Does the Cognitive Change Index Predict Future Cognitive and Clinical Decline? Longitudinal Analysis in a Demographically Diverse Cohort

Abstract

Background:

The Cognitive Change Index (CCI) is a widely-used measure of self-perceived cognitive ability and change. Unfortunately, it is unclear if the CCI predicts future cognitive and clinical decline.

Objective:

We evaluated baseline CCI to predict transition from normal cognition to cognitive impairment in nondemented older adults and in predementia groups including, subjective cognitive decline, motoric cognitive risk syndrome, and mild cognitive impairment. Different versions of the CCI were assessed to uncover any differential risk sensitivity. We also examined the effect of ethnicity/race on CCI.

Methods:

Einstein Aging Study participants (N = 322, M_age = 77.57±4.96, % female=67.1, M_education = 15.06±3.54, % non-Hispanic white = 46.3) completed an expanded 40-item CCI version (CCI-40) and neuropsychological evaluation (including Clinical Dementia Rating Scale [CDR], Montreal Cognitive Assessment, and Craft Story) at baseline and annual follow-up (M_{follow - up}=3.4 years). CCI-40 includes the original 20 items (CCI-20) and the first 12 memory items (CCI-12). Linear mixed effects models (LME) and generalized LME assessed the association of CCI total scores at baseline with rate of decline in neuropsychological tests and CDR.

Results:

In the overall sample and across predementia groups, the CCI was associated with rate of change in log odds on CDR, with higher CCI at baseline predicting faster increase in the odds of being impaired on CDR. The predictive validity of the CCI broadly held across versions (CCI-12, 20, 40) and ethnic/racial groups (non-Hispanic black and white).

Conclusions:

Self-perception of cognitive change on the CCI is a useful marker of dementia risk in demographically/clinically diverse nondemented samples. All CCI versions successfully predicted decline.

Keywords

Alzheimer’s disease cognitive change index ethnic/racial minoritized groups mild cognitive impairment motoric cognitive risk syndrome non-Hispanic Black older adults normal aging subjective cognitive decline

INTRODUCTION

Many individuals notice changes in cognition as they age [1]. A growing body of literature supports that older adults who report decline in cognitive ability in comparison with a prior level of functioning, even in the context of normal performance on neuropsychological test measures, are at an elevated risk for future cognitive decline and dementia [1 –3]. Referred to as subjective cognitive decline (SCD), this presentation may be one of the earliest symptoms on the Alzheimer’s disease and related disorders (ADRD) continuum [1 , 4]. For this reason, and to contextualize the extent of cognitive change relative to an individual’s baseline functioning, subjective cognitive concerns (SCC) are frequently included in criteria for preclinical dementia syndromes including mild cognitive impairment (MCI) (i.e., Petersen Criteria [4]; Diagnostic and Statistical Manual of Mental Disorders, 5th edition, [DSM-5] criteria [5]) and motoric cognitive risk syndrome (MCR) [6].

Although decline in self-perceived cognitive functioning may be related to an underlying neurodegenerative process, many individuals with SCC ultimately do not progress to dementia [1]. Other factors contributing to SCC may include mood or sleep disturbances, medical conditions, medication or substance use, personality, as well as sociodemographic variables including age, sex/gender, educational attainment, household income, and ethnic/racial background [1 , 7–12]. Given that such factors may confound the link between SCC and dementia risk, there has been debate in the MCI field (and to a lesser extent the MCR field [13]) about whether SCC may contribute to misdiagnosis of such predementia conditions [13 –15]. Furthermore, ethnic/racial minorities are highly underrepresented in aging and dementia research in the United States [16]. Thus far, research on SCC in ethnically/racially diverse groups is limited, and findings have been mixed; some studies demonstrate an association between SCC and current or incident cognitive impairment in ethnic/racial minority groups [17 –19], while others do not [20 –23]. This ambiguity surrounding SCC in diverse populations may compound health disparities in dementia prevention and treatment.

The Cognitive Change Index (CCI) [24, 25] is a widely used tool to evaluate perceived decline in various cognitive domains. The CCI includes both self-report and informant report forms, and asks individuals to rate their current level of cognitive ability compared to 5 years ago on a 5-point Likert scale, ranging from normal ability (no change or better) to severe problems (much worse than 5 years ago). Several versions of the CCI are currently utilized in research settings around the world (e.g., [24 –27]), including a 20-item version, which queries concerns across memory (12 items), executive functioning (5 items), and language (3 items) domains [25]. A 12 item CCI version, which comprises only episodic memory items, is a key measure of “significant memory concern” in the Alzheimer’s Disease Neuroimaging Initiative (ADNI) cohort [28], and was also included in the Women’s Health Initiative (WHI) [29] and the Multi-Ethnic Study of Atherosclerosis and Alzheimer’s Disease (MESA-MIND) [23]. Research utilizing the CCI has demonstrated its association with objective cognitive abilities [25, 30], AD biomarkers [28 , 32], mesial temporal lobe volume [24 , 34], and alterations in functional connectivity [35 –37], even in older adults without objective cognitive decline or dementia. Despite these associations, the utility of the CCI to predict future cognitive decline [23, 34], incident dementia is not well-established. Furthermore, conclusions from the current literature may be somewhat limited due to the use of differing CCI versions across studies. At present, it remains unclear which version of the CCI (memory items alone or items across broad cognitive domains) may most strongly predict future cognitive decline in healthy older adults and those with predementia diagnoses like SCD, MCR, and MCI. Finally, studies utilizing the CCI in ethnically/racially diverse aging populations are lacking, limiting the characterization of CCI in key health disparity groups at risk for dementia.

To address some of these important gaps, the CCI-20 [25] was expanded to 40 items (CCI-40) [27], broadening the cognitive domains assessed to now include both retrospective and prospective memory, attention/concentration, visuospatial functioning, mental efficiency, calculation, and orientation, in addition to executive functioning and language. The overarching objective of the present study was to determine how baseline CCI scores predict the transition from cognitive normality to cognitive impairment (as measured on the Clinical Dementia Rating Scale [CDR]) and decline in cognitive functioning (measured on neuropsychological tests).

Specifically, we sought to:

Aim 1. Baseline CCI. Cross-sectionally characterize three versions of the CCI self-report (CCI-12, 20, and 40) at baseline in a general, systematically recruited, community sample of nondemented older adults and within specific predementia groups that are often defined using self-perceived cognitive decline as diagnostic criteria (e.g., SCD, MCR, and MCI).

Aim 2. Baseline CCI to predict cognitive and clinical decline. Determine the predictive value of the CCI at baseline for change in risk of cognitive impairment and decline in objective measures of cognitive functioning during follow-up.

Secondary aim: Cross-sectional and longitudinal CCI Performance in non-Hispanic white and non-Hispanic black Participants. Given the lack of studies of the CCI in non-white older adults, we included a secondary aim to examine ethnic/racial differences in CCI at baseline and over follow up in non-Hispanic white (N-HW) and non-Hispanic black (N-HB) participants.

MATERIALS AND METHODS

Participants and procedures

This study included participants from the Einstein Aging Study (EAS), a longitudinal study of community-residing older adults in Bronx, New York, a demographically diverse urban setting. EAS participants are recruited systematically using Bronx County Voter Registration lists. Potential EAS participants are mailed introductory letters and called to complete an initial telephone screening. Inclusion criteria are age 70 and above, resident of Bronx, New York, noninstitutionalized, and English speaking. Exclusion criteria are the presence of active psychiatric symptomatology and/or perceptual (e.g., visual/auditory) impairments that could interfere with assessments and testing and non-ambulatory conditions [38]. All protocols were approved by the Einstein Institutional Review Board (IRB) and participants provided written informed consent at baseline assessment. Participants in the present study were free of dementia at baseline assessment and were administered the CCI-40 [26, 27]. Analyses were based on the longitudinal administration of the CCI-40 and neuropsychological measures. The CCI-40 was administered as part of the comprehensive in-person assessments that were conducted annually and included neurological, medical, psychosocial, and neuropsychological examinations (see [39] for a description of EAS methods and procedures). The EAS began including the CCI-40 in all annual in-person assessments in 2016.

Diagnostic classification

Clinical diagnosis was determined based on a consensus case conference attended by a study neurologist, neuropsychologist, and nurse practitioner at baseline and following each annual study visit. Dementia was defined according to Diagnostic and Statistical Manual, Fourth Edition criteria [40], and those with dementia at the baseline visit were excluded from this analysis (for details related to dementia diagnosis in EAS, see [38]). See Table 1 for a depiction of the criteria used to classify study participants into predementia diagnostic groups.

Table 1

Criteria used to classify study participants into diagnostic groups

Healthy Control	Subjective	Motoric Cognitive	Mild Cognitive
(HC)	Cognitive Decline	Risk Syndrome	Impairment
(SCD)	(MCR)	(MCI)
Slow Gait Speed			+
Objective Cognitive Decline				+
Self-Perceived Cognitive Decline^*		+	^*	^*
Normal Activities of Daily Living	+	+	+	+
No dementia	+	+	+	+

^*Self-perceived cognitive decline was defined as a response of “yes and this worries me” to the question “Do you feel like your memory is becoming worse?” for SCD. In MCR, SCC was defined as any participant who completed the CCI, as this was our variable of interest. In MCI, self-perceived cognitive decline was not required for diagnosis using Jak/Bondi criteria, but is utilized in other commonly used MCI diagnostic criteria (i.e., [4, 5]).

Motoric Cognitive Risk Syndrome (MCR). The diagnosis of MCR is based on established criteria [6] and includes: 1) slow gait, defined as ≥1 standard deviation (SD) below age and sex-appropriate mean values established within the same cohort. To capture gait speed (in centimeters per second), a four-meter walk was measured at usual pace in a quiet, well-lit room; 2) Preserved activities of daily living on the Lawton Brody scale, indicating independence on instrumental activities items [41]; 3) Absence of dementia diagnosis based on consensus case conference [38]; 4) Because subjective cognitive concerns as reported on the CCI were the focus of this study, we included any participant who met other MCR criteria and completed the CCI. Of note, MCR diagnosis does not exclude SCD nor MCI status.

Mild Cognitive Impairment (MCI). The classification of participants as MCI was based on Jak/Bondi criteria [15]. Ten neuropsychological measures that are part of the EAS neuropsychological battery measuring five cognitive domains were considered for this classification including: 1) Memory: Free recall from the Free and Cued Selective Reminding Test, Benson Complex Figure (Delayed); 2) Executive Function: Trail Making Test Part B (limit time 300 s), Phonemic Verbal Fluency (Letters F and L: 1 min each); 3) Attention: Trail Making Test Part A (limit 150 s), Number Span (forward and backwards); 4) Language: Multilingual Naming Test (MINT, total score), Category Fluency (Animals, Vegetables: 1 min each); 5) Visual-spatial: Benson Immediate Recall, WAIS III Block Design. Note that the cognitive change outcome measures in this study (described below) were not used to classify participants as MCI. MCI status was determined using the baseline neurocognitive measures. The following formula was used to assign group classification: 1) impaired scores, defined as > 1 SD below the age, sex and education adjusted normative mean, on both measures within at least one cognitive domain (i.e., memory, language, or speed/executive function); or 2) one impaired score, defined as > 1SD below the age, sex, and education adjusted normative mean, in each of three of the five cognitive domains assessed; or 3) a score of 4 on the Lawton Brody scale, indicating dependency on all 4 instrumental activities items [41]. MCI diagnosis does not exclude MCR status but does exclude SCD.

Subjective Cognitive Decline (SCD). Participants were classified as SCD if they did not meet criteria for MCI and responded “yes, and this worries me” to the CCI item “Do you feel like your memory is becoming worse?”, with other possible responses being “no” and “yes, but this does not worry me.” The use of this item to define SCD is consistent with the SCD plus criteria proposed by the Subjective Cognitive Decline Initiative (SCD-I) Working Group [2] which indicates that worries associated with SCD and SCD in memory specifically (rather than in other domains of cognition) increases the likelihood of association with preclinical Alzheimer’s disease [42, 43]. SCD diagnosis does not exclude MCR status. Participants who endorsed “no” or “yes, but this does not worry me” were eligible to be classified as HC, MCR, or MCI.

Healthy Control (HC). Participants were classified as Healthy Control (HC) if they did not meet criteria for SCD, MCR, or MCI.

Measures of interest

The Cognitive Change Index (CCI). We investigated three versions of the self-report CCI as well as each individual CCI item. These versions included: the CCI-20, the original 20 item CCI [25]; the CCI-40, an expanded 40-item version [27], which contains the original 20 items with 20 additional items; and the CCI-12, which includes the 12 episodic memory items from the CCI-20 that were summed to identify participants with “significant memory concerns” in the ADNI study [28]. The CCI-40 includes all the CCI items examined in this study, with the CCI-20 and CCI-12 being extracted from the longer CCI-40 version for analyses. The number of items within each cognitive domain on the CCI versions, in addition to sample CCI items, are presented in Table 2. Of note, an informant report version of the CCI exists [25] but was not administered in the EAS cohort.

Table 2

Number of items and sample items within each cognitive domain among CCI Versions: CCI-12, CCI-20, and CCI-40

Cognitive domain (item count)	CCI-12	CCI-20	CCI-40	Sample items
Memory	12	12	21	“Remembering the names and faces of new people I meet” “Remembering to take my medication on time”
Executive Functioning	–	5	6	“Making decisions about everyday matters” “Reasoning through a complicated problem”
Language Items	–	3	4	“Coming up with the word I want to use” “Expressing myself when speaking”
Attention/Concentration	–	–	3	“Focusing on more than one thing at once” “Paying attention to what is going on around me”
Visuospatial Items	–	–	2	“Navigating to locations” “Finding my way around familiar places”
Mental Clarity/Efficiency	–	–	2	“Thinking quickly” “Thinking clearly”
Calculation	–	–	1	“Solving everyday math problems”
Orientation	–	–	1	“Recalling the date without looking it up”

The CCI-40 includes 40 self-report items that probe concerns across various cognitive domains, including memory (21 items), executive functioning (6 items), language (4 items), attention/concentration (3 items), visuospatial (2 items), mental clarity/efficiency (2 items), orientation (1 item), and calculation (1 item). For all CCI versions, participants rate their current level of ability (approximately over the past month) compared to 5 years ago on a 5-point Likert scale: 1 = normal ability (no change or better than 5 years ago), 2 = slight/occasional problem (minimal change from 5 years ago), 3 = mild problem (some change from 5 years ago), 4 = moderate problem (clearly noticeable change from 5 years ago), and 5 = severe problem (much worse than 5 years ago). Total scores on the CCI range from 12–60 on the CCI-12, 20–100 on the CCI-20, and 40–200 on the CCI-40, with higher scores indicating more significant self-perceived cognitive difficulties/decline. There is an additional CCI item that precedes the Likert-type scale items and asks participants the following: “Do you feel like your memory is becoming worse?” with response options “No”/“Yes, but this does not worry me”/“Yes and this worries me” [44]. This item is not included in the total score and was used to classify individuals as having SCD (as discussed above).

Cognitive and Clinical Outcomes. The Clinical Dementia Rating Scale (CDR) [45] is a global rating scale for the study of individuals across the dementia spectrum. In the EAS, the CDR score is based upon a semi-structured interview with the participant and the clinical judgement of study clinician. The CDR is constructed of clinically relevant cognitive and behavioral domains such as memory, orientation, judgment and problem solving, community affairs, home and hobbies performance, and personal care. The CDR score is tabulated on a scale of 0–3, ranging from no dementia (CDR = 0), questionable dementia (CDR = 0.5), mild dementia (CDR = 1), moderate dementia (CDR = 2), to severe dementia (CDR = 3). For the purposes our analyses, we generated dichotomized CDR scores representing cognitively impaired and non-cognitively impaired CDR performances, with CDR scores of 0 being classified as non-cognitively impaired and CDR scores≥0.5 classified as cognitively impaired.

In addition to the CDR, participants completed a comprehensive neuropsychological assessment (a modified version of the National Alzheimer’s Coordinating Center Uniform Data Set Neuropsychological Battery [46, 47]), which included a cognitive screen (The Montreal Cognitive Assessment [MoCA]) and a test of learning/memory (Craft Story 21). The MoCA and Craft Story were selected as cognitive outcomes to avoid diagnostic circularity, given that these measures were not used to assign MCI diagnosis. The MoCA [48] is a rapid screening instrument for mild cognitive dysfunction. It assesses different cognitive domains: attention and concentration, executive functions, memory, language, visuo-constructional skills, conceptual thinking, calculations, and orientation. The total possible score is 30 points, with higher scores representing better performance.

Craft Story 21 [49] immediate and delayed recall assesses verbal memory by examining an individual’s ability to encode and recall a short story. Participants are asked to listen to a short story and to repeat it back to the examiner (Craft Story 1). After a 20-minute delay, participants are once again asked to recall every detail that they can from the story (Craft Story 2). Participants are scored on their ability to produce details from the story, with higher scores reflecting better recall (scores range from 0 to 25 for both Craft Story 1 and Craft Story 2).

Covariates

Demographic information included self-reported ethnicity/race as defined by the U.S. Census Bureau in 1994, number of years of education, biological sex, and age at baseline. An adjusted version of the Geriatric Depression Scale (GDS, short form) total score at baseline was used to screen for depressive symptoms. The GDS contains dichotomous (yes/no) items and when summed, the total scores range from 0 to 15, with scores of six or above suggestive of clinically significant depressive symptoms [50]. For this study, we excluded GDS item 10 (“Do you feel you have more problems with your memory than most?”) from the total GDS score, as this item is commonly used in MCR studies to assess self-perceived cognitive functioning [13].

Statistical analysis

Aim 1. Baseline CCI. Baseline descriptive characteristics were reported for the CCI, sociodemographic variables, and cognitive measures in the overall sample, by diagnostic group status, and by ethnicity/race. At baseline, ANOVA, linear regression for continuous measures, and Chi-square test, logistic regression, or multinomial logistic regression for categorical measures were used to assess the association of demographic characteristics, depressive symptoms, and objective cognitive functioning with CCI in the overall sample, and in predementia and ethnic/racial subgroups. Of note, all total score CCI analyses were conducted separately using the CCI-12, CCI-20, and CCI-40. We also present Pearson correlation coefficient (continuous variables) and Spearman rank correlation coefficient (binary variables) to demonstrate the association between CCI and variables of interest.

Aim 2. Baseline CCI to predict cognitive and clinical decline. We used linear mixed effects models (LME) for continuous, and generalized LME for binary longitudinal cognitive outcomes, to examine the association of CCI total scores (CCI-12, CCI-20, and CCI-40) at baseline with rate of decline in cognition (measured on the MoCA and Craft Story) and transition to cognitive impairment (measured on dichotomized CDR scores [described above]) in the overall sample and within each diagnostic group (normal, SCD, MCR and MCI) over follow-up. Each CCI total score or item, time, and their interaction were included in the model as predictors, adjusting for covariates including age, sex, education, ethnicity/race, and adjusted GDS score. The interaction term, which estimates the effect of CCI scores on the rate of change in cognition over follow-up, was reported. Separate models for each cognitive outcome of interest were used. These analyses were also conducted in each ethnic/racial demographic group as a secondary analysis (ethnicity/race was removed as a covariate in these models). A false discovery rate (FDR) correction was applied to correct for multiple comparisons within the primary analysis in the overall sample. To show the added utility of the CCI to predict CDR decline above and beyond the MoCA, we repeated the above described generalized LME model in the overall sample with the additional covariate of the MoCA. We also performed a sensitivity analysis using cox proportional hazards models among a subsample of particpants who were CDR = 0 at baseline to examine the association of CCI at baseline with the time to CCI≥0.5 (survival outcome) over follow up An α value of 0.05 was used for determining statistical significance for all hypothesis-driven analyses. SPSS version 25 software (IBM Corp., Armonk, NY, USA) and SAS 9.4 (SAS Institute Inc., Cary, NC, USA) were utilized.

RESULTS

Overview

Participants’ (N = 322) age at baseline ranged from 70.4 to 93.7 (mean = 77.57, SD = 4.96) years. The sample was 67.1% female and educational achievement averaged 15.06±3.54 years. The sample was 46.3% non-Hispanic white (N-HW), 40.1% non-Hispanic black (N-HB), with the remaining participants combined into an “Other” ethnic/racial group (13.7%) included 31 Hispanic white (9.6%), 9 Hispanic black (2.8%), 3 Asian (0.9%), and 1 (0.3%) “more than 1 race” participant. At baseline, a total of 57 participants met criteria for SCD, 42 met criteria for MCR, 99 met criteria for MCI, and 126 participants were classified as HC (normal cognition and gait). Consistent with previous research, which shows an overlap between MCR and MCI diagnoses [6], 20 participants met criteria for both MCR and MCI (47.6% of the MCR group). MCR and SCD diagnoses also overlapped (n = 6, 14.3% of MCR). Due to this diagnostic overlap, differences between diagnostic groups were not tested. Baseline characteristics of the sample are presented in Table 3.

Table 3

Baseline Characteristics in the overall sample and by diagnostic groups

	Total Sample		HC		SCD		MCR		MCI
N	322		126		57		42		99
Demographic Variables	Mean	SD	Mean	SD	Mean	SD	Mean	SD	Mean	SD
Age	77.576	4.9635	76.789	4.3136	76.126	4.1253	78.01	5.5758	78.64	5.142
Education	15.06	3.541	15.06	3.724	15.49	3.757	15.21	3.418	14.91	3.274
% female	67.1		66.7		61.4		66.7		67.7
% n-H white	46.3		50.8		49.1		42.9		35.4
% n-H black	40.1		35.7		36.8		50		50.5
% other	13.7		13.5		14		7.1		14.1
Depression (GDS)	2.0932	1.86558	1.6349	1.29987	2.5088	1.90978	2.7619	2.41751	2.404	2.20352
Gait Speed	98.4013	27.84835	107.0303	23.55995	99.2557	27.01123	58.2891	11.29801	93.1891	28.55289
Neuropsychological
Variables
MoCA	23.66	3.504	24.67	2.805	24.53	3.328	23.07	3.612	21.3	3.579
Craft Story 1	14.81	4.089	15.37	4.023	15.63	3.867	14.48	3.611	13.13	4.115
Craft Story 2	13.27	4.288	13.78	4.087	14.26	4.151	12.62	4.114	11.38	4.137
CCI Total Scores
CCI 40	63.4348	18.26877	58.0794	13.43822	72.7544	17.52272	66.119	22.75447	67.2323	21.67234
CCI 20	34.2267	11.02934	31.2619	8.83283	40.193	10.08967	35.9762	13.15107	36.3232	12.83158
CCI 12	22.4627	7.76766	20.5873	6.61304	26.8596	6.95506	23.5	8.84073	23.7879	8.84478

HC, healthy control; SCD, subjective cognitive decline; MCR, motoric cognitive risk syndrome; MCI, mild cognitive impairment; n-H white, non-Hispanic white; n-H black, non-Hispanic black; other, other ethic/racial groups; GDS, Geriatric Depression Scale; MoCA, Montreal Cognitive Assessment; CCI, Cognitive Change Index. Results for n-H white and n-H black participants are presented in Supplementary Table 2.

Aim 1. Baseline CCI

Summaries of total scores for all 3 versions of the CCI at baseline are presented in Table 3, in the overall sample and by diagnostic group status. The most elevated CCI total scores were observed in participants with SCD, with highly comparable performances observed in both the MCR and MCI groups, while healthy controls showed the lowest CCI scores.

Demographic, cognitive, and depression variable association with CCI-40 total score are presented in Table 4. In the entire sample, higher CCI-40 total score was significantly associated with lower levels of educational achievement (b=−0.91, p < 0.001), higher GDS score (b = 2.174, p < 0.001), and N-HB ethnicity/race (b = 5.690, p = 0.006) and “other” ethnic/racial group (b = 6.604, p = 0.026) status compared to N-HW. In terms of cognition, in the overall sample, higher CCI-40 scores were associated with worse scores on an objective cognitive screen (MoCA, b =−0.955, p = 0.001), immediate memory (Craft Story 1, b =−0.957, p < 0.001), and delayed memory recall (Craft Story 2, b = −0.799, p < 0.001). Correlations between CCI-40 and variables of interest in the overall sample are presented in Supplementary Table 1.

Table 4

Demographic and neuropsychological variable association of CCI-40 at baseline

	Total Sample		HC		SCD		MCR		MCI
Demographic Variables	b	p	b	p	b	p	b	p	b	p
Age	−1.73	ns	0.026	ns	−0.509	ns	−0.359	ns	0.116	ns
Education	−0.911	0.001	−1.102	<0.001	−0.637	ns	−1.572	ns	−0.757	ns
Female	−0.139	ns	−2.595	ns	−1.214	ns	6.714	ns	3.206	ns
n-H black	5.690	0.006	5.165	0.038	4.460	ns	10.714	ns	1.793	ns
“Other”	6.604	0.026	2.900	ns	6.829	ns	2.744	ns	12.708	0.041
Depression (GDS)	2.174	<0.001	1.405	ns	2.576	0.034	0.226	ns	1.538	ns
Neuropsychological
Variables
MoCA	−0.955	0.001	−1.024	0.016	−0.877	ns	−1.876	ns	−0.233	ns
Craft Story 1	−0.957	<0.001	−0.718	0.016	−0.767	ns	−1.759	ns	−0.935	ns
Craft Story 2	−0.799	<0.001	−0.715	0.014	−0.324	ns	−1.930	0.024	−0.943	ns

A similar pattern of results for CCI-20 and CCI-12 were observed and presented in Supplementary Table 1. HC, healthy control; SCD, subjective cognitive decline; MCR, motoric cognitive risk syndrome; MCI, mild cognitive impairment; n-H black, non-Hispanic black; other, other ethic/racial groups; GDS, Geriatric Depression Scale; MoCA, Montreal Cognitive Assessment; CS, Craft Story; CCI, Cognitive Change Index. Results for CCI-40 in non-Hispanic white and black participants are presented in Supplementary Table 3.

In HC participants, high CCI-40 total score was associated with lower education levels (b=−1.102, p < 0.001), N-HB ethnic/racial status (b = 5.165, p = 0.038) compared to N-HW, and worse performance on a cognitive screen (MoCA, b=−1.024, p = 0.016), immediate memory (Craft Story 1, b = 0.718, p = 0.016), and delayed memory (Craft Story 2, b=−0.014, p = 0.014). In participants with SCD, CCI-40 was positively associated with depressive symptoms (b = 2.576, p = 0.034), but CCI-40 total score was not associated with objective cognitive performance. In participants with MCR, higher CCI-40 scores were associated with worse performance on objective cognitive measures of memory recall (Craft Story 2, b =−1.930, p = 0.024), and approached a significant association with a global cognitive screen (MoCA, b =−1.876, p = 0.055). In participants with MCI, CCI-40 scores were not associated with any cognitive measures. Results for CCI-20 and CCI-12 are highly consistent with those observed for CCI-40 and are presented in Supplementary Table 2.

Aim 2. Baseline CCI to predict cognitive and clinical decline

Among the 322 participants, the average number of annual follow-up assessments was 3.4 (median = 4, range 1–7). The longitudinal findings among the CCI versions were highly similar; thus, for simplicity, we present only results for CCI-40 here. Results for CCI-20 and CCI-12 are presented in Supplementary Table 3. Unadjusted p values are presented in Table 5, while FDR adjusted values for the primary analysis in the overall sample are presented here in text. In the overall sample of nondemented older adults, baseline CCI-40 (βinteraction = 0.013, p = 0.004, FDR-adjusted) was significantly associated with rate of change in log odds on CDR, with higher CCI at baseline predicting faster increase in the odds of being impaired defined by CDR (i.e., CDR > 0). To show the added utility of the CCI to predict CDR decline above and beyond other commonly used neuropsychological screeners, we repeated this analysis, additionally adjusting for MoCA score. The CCI-40 remained significantly associated with rate of change in log odds on CDR (βinteraction = 0.014, p < 0.001) (Results of this analysis for CCI-20 and 12 are presented in Supplementary Table 4). We also performed a sensitivity analysis among the subsample of participants who were cognitively normal at baseline (CDR = 0) to examine the association of CCI at baseline incidence of cognitive impairment defined by CDR≥0.5 using cox proportional hazards models, adjusting for age, sex, race/ethnicity, education, and depression. In the overall sample, 161 were CDR = 0 at baseline and 41 of these participants became impaired (CDR≥0.5) over follow up. The CCI-40 remained a significant predicter of risk for incidence of CDR≥0.5 (aHR = 1.041, SE = 0.0101, p < 0.001) (results for CCI-20 and 12 are presented in Supplementary Table 5).

Table 5

Baseline CCI-40 association with rate of decline on neuropsychological variables and clinical dementia status

	MoCA			Craft Story 1			Craft Story 2			CDR
	estimate	stderr	p	estimate	stderr	p	estimate	stderr	p	estimate	stderr	p
Total Sample
CCI-40	−0.00223	0.003058	0.4669	0.003693	0.004256	0.3861	0.001851	0.004341	0.6701	0.01347	0.004043	0.0009^*
HC
CCI-40	−0.00893	0.007918	0.2611	−0.01585	0.01045	0.1314	−0.03252	0.01103	0.0038^*	0.01699	0.008423	0.0446^*
SCD
CCI-40	−0.00208	0.005404	0.7018	0.02023	0.009643	0.0396	0.01586	0.0092	0.0891	0.02971	0.01021	0.0043^*
MCR
CCI-40	0.00309	0.006111	0.615	0.000677	0.007727	0.9306	0.009652	0.007758	0.2209	0.03855	0.01605	0.0188^*
MCI
CCI-40	−0.00732	0.005308	0.1722	−0.0027	0.007213	0.7087	0.000837	0.007006	0.9052	0.00337	0.007612	0.6586

Analyses are adjusted for age, sex, education, ethnicity/race, and adjusted GDS score. Results for CCI-20 and CCI-12 are presented in Supplementary Table 2. HC, healthy control; SCD, subjective cognitive decline; MCR, motoric cognitive risk syndrome; MCI, mild cognitive impairment; MoCA, Montreal Cognitive Assessment; CCI, Cognitive Change Index; CDR, Clinical Dementia Rating Scale.

In terms of neuropsycholical tests in the overall sample, no significant associations were observed between baseline CCI-40 total score and rate of decline on neuropsychological measures in the overall sample; MoCA βinteraction =−0.002, p = 0.6225; Craft Story 1 βinteraction = 0.004, p = 0.6225; or Craft Story 2 βinteraction = 0.002, p = 0.6701 (FDR-adjusted p values).

Analyses were also conducted within each diagnostic group. In the HC participants, baseline CCI-40 (βinteraction = 0.017, p = 0.045) significantly predicted rate of change in log odds on CDR. In terms of cognitive decline, high baseline endorsement on the CCI-40 was associated with rate of decline on memory recall performance (Craft Story 2, βinteraction =−0.032, p = 0.004). In participants with SCD, baseline CCI-40 (βinteraction = 0.029, p = 0.004) significantly impacted rate of change in log odds on a global dementia scale (CDR). No significant associations between CCI-40 and decline on cognitive measures was observed in SCD. In participants with MCR, baseline CCI-40 (βinteraction = 0.038, p = 0.019) significantly impacted rate of change in log odds on a global dementia scale (CDR). No significant associations between the baseline CCI-40 total score was observed with decline on cognitive measures in MCR. In participants with MCI, no significant associations between baseline CCI total scores were observed with decline on CDR or cognitive measures.

Secondary aim: Cross-sectional and longitudinal CCI performance in non-Hispanic white and non-Hispanic black participants

As secondary analyses, we examined CCI performance at baseline and longitudinally by ethnic/racial groups. There were 149 non-Hispanic white (N-HW) participants and 129 non-Hispanic black (N-HB) participants. We limit our secondary analyses on the N-HW and N-HB groups due to the highly heterogenous and small sample of “Other” ethnic/racial population (n = 45; ethnic/racial details described above). The N-HB group had significantly fewer years of education (N-HB m = 14.25 versus N-HW m = 16.20; t(277) = 4.749, p < 0.001), larger proportion of women in the sample (N-HB % female = 76.7 versus N-HW % female = 56.4; χ²(1, N = 279)=12.45, p < 0.001), and slower gait speed (N-HB m = 93.48 versus N-HW m = 103.15; t(238) = 2.6667, p = 0.009) than the N-HW older adults. The N-HB and N-HW groups did not differ in terms of age or depressive symptoms. The N-HB group performed significantly worse on all neuropsychological measures, including a cognitive screen (MoCA: N-HB m = 22.69 versus N-HW m = 24.69; t(276) = 5.192, p < 0.001), and a test of immediate memory (Craft Story 1: N-HB m = 14.10 versus N-HW m = 15.73; t(277) = 3.235, p < 0.001), and delayed memory (Craft Story 2: N-HB m = 12.16 versus N-HW m = 14.46; t(277) = 4.461, p < 0.001). N-HB endorsed higher CCI total scores on all versions, including CCI-40 (N-HB m = 66.85 versus N-HW m = 58.79; t(277) = 3.982, p < 0.001), CCI-20 (N-HB m = 36.56 versus N-HW m = 31.34; t(277)=4.218, p < 0.001), and CCI-12 (N-HB m = 24.09 versus N-HW m = 20.60; t(277)=3.91, p < 0.001). These cross-sectional results are presented in Supplementary Table 6.

At baseline, higher CCI scores in N-HW participants were associated with depressive symptoms (b = 0.022, p = 0.046), and a cognitive screen (MoCA, b =−0.062, p = 0.002). Higher CCI scores in N-HB participants were associated with lower levels of education (b =−0.03, p = 0.031), depressive symptoms (b = 0.024, p = 0.004), immediate memory (Craft Story 1, b =−0.048, p = 0.010), and delayed memory recall (Craft Story 2, b =−0.042, p = 0.031). These cross-sectional associations are presented in Supplementary Table 7.

Longitudinally, in N-HW participants, baseline CCI-40 (βinteraction=0.018, p = 0.039) total score significantly predicted rate of change in log odds for CDR. No significant associations between any of the CCI total scores were observed with decline on cognitive measures in the N-HW group. In N-HB participants, baseline CCI-40 (βinteraction = 0.018, p = 0.005) total score significantly predicted rate of change in log odds on CDR. No significant associations between CCI-40 total score was observed with decline on cognitive measures in the N-HB group. These longitudinal results are presented in Supplementary Table 8.

DISCUSSION

The Cognitive Change Index (CCI) is a widely used measure of self-perceived cognitive functioning for older adults in clinical and research settings [24, 25]. The CCI has informant- and self-report forms, versions of different lengths (12, 20, or 40 items), and versions that capture various cognitive domains (solely memory items versus multiple cognitive domains). Despite its widespread use, little is understood about the association of specific CCI versions with future cognitive decline. The present study fills an important gap by demonstrating that high self-reported CCI endorsement of cognitive concerns at baseline is associated with faster rates of conversion from normal cognitive status to cognitive impairment in a clinically and demographically diverse community sample of older adults initially free of dementia. We documented that the predictive validity of the CCI was similar across all three self-report versions (CCI-12, 20, or 40), and when controlling for key demographic variables (age, education, sex, and ethnicity/race) and depressive symptoms. Overall, results support the inclusion of the CCI in clinical and research protocols to optimize detection of at-risk individuals, including cognitively unimpaired older adults.

The present study builds upon prior research that shows cross-sectional associations between concurrent CCI scores and neuropsychological test performance [25, 31], and a broader literature that demonstrates a modest but significant relationship between subjective cognitive concerns and objective cognitive functioning (see [51] for a review). In our overall sample, high CCI endorsement was associated with worse performance on a general cognitive screen (MoCA) and a learning/memory measure (Craft Story 21) at baseline. However, the association between CCI and objective neuropsychological performance differed by group status. Although the MCR and MCI groups showed similar levels of cognitive concerns, the CCI was associated with worse cognitive performance in MCR, but not in MCI. Given that objective cognitive impairment is a key criterion for MCI diagnosis, but not MCR, this lack of association between CCI and neuropsychological tests in MCI may result from reduced insight in those with cognitive impairment [52]. Indeed, prior research by Rattanabannakit and colleagues [25] showed an advantage for informant-report CCI in predicting neuropsychological performance, over self-report CCI, particularly in those with MCI, suggesting that reduced self-perception of cognitive problems may decrease the association between CCI and objective indicators of cognitive performance [25, 52]. Additionally, a restricted range of cognitive performance in the MCI group may attenuate an association with CCI. In the context of MCR, the association between CCI and neuropsychological tests highlights the promising utility of comprehensive SCC assessment in this syndrome for detecting objective cognitive vulnerabilities. More broadly, there is substantial need for increased research on SCC in MCR, as such concerns are one of the two key criteria for diagnosis, yet have been poorly defined or standardized in the field [13]. To this end, the present research is one of the first studies to comprehensively characterize the CCI in this population.

Although individuals with SCD exhibited higher CCI scores at baseline than those without SCD, this elevated endorsement of cognitive concerns was associated with contemporaneous depressive symptoms, but not objective cognitive performance cross-sectionally. This is unsurprising, given that (a) neuropsychological test scores are, by definition, unimpaired in individuals with SCD [1, 2]; and (b) there is a well-established association between SCC and depressive symptoms [1 , 53]. Indeed, both depression and SCC are independent risk factors for dementia and are also strongly correlated conditions [53, 54]. The co-occurrence of both depression and SCD confers a higher risk for major neurocognitive disorder, than the presence of depression or SCD alone [53]. The role of depression in SCC is controversial, challenging to disentangle, and may depend on study sample. For example, research conducted in “worried well,” self-referring populations suggests that SCCs are most likely indicative of depression [51]. By contrast, in general, community populations, as exemplified by the present and prior reports, SCCs are a risk factor even after adjusting for depressive symptoms and likely represent an independent risk factor [51]. It is also important to note that the level of depressive symptoms in SCD individuals in our sample was well below clinical threshold, suggesting that a diagnosable depressive syndrome was likely not primarily driving their cognitive concerns.

Our longitudinal findings clearly demonstrated that high baseline CCI endorsement was associated with rate of transition from cognitive normality to cognitive impairment over follow-up, controlling for depressive symptoms and demographic covariates. This finding was observed across CCI versions (CCI-12, 20, or 40). This is reassuring, given that various iterations of the CCI are utilized across clinical and research settings, as it shows that all CCI versions may be broadly equally useful in predicting risk in a sample of nondemented older adults. Our study builds on recent research [34], which found that SCD defined by CCI-12 was associated with future cognitive decline, by demonstrating that all CCI versions predict cognitive risk. The CCI appeared to be a worthwhile tool to predict future cognitive decline in participants who were cognitively unimpaired. This suggests that the CCI may be sensitive to very subtle, early cognitive changes in at risk individuals. In the MCI group, baseline CCI endorsement was not associated with cognitive decline. This finding may further serve to underscore the anosognosia for cognitive deficits that may arise as part of cognitive impairment [55, 56], leading to inaccuracies in SCC self-report [52]. Although the CCI versions all performed broadly similarly in predicting decline, the CCI-40 was the only version that was uniformly associated with transition from normal cognition to cognitive impairment across the clinical and demographic groups analyzed (aside from MCI). As such, the CCI-40 may be marginally superior over the more abbreviated versions to capture cognitive risk, depending on study or clinical setting and/or time allotted for SCC assessment (given that the CCI-40 is double the length of CCI-20). Future research should explore if CCI cognitive domain sub-scores (i.e., executive functioning items) or if certain CCI items may provide additional utility to create a brief, targeted CCI screen optimized for risk detection.

We also investigated the CCI’s association with demographic variables, specifically ethnicity/race, which have shown an association with self-perception of cognition functioning in prior research [1 , 58]. Shifting demographic trends predict larger percentages of older adults from ethnic/racial minoritized groups in the United States [58] and there is disproportionate burden of cognitive impairment and dementia in such groups [59]. These disparities are compounded by the systemic underrepresentation of such populations in aging studies [16, 59], as such, research investigating the role of SCCs in dementia risk in diverse populations is an essential step toward brain health equity. In our overall sample, high endorsement on the CCI was associated with ethnicity/race minority status and lower level of education. We further explored the CCI in non-Hispanic black (n = 129) older adults. N-HB participants showed significantly higher scores on the CCI at baseline (compared with N-HW), consistent with recent findings from a Centers for Disease Control and Prevention report, which also showed disparity in SCD burden across ethnic/racial groups [57]. High CCI scores in N-HB older adults were associated with depressive symptoms, lower education, and worse memory performance. Further, baseline cognitive concerns expressed on the CCI in this ethnic/racial group predicted the change of likelihood to cognitive impairment based on CDR over follow up as well as incidence of cognitive impairment based on CDR in the overall sample. Our results add to the growing literature on SCC in ethnic/racially diverse older adult populations. Recent research has highlighted the interaction between factors such as discrimination, marginalization, and socioeconomic status, which result in an “accelerated aging” phenotype, ultimately driving SCC disparities and dementia risk in minoritized populations [60, 61]. The current results are consistent with prior studies which have found SCC to confer similar or greater levels of dementia risk in minoritized groups when compared with non-Hispanic white populations [17, 18]. Importantly we demonstrate that the CCI specifically may be a useful measure to capture self-perceived cognitive decline in health disparity older adult groups.

The present study has several notable strengths, including being the first to comprehensively explore various CCI versions cross-sectionally and longitudinally in a demographically and clinically diverse nondemented older adult population. Some study limitations, as well as future research opportunities, warrant mention. Our overall sample size was somewhat small (N = 322) with relatively few years of annual follow up data (median = 4 years). For this reason, we were not able to examine incident dementia as an outcome. Additionally, attrition bias, due to sample size reductions at annual follow-up (N = 234, 245, 213, 167, 109 at waves 0–5), may influence our longitudinal results. Future research in larger samples with longer follow up periods should explore the ability of the CCI to predict dementia. Unfortunately, informant report CCI was not included in the EAS protocol, so we were not able to assess informant reported CCI as a predictor of decline or change in CCI report over time. Future research should explore the relative contribution of self and informant reported CCI to predict risk. Unfortunately, we had a small number of Hispanic (n = 40) and a very small number of Asian (n = 3) participants, which may limit the generalizability of our findings in those diverse groups. To avoid diagnostic circularity with MCI, the number of neuropsychological outcomes we were able to explore were limited. Although beyond the scope of the current work, future research should explore the CCI at an item-by-item level as well as explore intra-individual variability in CCI responding longitudinally, to identify specific patterns of responding on the CCI that may be optimal for risk detection.

In conclusion, the CCI self-report instrument is useful for predicting future progression to cognitive impairment. CCI scores vary by baseline predementia diagnostic status, with SCD participants showing highest levels of concerns, MCI and MCR showing similar levels, and healthy older adults showing the least. The CCI total scores are associated with depressive symptoms, low levels of education, ethnic/racial minority status, and performance on objective neuropsychological testing. Non-Hispanic Black older adults show higher levels of cognitive concerns than non-Hispanic white older adults, and these concerns are associated with risk for decline regardless of ethnic/racial group. In sum, self-perception of cognitive change, as captured on the CCI, is a useful marker of dementia risk in demographically and clinically diverse nondemented populations. All self-report CCI versions (CCI-12, 20, and 40) were associated with decline. The CCI-40, although double in length, may provide marginal benefit for prediction of outcomes in clinical and research settings with diverse at-risk populations.

AUTHOR CONTRIBUTIONS

Caroline O. Nester (Conceptualization; Formal analysis; Funding acquisition; Writing – original draft); Qi Gao (Formal analysis; Writing – review & editing); Mindy J. Katz (Funding acquisition; Project administration; Resources; Writing – review & editing); Jacqueline A. Mogle (Supervision; Writing – review & editing); Cuiling Wang (Formal analysis; Writing – review & editing); Carol A. Derby (Supervision; Writing – review & editing); Richard B. Lipton (Conceptualization; Funding acquisition; Project administration; Supervision; Writing – review & editing); Andrew J. Saykin (Conceptualization; Supervision; Writing – review & editing); Laura A. Rabin (Conceptualization; Supervision; Writing – review & editing).

Footnotes

ACKNOWLEDGMENTS

The authors have no acknowledgements to report.

FUNDING

This work was supported by the National Institutes of Health, National Institute of Neurological Disorders and Stroke [F31NS127520 to CN]; National Institute on Aging [3P01AG003949 to RL]; and The CUNY Graduate Center Dissertation Fellowship to CN. AS receives support from multiple NIH grants (P30 AG010133, P30 AG072976, R01 AG019771, R01 AG057739, U19 AG024904, R01 LM013463, R01 AG068193, T32 AG071444, U01 AG068057, U01 AG072177, and U19 AG074879). Support was also provided by the Leonard and Sylvia Marx Foundation, the Czap Foundation, and the Hollander Family Foundation.

CONFLICT OF INTEREST

AS acknowledges support from Avid Radiopharmaceuticals, a subsidiary of Eli Lilly (in kind contribution of PET tracer precursor); Bayer Oncology (Scientific Advisory Board); Eisai (Scientific Advisory Board); Siemens Medical Solutions USA, Inc. (Dementia Advisory Board); NIH NHLBI (MESA Observational Study Monitoring Board); Springer-Nature Publishing (Editorial Office Support as Editor-in-Chief, Brain Imaging and Behavior).

The other authors have no conflict of interest to report.

DATA AVAILABILITY

The data supporting the findings of this study are available on request from the corresponding author.

The supplementary material is available in the electronic version of this article: .

References

Jessen

, Amariglio

, Buckley

, van der Flier

, Han

, Molinuevo

, Rabin

, Rentz

, Rodriguez-Gomez

, Saykin

(2020) The characterisation of subjective cognitive decline. Lancet Neurol 19, 271–278.

Jessen

, Amariglio

, Van Boxtel

, Breteler

, Ceccaldi

, Chételat

, Dubois

, Dufouil

, Ellis

, Van Der Flier

(2014) A conceptual framework for research on subjective cognitive decline in preclinical Alzheimer’s disease. Alzheimers Dement 10, 844–852.

Mitchell

, Beaumont

, Ferguson

, Yadegarfar

, Stubbs

(2014) Risk of dementia and mild cognitive impairment in older people with subjective memory complaints: Meta-analysis. Acta Psychiatr Scand 130, 439–451.

Petersen

, Caracciolo

, Brayne

, Gauthier

, Jelic

, Fratiglioni

(2014) Mild cognitive impairment: A concept in evolution. J Intern Med 275, 214–228.

American Psychiatric Association (2013) Diagnostic and statistical manual of mental disorders: DSM-5, American Psychiatric Association, Washington, DC.

Verghese

, Ayers

, Barzilai

, Bennett

, Buchman

, Holtzer

, Katz

, Lipton

, Wang

(2014) Motoric cognitive risk syndrome: Multicenter incidence study. Neurology 83, 2278–2284.

Ficker

, Lysack

, Hanna

, Lichtenberg

(2014) Perceived cognitive impairment among African American elders: Health and functional impairments in daily life. Aging Ment Health 18, 471–480.

Flatt

, Johnson

, Karpiak

, Seidel

, Larson

, Brennan-Ing

(2018) Correlates of subjective cognitive decline in lesbian, gay, bisexual, and transgender older adults. J Alzheimers Dis 64, 91–102.

Yue

, Hu

, Zhang

, Wen

, Wu

, Li

, Sun

, Li

, Wang

, Li

(2021) Prediction of 7-year’s conversion from subjective cognitive decline to mild cognitive impairment. Hum Brain Mapp 42, 192–203.

10.

Jenkins

, Tree

, Thornton

, Tales

(2019) Subjective cognitive impairment in 55-65-year-old adults is associated with negative affective symptoms, neuroticism, and poor quality of life. J Alzheimers Dis 67, 1367–1378.

11.

Peterson

, Carvajal

, McGuire

, Fain

, Bell

(2019) State inequality, socioeconomic position and subjective cognitive decline in the United States. SSM Popul Health 7, 100357.

12.

Rabin

, Smart

, Amariglio

(2017) Subjective cognitive decline in preclinical Alzheimer’s disease. Annu Rev Clin Psychol 13, 369–396.

13.

Nester

, Ayers

, Rabin

, Verghese

(2020) Non-memory subjective cognitive concerns predict incident motoric cognitive risk syndrome. Eur J Neurol 27, 1146–1154.

14.

Edmonds

, Delano-Wood

, Galasko

, Salmon

, Bondi

(2014) Subjective cognitive complaints contribute to misdiagnosis of mild cognitive impairment. J Int Neuropsychol Soc 20, 836–847.

15.

Bondi

, Edmonds

, Jak

, Clark

, Delano-Wood

, McDonald

, Nation

, Libon

, Au

, Galasko

(2014) Neuropsychological criteria for mild cognitive impairment improves diagnostic precision, biomarker associations, and progression rates. J Alzheimers Dis 42, 275–289.

16.

Brewster

, Barnes

, Haan

, Johnson

, Manly

, Nápoles

, Whitmer

, Carvajal-Carmona

, Early

, Farias

(2019) Progress and future challenges in aging and diversity research in the United States. Alzheimers Dement 15, 995–1003.

17.

Chapman

, Rentería

, Dworkin

, Garriga

, Barker

, Avila-Rieger

, Gonzalez

, Joyce

, Vonk

, Soto

(2023) Association of subjective cognitive decline with progression to dementia in a cognitively unimpaired multiracial community sample. , . Neurology 100, e1020–e1027.

18.

Arvanitakis

, Leurgans

, Fleischman

, Schneider

, Rajan

, Pruzin

, Shah

, Evans

, Barnes

, Bennett

(2018) Memory complaints, dementia, and neuropathology in older blacks and whites. Ann Neurol 83, 718–729.

19.

Rodríguez

, Ayers

, Weiss

, Verghese

(2021) Cross-cultural comparisons of subjective cognitive complaints in a diverse primary care population. J Alzheimers Dis 81, 545–555.

20.

Ferraro

, Sauerteig-Rolston

, Barnes

, Friedman

, Sands

, Thomas

(2023) Subjective memory decline predicts incident cognitive impairment among White—but not Black or Hispanic—older adults. Gerontologist 63, 690–699.

21.

Sims

, Whitfield

, Ayotte

, Gamaldo

, Edwards

, Allaire

(2011) Subjective memory in older African Americans. Exp Aging Res 37, 220–240.

22.

Jackson

, Rentz

, Aghjayan

, Buckley

, Meneide

, Sperling

, Amariglio

(2017) Subjective cognitive concerns are associated with objective memory performance in Caucasian but not African-American persons. Age Ageing 46, 988–993.

23.

Hughes

, Winkey

, Sachs

, Fitzpatrick

, Craft

, Baker

, Heckbert

, Jacobs

Jr , Rapp

, Hayden

(2020) Subjective cognitive impairment differs by racial/ethnic groups and sociodemographic factors: The Multi-Ethnic Study of Atherosclerosis and Alzheimer’s Disease (MESA-MIND) featured research and focused topic sessions: Examining subjective cognitive decline in diverse, ethnoracial populations. Alzheimers Dement 16, e040429.

24.

Saykin

, Wishart

, Rabin

, Santulli

, Flashman

, West

, McHugh

, Mamourian

(2006) Older adults with cognitive complaints show brain atrophy similar to that of amnestic MCI. Neurology 67, 834–842.

25.

Rattanabannakit

, Risacher

, Gao

, Lane

, Brown

, McDonald

, Unverzagt

, Apostolova

, Saykin

, Farlow

(2016) The cognitive change index as a measure of self and informant perception of cognitive decline: Relation to neuropsychological tests. J Alzheimers Dis 51, 1145–1155.

26.

Katz

, Qin

, Wang

, Rabin

(2023) Subjective cognitive concerns differ by race/ethnicity among einstein aging study (EAS) participants: Implications for assessment and interpretation of self-perceived cognitive functioning. Alzheimers Dement 19, e065943.

27.

Nester

CO MJ

, Katz

, Saykin

, Lipton

, Rabin

(2022) The Cognitive Change Index Supplement (CCI-40): Broadening the assessment of subjective cognitive concerns in demographically diverse community dwelling older adults. In International Neuropsychological Society Annual Meeting, Virtual.

28.

Risacher

, Kim

, Nho

, Foroud

, Shen

, Petersen

, Jack

Jr , Beckett

, Aisen

, Koeppe

(2015) APOE effect on Alzheimer’s disease biomarkers in older adults with significant memory concern. Alzheimers Dement 11, 1417–1429.

29.

Hayden

, Leng

, Rapp

, Shumaker

, Espeland

, Weitlauf

, Wallace

(2019) P1-289: Subjective cognitive impairment and anosognosia in the Women’s Health Initiative Memory Study. Alzheimers Dement 15, P353–P354.

30.

Slot

, Verfaillie

, Overbeek

, Timmers

, Wesselman

, Teunissen

, Dols

, Bouwman

, Prins

, Barkhof

(2018) Subjective Cognitive Impairment Cohort (SCIENCe): Study design and first results. Alzheimers Res Ther 10, 76.

31.

Risacher

, West

, McDonald

, Tallman

, Glazier

, Gao

, Brown

, Apostolova

, Brosch

, Farlow

(2017) [IC-P-214]: Optimizing cognitive change index cutoffs based on cognitive decline and biomarker positivity in cognitively normal older adults. Alzheimers Dement 13, P154.

32.

Risacher

, WuDunn

, Tallman

, West

, Gao

, Farlow

, Brosch

, Apostolova

, Saykin

(2020) Visual contrast sensitivity is associated with the presence of cerebral amyloid and tau deposition. , fcaa. Brain Commun 2, 019.

33.

Risacher

, West

, McDonald

, Tallman

, Gandhi

, Glazier

, Gao

, Brown

, O’Neill

, Crosbie

(2016) P4-166: Association of the cognitive change index with hippocampal subfield volumes. Alzheimers Dement 12, P1080.

34.

Morrison

, Dadar

, Shafiee

, Villeneuve

, Louis Collins

; for Alzheimer’s Disease Neuroimaging Initiative (2022) Regional brain atrophy and cognitive decline depend on definition of subjective cognitive decline. Neuroimage Clin 33, 102923.

35.

, Song

, Liang

, Xing

, Qin

, Wu

(2021) Decreased resting-state functional complexity in elderly with subjective cognitive decline. Entropy (Basel) 23, 1591.

36.

Peng

, Feng

, Ma

, Xu

, Gao

(2022) Rich-club organization disturbances of the individual morphological network in subjective cognitive decline. Front Aging Neurosci 14, 834145.

37.

Contreras

, Goñi

, Risacher

, Amico

, Yoder

, Dzemidzic

, West

, McDonald

, Farlow

, Sporns

(2017) Cognitive complaints in older adults at risk for Alzheimer’s disease are associated with altered resting-state networks. Alzheimers Dement (Amst) 6, 40–49.

38.

Katz

, Lipton

, Hall

, Zimmerman

, Sanders

, Verghese

, Dickson

, Derby

(2012) Age-specific and sex-specific prevalence and incidence of mild cognitive impairment, dementia, and Alzheimer dementia in blacks and whites: A report from the Einstein Aging Study. Alzheimer Dis Assoc Disord 26, 335–343.

39.

Katz

, Wang

, Nester

, Derby

, Zimmerman

, Lipton

, Sliwinski

, Rabin

(2021) T-MoCA: A valid phone screen for cognitive impairment in diverse community samples. Alzheimers Dement (Amst) 13, e12144.

40.

American Psychiatric Association (1994) Diagnostic and statistical manual of mental disorders: DSM-IV, American psychiatric Association, Washington, DC.

41.

Lawton

, Brody

(1969) Assessment of older people: Self-maintaining and instrumental activities of daily living. Gerontologist 9, 179–186.

42.

Molinuevo

, Rabin

, Amariglio

, Buckley

, Dubois

, Ellis

, Ewers

, Hampel

, Klöppel

, Rami

(2017) Implementation of subjective cognitive decline criteria in research studies. Alzheimers Dement 13, 296–311.

43.

Jessen

, Wiese

, Bachmann

, Eifflaender-Gorfer

, Haller

, Kölsch

, Luck

, Mösch

, van den Bussche

, Wagner

(2010) Prediction of dementia by subjective memory impairment: Effects of severity and temporal association with cognitive impairment. Arch Gen Psychiatry 67, 414–422.

44.

Geerlings

, Jonker

, Bouter

, Adèr

, Schmand

(1999) Association between memory complaints and incident Alzheimer’s disease in elderly people with normal baseline cognition. Am J Psychiatry 156, 531–537.

45.

Morris

(1997) Clinical dementia rating: A reliable and valid diagnostic and staging measure for dementia of the Alzheimer type. Int Psychogeriatr 9, 173–176.

46.

Weintraub

, Besser

, Dodge

, Teylan

, Ferris

, Goldstein

, Giordani

, Kramer

, Loewenstein

, Marson

, Mungas

, Salmon

, Welsh-Bohmer

, Zhou

, Shirk

, Atri

, Kukull

, Phelps

, Morris

(2018) Version 3 of the Alzheimer Disease Centers’ neuropsychological test battery in the Uniform Data Set (UDS). Alzheimer Dis Assoc Disord 32, 10–17.

47.

Wang

, Katz

, Chang

, Qin

, Lipton

, Zwerling

, Sliwinski

, Derby

, Rabin

(2021) UDSNB 3.0 neuropsychological test norms in older adults from a diverse community: Results from the Einstein Aging Study (EAS). J Alzheimers Dis 83, 1665–1678.

48.

Nasreddine

, Phillips

, Bédirian

, Charbonneau

, Whitehead

, Collin

, Cummings

, Chertkow

(2005) The Montreal Cognitive Assessment, MoCA: A brief screening tool for mild cognitive impairment. J Am Geriatr Soc 53, 695–699.

49.

Craft

, Newcomer

, Kanne

, Dagogo-Jack

, Cryer

, Sheline

, Luby

, Dagogo-Jack

, Alderson

(1996) Memory improvement following induced hyperinsulinemia in Alzheimer’s disease. Neurobiol Aging 17, 123–130.

50.

Yesavage

, Brink

, Rose

, Lum

, Huang

, Adey

, Leirer

(1983) Development and validation of a geriatric depression screening scale: A preliminary report. J Psychiatr Res 17, 37–49.

51.

Burmester

, Leathem

, Merrick

(2016) Subjective cognitive complaints and objective cognitive function in aging: A systematic review and meta-analysis of recent cross-sectional findings. Neuropsychol Rev 26, 376–393.

52.

Edmonds

, Weigand

, Thomas

, Eppig

, Delano-Wood

, Galasko

, Salmon

, Bondi

(2018) Increasing inaccuracy of self-reported subjective cognitive complaints over 24 months in empirically derived subtypes of mild cognitive impairment. J Int Neuropsychol Soc 24, 842–853.

53.

Liew

(2019) Depression, subjective cognitive decline, and the risk of neurocognitive disorders. Alzheimers Res Ther 11, 70.

54.

Dafsari

, Jessen

(2020) Depression—an underrecognized target for prevention of dementia in Alzheimer’s disease. Transl Psychiatry 10, 160.

55.

Galeone

, Pappalardo

, Chieffi

, Iavarone

, Carlomagno

(2011) Anosognosia for memory deficit in amnestic mild cognitive impairment and Alzheimer’s disease. Int J Geriatr Psychiatry 26, 695–701.

56.

Lehrner

, Kogler

, Lamm

, Moser

, Klug

, Pusswald

, Dal-Bianco

, Pirker

, Auff

(2015) Awareness of memory deficits in subjective cognitive decline, mild cognitive impairment, Alzheimer’s disease and Parkinson’s disease. Int Psychogeriatr 27, 357–366.

57.

Wooten

, McGuire

, Olivari

, Jackson

, Croft

(2023) Racial and ethnic differences in subjective cognitive decline—United States, 2015–2020. MMWR Moeb Mortal Wkly Rep 72, 249–255.

58.

Gupta

(2021) Racial and ethnic disparities in subjective cognitive decline: A closer look, United States, 2015–2018. BMC Public Health 21, 1173.

59.

Barnes

(2022) Alzheimer disease in African American individuals: Increased incidence or not enough data? . Nat Rev Neurol 18, 56–62.

60.

Hooten

, Pacheco

, Smith

, Evans

(2022) The accelerated aging phenotype: The role of race and social determinants of health on aging. Ageing Res Rev 73, 101536.

61.

Simons

, Ong

, Beach

SRH

, Lei

, Philibert

, Mielke

(2023) Direct and indirect effects of socioeconomic status and discrimination on subjective cognitive decline: A longitudinal Study of African American women. J Gerontol B Psychol Sci Soc Sci 78, 799–808.