Improving the Methodology for Identifying Mild Cognitive Impairment in Intellectually High-Functioning Adults Using the NIH Toolbox Cognition Battery

Cognitive and everyday functioning after bacterial brain abscess: a prospective study of functional recovery from 8 weeks to 1 year post-treatment

OBJECTIVE

A bacterial brain abscess may damage surrounding brain tissue by mass effect, inflammatory processes, and bacterial toxins. The aim of this study was to examine cognitive and functional outcomes at 8 weeks and 1 year following acute treatment.

METHODS

Prospective study of 20 patients with bacterial brain abscess (aged 17-73 years; 45% females) with neuropsychological assessment at 8 weeks and 1 year post-treatment. Behavior Rating Inventory of Executive Function-Adult Version (BRIEF-A) and Patient Competence Rating Scale (PCRS) were used to assess everyday functioning and administered to patients and informants.

RESULTS

Cognitive impairment was found in 30% of patients at 8 weeks and 22% at 1 year. Significant improvements were seen on tests of perceptual reasoning, attention, verbal fluency, and motor abilities (p < 0.05). At 1 year, 45% had returned to full-time employment. Nevertheless, patients and their informants obtained scores within the normal range on measures of everyday functioning (PCRS and BRIEF-A) at 8 weeks and 1 year. No significant improvements on these measures emerged over time.

CONCLUSION

Residual long-term cognitive impairment and diminished work ability affected 22% and 45% of patients one year after BA. Persistent cognitive impairment emphasizes the importance of prompt acute treatment and cognitive rehabilitation.

Estimating the prevalence of Non-Verbal Learning Disability (NVLD) from the ABCD sample

Non-Verbal Learning Disability (NVLD) is a neurodevelopmental disorder characterized by deficits in processing visuospatial information but with age-appropriate verbal skills. This cognitive profile has been hypothesized to be associated with atypical white matter, but at the present there is a lack of evidence for this hypothesis. Currently, the condition is not characterized within the main diagnostic systems, in part because no clear set of criteria for characterizing the disorder exists. This report is the first attempt to estimate NVLD prevalence, using two sets of diagnostic criteria, in a large sample of over 11,000 children who were selected without regards to problems of specific nature, either psychological, neurological, physical and/or social. Furthermore, it examined the association between the profile of cognitive abilities and aspects of whole-brain white matter measures in children with and without symptoms associated with NVLD. Participants were drawn from the Adolescent Brain Cognitive Development (ABCD) study, a 10-year longitudinal study of 11,876 children in the U.S. The data used in the present study were drawn from the initial testing point at which the children were 9-10 years old. Prevalence of NVLD based on two distinct sets of criteria, correlations between the measures used to create the criteria, correlations between criteria measures and measures of white matter integrity. The cognitive criteria included measures of visuospatial processing, reading, intelligence and social skills. By varying the cut-offs applied to social skills in conjunction with visuo-spatial difficulties, spared reading skills and intelligence scores, we calculated prevalence for two NVLD groups. White matter characteristics were measures of volume, fractional anisotropy and mean diffusivity. Based on the criteria used, the estimated prevalence of NVLD varied from 1 to 8%. Furthermore, children with NVLD showed a dissociation between measures of visuo-spatial processing not observed in non-NVLD children. At the neurological level, findings provide preliminary evidence of associations between the cognitive profile of NVLD and abnormalities in white matters tracts. The present study documents that exists, within this large non-selected sample, a proportion of youth who show evidence of NVLD. Given those results, it appears essential to establish the best diagnostic criteria, to improve the treatment options and quality of life for children with this disorder.

Detecting cognitive decline in high-functioning older adults: The relationship between subjective cognitive concerns, frequency of high neuropsychological test scores, and the frontoparietal control network

OBJECTIVE

Neuropsychologists have difficulty detecting cognitive decline in high-functioning older adults because greater neurological change must occur before cognitive performances are low enough to indicate decline or impairment. For high-functioning older adults, early neurological changes may correspond with subjective cognitive concerns and an absence of high scores. This study compared high-functioning older adults with and without subjective cognitive concerns, hypothesizing those with cognitive concerns would have fewer high scores on neuropsychological testing and lower frontoparietal network volume, thickness, and connectivity.

METHOD

Participants had high estimated premorbid functioning (e.g., estimated intelligence ≥75th percentile or college-educated) and were divided based on subjective cognitive concerns. Participants with cognitive concerns (n = 35; 74.0 ± 9.6 years old, 62.9% female, 94.3% White) and without cognitive concerns (n = 33; 71.2 ± 7.1 years old, 75.8% female, 100% White) completed a neuropsychological battery of memory and executive function tests and underwent structural and resting-state magnetic resonance imaging, calculating frontoparietal network volume, thickness, and connectivity.

RESULTS

Participants with and without cognitive concerns had comparable numbers of low test scores (≤16th percentile), p = .103, d = .40. Participants with cognitive concerns had fewer high scores (≥75th percentile), p = .004, d = .71, and lower mean frontoparietal network volumes (left: p = .004, d = .74; right: p = .011, d = .66) and cortical thickness (left: p = .010, d = .66; right: p = .033, d = .54), but did not differ in network connectivity.

CONCLUSIONS

Among high-functioning older adults, subjective cognitive decline may correspond with an absence of high scores on neuropsychological testing and underlying changes in the frontoparietal network that would not be detected by a traditional focus on low cognitive test scores.

Detection of Cognitive Impairment From eSAGE Metadata Using Machine Learning

OBJECTIVE

Using the metadata collected in the digital version of the Self-Administered Gerocognitive Examination (eSAGE), we aim to improve the prediction of mild cognitive impairment (MCI) and dementia (DM) by applying machine learning methods.

PATIENTS AND METHODS

A total of 66 patients had a diagnosis of normal cognition (NC), MCI, or DM, and eSAGE scores and metadata were used. eSAGE scores and metadata were obtained. Each eSAGE question was scored and behavioral features (metadata) such as the time spent on each test page, drawing speed, and average stroke length were extracted for each patient. Logistic regression (LR) and gradient boosting models were trained using these features to detect cognitive impairment (CI). Performance was evaluated using 10-fold cross-validation, with accuracy, precision, recall, F1 score, and receiver operating characteristic area under the curve (AUC) score as evaluation metrics.

RESULTS

LR with feature selection achieved an AUC of 89.51%, a recall of 87.56%, and an F1 of 85.07% using both behavioral and scoring. LR using scores and metadata also achieved an AUC of 84.00% in detecting MCI from NC, and an AUC of 98.12% in detecting DM from NC. Average stroke length was particularly useful for prediction and when combined with 4 other scoring features, LR achieved an even better AUC of 92.06% in detecting CI. The study shows that eSAGE scores and metadata are predictive of CI.

CONCLUSIONS

eSAGE scores and metadata are predictive of CI. With machine learning methods, the metadata could be combined with scores to enable more accurate detection of CI.

Normative Reference Values for Crystallized-Fluid Discrepancy Scores for the NIH Toolbox Cognition Battery

INTRODUCTION

The purpose of this study was to translate NIH Toolbox Cognition Battery (NIHTB-CB) Crystallized-Fluid discrepancy scores into research and clinical practice with adults by providing normative data for discrepancy scores for both age-adjusted standard scores (SSs) and demographically adjusted T-scores.

METHOD

We included adult participants from the NIHTB-CB standardization sample who denied having neurodevelopmental, medical, psychiatric, or neurological conditions (n = 730; M = 47.4 years old, SD = 17.6, range: 18-85; 64.4% women; 63.1% White). Descriptive statistics were calculated for the Fluid and Crystallized composite scores and Crystallized-Fluid discrepancy score, along with correlations between the composite scores and reliability estimates of the discrepancy score. Percentiles were calculated for the discrepancy score, with stratifications by the gender, education, and Crystallized composite for the age-adjusted SSs and demographically adjusted T-scores (T).

RESULTS

Crystallized-Fluid discrepancy scores ranged from -40 to 44 (M = -0.63, SD = 14.89, Mdn = -1, interquartile range [IQR]: -11 to 10) for age-adjusted SSs and from -29 to 27 (M = -0.39, SD = 10.49, Mdn = -1, IQR = -8 to 7) for demographically adjusted T-scores. Crystallized-Fluid discrepancy scores of SS = 15 and T = 11 were at the 16th percentile (1 SD below the mean) and discrepancy scores of SS = 21 and T = 15 were at the 7th percentile (1.5 SD below the mean).

CONCLUSIONS

Crystallized-Fluid discrepancy scores may be, with future research, a useful within-person interpretive approach for detecting a decline from pre-injury or pre-disease levels of cognitive functioning. These normative reference values assist clinicians and researchers in determining the frequency at which given Crystallized-Fluid discrepancy scores occurred among healthy adults in the normative sample.

Multivariate Base Rates of Low Neuropsychological Test Scores in Cognitively Intact Older Adults with Subjective Cognitive Decline from a Specialist Memory Clinic

OBJECTIVE

To avoid misdiagnosing mild cognitive impairment (MCI), knowledge of the multivariate base rates (MVBRs) of low scores on neuropsychological tests is crucial. Base rates have typically been determined from normative population samples, which may differ from clinically referred samples. The current study addresses this limitation by calculating the MVBR of low or high cognitive scores in older adults who presented to a memory clinic experiencing subjective cognitive decline but were not diagnosed with MCI.

METHOD

We determined the MVBRs on the Kaplan-Baycrest Neurocognitive Assessment for 107 cognitively healthy older adults (M age = 75.81), by calculating the frequency of patients producing n scores below or above different cut-off values (i.e., 1, 1.5, 2.0, 2.5 SD from the mean), stratifying by education and gender.

RESULTS

Performing below or above cut-off was common, with more stringent cut-offs leading to lower base rates (≥1 low scores occurred in 84.1% of older adults at -1 SD, 55.1% at -1.5 SD, and 39.3% at -2 SD below the mean; ≥1 high scores occurred in 80.4% of older adults at +1 SD, 35.5% at +1.5 SD, and 16.8% at +2 SD above the mean). Higher education was associated with varying base rates. Overall, the MVBR of obtaining a low cognitive test score was higher in this clinic sample, compared with prior studies of normative samples.

CONCLUSIONS

MVBRs for clinically referred older adults experiencing memory complaints provide a diagnostic benefit, helping to prevent attributing normal variability to cognitive impairment and limiting false positive diagnoses.