| Background:Assessment of cognitive reserve can identify the susceptible population of cognitive frailty early,which is particularly important for advancing the nursing intervention window and promoting the scientific and refined evaluation of nursing intervention effects.The traditional cognitive reserve assessment scales can assess cognitive reserve statically,but it is limited by the measurement time interval,cannot be measured dynamically,and is susceptible to the influence of the evaluator.Neuroimaging assessment methods can dynamically assess cognitive reserve,but there are some problems such as high costs and immobility of the instruments.Therefore,it is necessary to explore portable and cost-effective cognitive reserve assessment methods for the older adults with cognitive frailty from the static and dynamic perspectives.Objectives:(1)To select and introduce the recommended cognitive reserve scale and validate it in the Chinese population with cognitive frailty.(2)To verify the feasibility of Functional Near Infrared Spectroscopy(fNIRS)as an assessment tool for cognitive reserve.Subsequently,to build and evaluate cognitive reserve assessment models for older adults with cognitive frailty using resting and task state fNIRS.Methods:This study consisted of three parts.Part Ⅰ focused on the selection of cognitive reserve assessment scales.Literature research method was used to systematically search literature on cognitive reserve assessment scales.2018 Version of Consensus-Based Standards for the Selection of Health Measurement Instruments(COSMIN)was utilized to evaluate the methodological quality(i.e.,risk of bias)of the included studies and the quality of the measurement properties of the cognitive reserve assessment scales.Based on the guidelines for grading the quality of evidence,recommendations for the optimal cognitive reserve assessment scales were formed by considering their interpretability and applicability.Part Ⅱ focused on the introduction,reliability,and validity study of cognitive reserve assessment scale.Based on the final recommendations,a cognitive reserve questionnaire suitable for older adults was translated into Chinese.After cultural adaptation,eight experts were invited to evaluate the content validity.Then,a convenience sample of 231 older adults with cognitive frailty from the Fangzhuang community in Beijing was selected to validate the convergent validity,known-groups validity,and test-retest reliability of the questionnaire.Part Ⅲ was the construction and evaluation of the cognitive reserve assessment model based on fNIRS in older adults with cognitive frailty.This cross-sectional study recruited 71 older adults with cognitive frailty from a community in Beijing.The fNIRS was used to measure brain function during resting state(8 minutes),Stroop color-word naming task,and n-back working memory task(n=0,1,2).Correlation analysis was performed on the 60-channel fNIRS,and functional connectivity strength between channels was calculated using Z transformation.The brain network analysis based on graph theory was used to calculate the global network and node metrics.General linear models were used to estimate degree of brain activation(β values)caused by cognitive tasks.Finally,independent sample t test and FDR multiple comparison corrections were used to compare the differences in brain network metrics and β values between the high and low cognitive reserve groups.Based on this,the dataset was randomly divided into two sets,with 70%used to construct the training set and 30%used to form the validation set.Support vector machine was used to build the cognitive reserve assessment models of the older adults with cognitive frailty in resting state,task state,as well as the combined model.The performance of each model was evaluated using accuracy,precision,recall(sensitivity),specificity,F1 score and area under the receiver operating characteristic curve.ResultsPart Ⅰ:Seven articles related to cognitive reserve assessment scales for older adults were included in the study,including five different cognitive reserve assessment scales.Based on the summary of measurement properties,interpretability,and feasibility,five scales were all rated as "B" levels.After a comprehensive comparison,the Cognitive Reserve Index questionnaire(CRIq),which has a shorter duration and simpler calculation,was recommended for introduction and usage.Part Ⅱ:The item-level content validity index(I-CVI)of CRIq ranged from 0.875 to 1.The universal agreement S-CVI was 0.82,and the average S-CVI was 0.98.Statistical significance was observed in the standardized coefficients obtained through partial least squares regression,with all three dimensions having an estimated coefficient of 0.44(P<0.001).The cognitive reserve index could be regarded as the formative second-order factor of the three dimensions of education,working activities and leisure activities.The Chinese version of CRIq exhibited good convergent validity,as evidenced by the significant correlation coefficients of 0.68,0.62,and 0.67(P<0.001)with verbal intelligence,performance intelligence,and full-scale intelligence,respectively.Knowngroups validity verified the hypothesis of the Chinese version of CRIq in terms of sex(P values were 0.005,0.002 and 0.002,respectively).The test-retest reliability of the Chinese version of CRIq was 0.85(P<0.001).Part Ⅲ:Seventy-one older people with cognitive frailty completed the fNIRS test.According to the median scores of the CRIq,they were divided into a high cognitive reserve group(n=36)and a low cognitive reserve group(n=35).(1)Resting state:Compared with the low cognitive reserve group,the high cognitive reserve group showed significantly increased functional connectivity.When the sparsity was 0.05,the global efficiency of the high cognitive reserve group(0.15±0.02)was significantly higher than that of the low cognitive reserve group(0.13±0.02)(P<0.001),and the characteristic path length of the high cognitive reserve group(6.70±1.06)was significantly shorter than that of the low cognitive reserve group(7.57±.02)(P<0.001).The standardized characteristic path length of the high cognitive reserve group(1.3 1±0.25)was significantly shorter than that of the low cognitive reserve group(1.46±0.28)(P<0.001).(2)Task state:During the Stroop task,the dorsolateral prefrontal cortex of the low cognitive reserve group showed significantly higher activation compared to the high cognitive reserve group(P<0.05).In the 0-back task,the high cognitive reserve group exhibited significantly higher activation in the frontal eye field,primary motor cortex,and somatosensory association cortex compared to the low cognitive reserve group(P<0.05).During the 1-back task,the high cognitive reserve group showed significantly higher activation in the somatosensory association cortex compared to the low cognitive reserve group(P<0.05).During the 2back task,compared with the low cognitive reserve group,the high cognitive reserve group showed significant activation in the frontal eye field,premotor area,supplementary motor cortex,primary motor cortex,and somatosensory association cortex(P<0.05).(3)Classification assessment model:The assessment models based on the significant activated β values obtained from combining the Stroop,0-back,and 1-back tasks(duration:657 seconds)achieved the best performance with the shortest time for cognitive reserve classification of older adults with cognitive frailty,with an accuracy of 0.727,precision of 0.545,recall(sensitivity)of 0.857,F1 score of 0.667,and specificity of 0.667.The area under the receiver operating characteristic curve was 0.829.Conclusions:(1)Cognitive Reserve Index questionnaire was recommended to introduce to further validate the measurement properties.(2)The Chinese version of CRIq has good reliability and validity in community-dwelling older adults with cognitive frailty,making it suitable for assessing cognitive reserve in this population.(3)Resting and task-state fNIRS can be used to assess cognitive reserve in older adults with cognitive frailty.The prefrontal and superior parietal cortex were used as optode placement areas to allocate brain regions during the Stroop,0-back,and 1-back tasks,and support vector machine was used to classify cognitive reserve.To provide a comprehensive,accurate,real-time,and non-invasive assessment of cognitive reserve in community-dwelling older individuals with cognitive frailty,broaden the assessment scenario of cognitive reserve of community nurses,and provides novel strategies and perspectives for evaluating the effectiveness of precision nursing interventions. |
PDF Full Text Request