitive impairment (mSTS-MCI) was developed and validated to address the low sensitivity and specificity of the Montreal Cognitive Assessment (MoCA) widely used clinically. Objective: This study was to evaluate the efficacy machine learning algorithms based on the mSTS-MCI and Korean version of MoCA. Method: In total, 103 healthy individuals and 74 patients with MCI were randomly divided into training and test data sets, respectively. The algorithm using TensorFlow was trained based on the training data set, and then its accuracy was calculated based on the test data set. The cost was calculated via logistic regression in this case. Result: Predictive power of the algorithms was higher than those of the original tests. In particular, the algorithm based on the mSTS-MCI showed the highest positive-predictive value. Conclusion: The machine learning algorithms predicting MCI showed the comparable findings with the conventional screening tools.

Download Full-text

Evidence for the Utility of Actuarial Neuropsychological Criteria Across the Continuum of Normal Aging, Mild Cognitive Impairment, and Dementia

Journal of Alzheimer s Disease ◽

10.3233/jad-200778 ◽

2020 ◽

Vol 78 (1) ◽

pp. 371-386

Author(s):

Lisa V. Graves ◽

Emily C. Edmonds ◽

Kelsey R. Thomas ◽

Alexandra J. Weigand ◽

Shanna Cooper ◽

...

Keyword(s):

Cognitive Impairment ◽

Mild Cognitive Impairment ◽

Diagnostic Errors ◽

Normal Aging ◽

Lower Proportion ◽

Diagnostic Methods ◽

Consensus Methods ◽

Data Set ◽

Cognitively Normal ◽

The Continuum

Background: Research suggests that actuarial neuropsychological criteria improve the accuracy of mild cognitive impairment (MCI) diagnoses relative to conventional diagnostic methods. Objective: We sought to examine the utility of actuarial criteria relative to consensus diagnostic methods used in the National Alzheimer’s Coordinating Center (NACC) Uniform Data Set (UDS), and more broadly across the continuum of normal aging, MCI, and dementia. Methods: We compared rates of cognitively normal (CN), MCI, and dementia diagnoses at baseline using actuarial versus consensus diagnostic methods in 1524 individuals from the NACC UDS. Results: Approximately one-third (33.59%) of individuals diagnosed as CN and more than one-fifth (22.03%) diagnosed with dementia based on consensus methods, met actuarial criteria for MCI. Many participants diagnosed with MCI via consensus methods also appeared to represent possible diagnostic errors. Notably, the CNa/CNc group (i.e., participants diagnosed as CN based on both actuarial [a] and consensus [c] criteria) had a lower proportion of apolipoprotein E ɛ4 carriers than the MCIa/MCIc group, which in turn had a lower proportion of ɛ4 carriers than the dementia (Dem)a/Demc group. Proportions of ɛ4 carriers were comparable between the CNa/CNc and CNa/MCIc, MCIa/MCIc and MCIa/CNc, MCIa/MCIc and MCIa/Demc, and Dema/Demc and Dema/MCIc groups. These results were largely consistent with diagnostic agreement/discrepancy group comparisons on neuropsychological performance. Conclusion: The present results extend previous findings and suggest that actuarial neuropsychological criteria may enhance diagnostic accuracy relative to consensus methods, and across the wider continuum of normal aging, MCI, and dementia. Findings have implications for both clinical practice and research.

Download Full-text

Predicting the Progression of Mild Cognitive Impairment Using Machine Learning: A Systematic, Quantitative and Critical Review

10.1101/2020.09.01.20185959 ◽

2020 ◽

Author(s):

Manon Ansart ◽

Stephane Epelbaum ◽

Giulia Bassignana ◽

Alexandre Bone ◽

Simona Bottani ◽

...

Keyword(s):

Machine Learning ◽

Cognitive Impairment ◽

Clinical Practice ◽

Mild Cognitive Impairment ◽

Parameter Tuning ◽

Methodological Issues ◽

Cognitive Assessments ◽

Data Set ◽

Mixed Effect ◽

The Impact

We performed a systematic review of studies focusing on the automatic prediction of the progression of mild cognitive impairment to Alzheimer's disease (AD) dementia, and a quantitative analysis of the methodological choices impacting performance. This review included 172 articles, from which 234 experiments were extracted. For each of them, we reported the used data set, the feature types, the algorithm type, performance and potential methodological issues. The impact of these characteristics on the performance was evaluated using a multivariate mixed effect linear regressions. We found that using cognitive, fluorodeoxyglucose-positron emission tomography or potentially electroencephalography and magnetoencephalography variables significantly improved predictive performance compared to not including them, whereas including other modalities, in particular T1 magnetic resonance imaging, did not show a significant effect. The good performance of cognitive assessments questions the wide use of imaging for predicting the progression to AD and advocates for exploring further fine domain-specific cognitive assessments. We also identified several methodological issues, including the absence of a test set, or its use for feature selection or parameter tuning in nearly a fourth of the papers. Other issues, found in 15% of the studies, cast doubts on the relevance of the method to clinical practice. We also highlight that short-term predictions are likely not to be better than predicting that subjects stay stable over time. These issues highlight the importance of adhering to good practices for the use of machine learning as a decision support system for the clinical practice.

Download Full-text