Memory Complaints and Impairment Among Middle-Aged and Older Brazilians: ELSI-Brazil Study

Abstract

Objective

To measure awareness of memory impairment (MI) among individuals with MI, the prevalence of MI among those with memory complaints (MC), and to identify associated factors to both in a nationally representative sample of Brazilians aged ≥50.

Methods

MI was defined as a z-score ≤ −1.5 derived from immediate and delayed recall tests, adjusted for demographics. MC were identified by self-report. Logistic regression models examined sociodemographic and health-related factors associated with awareness of MI and with MI among individuals reporting MC.

Results

Of 7831 participants, 739 (9.3%) had MI, of whom 52% were aware of their impairment. Greater awareness was associated with a higher number of chronic illnesses, disability in instrumental activities of daily living (i-ADL), and more depressive symptoms. Among 3402 (41.7%) participants with MC, 11.5% had objective MI. In this group, higher education, lower income, fewer chronic illnesses, disability in i-ADL and advanced ADL (a-ADL), and more depressive symptoms were associated with MI.

Conclusion

Half of individuals with objective MI demonstrated awareness of their impairment, and one in ten individuals with MC showed objective cognitive decline. These findings emphasize the need for cognitive screening in middle aged and older adults and for clinical evaluation of other causes of memory complaints, particularly mood and functional factors.

Keywords

cognitive decline cognitive impairment memory epidemiology elderly

Introduction

Awareness of memory impairment plays an intricate role in cognitive decline. While memory complaints (MC) might be considered one of the earliest symptoms of Alzheimer’s disease,^1,2 unawareness of memory impairment is frequently observed in individuals with dementia. It can occur at the mild cognitive impairment (MCI) stage,³ with a trend to become more frequent as the cognitive impairment progressively increases.⁴ The prevalence of unawareness of memory impairment has been estimated to be 60-80% in people with dementia, and 15-30% in people with MCI.³ It has been associated with sociodemographic characteristics³ such as sex, education and age, cognitive abilities,⁵ having fewer depressive and anxiety symptoms, more apathy,¹ and higher caregiver burden,⁶ reinforcing the importance of studying this phenomenon.

Memory complaints (MC) can occur with or without memory impairment (MI). When MC are present in individuals without objective MI, the condition is referred to as subjective cognitive decline (SCD). SCD is considered part of the continuum of cognitive decline⁷ and may represent a pre-clinical stage of dementia.⁷ Individuals with SCD often notice subtle early cognitive changes that are not yet evident in daily activities or observable by others, even in standard cognitive tests. However, not all MC without MI indicate early neurodegeneration and the presence of MC has been associated with other factors such as depression,⁸ self-rated health, neuroticism, and psychosocial factors,⁹ independently of cognitive functioning. Assessing whether individuals with MC show objective cognitive decline is crucial for early diagnosis, prevention and treatment of dementia.

The global burden of dementia is projected to rise sharply, particularly in low- and middle-income countries (LMICs).¹⁰ However, most studies, including those on SCD, have taken place in high-income countries (HICs), often with less diverse populations, thus limiting their generalizability to the countries where they might be most needed.¹¹ One such country is Brazil, an LMIC with significant heterogeneity that is undergoing rapid population aging, with dementia prevalence estimated to be 5.8% in individuals aged 60 and older.^10,12 The few studies that have been conducted on MC, awareness of MI, and cognitive decline in Brazil have largely relied on small, non-population-based samples,^13-15 Thus, in this study we aimed to measure these factors in a large representative sample of Brazilians aged 50 or older.

Methods

Sample

We used the baseline data from the Brazilian Longitudinal Study of Aging (ELSI-Brazil), a nationally representative sample of noninstitutionalized Brazilians aged 50 years or older. Baseline data was collected between 2015-2016 and included 9412 individuals from 70 municipalities across the five Brazilian geopolitical regions from both urban and rural settings. To characterize the sociodemographic variety of middle aged and older adults in Brazil, participants were selected through a random sampling process using municipalities, census tract, and households. An inverse sampling design was used to compensate for non-responders(More details of the ELSI study can be found in a previous publication¹⁶). ELSI-Brazil follows a conceptual framework similar to the Health and Retirement Study (HRS) partner studies.¹⁷

Ethics

The ELSI-Brazil study was approved by the Ethics Committee of the Oswaldo Cruz Foundation, Minas Gerais (CAAE 34649814.3.0000.5091).

Measurements

Objective memory impairment (MI): The ELSI-Brazil cognitive function module was designed to be like other HRS sister studies. MI was defined through a z-score calculated using a regression-based-norms approach (based on age, education, and gender/sex) in a healthy subset of individuals from the ELSI-Brazil study, namely: (a) visual and hearing deficits that could affect test performance; (b) self-report of a previous medical diagnosis of depression or clinical depressive symptoms according to the Center for Epidemiological Scale—Depression (cutoff of 4 points; CESD-8), (c) self-report or informant-based diagnosis history of Alzheimer’s disease (AD), Parkinson disease, or stroke; (d) heavy drinking based on the National Institute of Alcohol Abuse and Alcoholism criteria (weekly use of 14 doses or daily use of 4 doses for men, and weekly use of 7 doses or daily use of 3 doses for women); (e) self-report or informant-based memory complaints; (f) self-report of impairment in 4 gender-independent instrumental activities of daily living (IADLs; money managing, using transportation, using the telephone, and taking medications); and (g) missing cognitive data).¹² The memory score is based on a combination of immediate and delayed responses from the episodic memory test (10-word list), with individuals with scores equal or under −1.5 classified as having memory impairment.

Memory Complaints (MC): We assessed MC based on responses to the question: “Currently, how would you rate your memory?”, with answers: excellent, very good, good, fair, and poor. Those participants who perceived their memory as fair or poor were identified as having MC. Individuals were considered aware of MI if they presented both MI and MC, and unaware if they presented MI but did not present MC.

Sociodemographic variables: We gathered data on age, sex, education (years), and monthly household income per capita.

Health-Related Variables

Current depressive symptoms were measured using the total score of the 8-item version of the Centre for Epidemiologic Studies Depression Scale (CES-D-8), based on dichotomous answers (yes/no) for how the participant felt in the previous week, with higher scores indicating a greater number of symptoms.¹⁸

Self-rated health was based on the participants self-reported evaluation of their health. We divided the participants into those who reported having regular, good, and very good health and those who reported having poor or very poor health.

The number of chronic illnesses was computed as the total of previous self-reported medical diagnoses of the following diseases: hypertension, diabetes mellitus, hypercholesterolemia, cardiovascular illness (stroke, heart attack, cardiac insufficiency, and angina), asthma, chronic obstructive pulmonary disease (COPD), arthritis, osteoporosis, back pain, depression, cancer, and renal insufficiency.

The Activities of Daily Living (ADL) total score was calculated based on the number of reported difficulties, with each positive response indicating a disability in one specific item. B-ADL and i-ADL were collected using a Likert scale and only those who reported great difficulty (does the activity alone but only with great effort) and those who could not do it (can do the activity only with help) were included as a disability, while a-ADL data was collected using dichotomic answers (yes/no). All ADLs were harmonized so that higher scores indicate greater functional disability in a specific domain. The questionnaires used allowed for direct comparisons with other HRS sister studies.

• Basic activities of daily living (b-ADL) included walking, transferring, dressing, bathing, eating, toilet hygiene, and fecal and urinary incontinence.

• Instrumental activities of daily living (i-ADL) included personal care, cooking, finance management, use of transportation, shopping, using a telephone/cellphone, medication management, and doing light and heavy house cleaning.

• Advanced activities of daily living (a-ADL) included contacting others (through letters, emails, or telephone), visiting friends or family at their houses, receiving friends or family at their own house, going with others to public places, participating in organized social activities, participating in public associations, using computers, driving, playing games with friends, doing manual activities or hobbies, making short trips, making long trips, participating in volunteer work and caring for others.

Statistical Analysis

Statistical analyses were conducted using the Stata/SE16 software (StataCorp. 2021, College Station, TX, USA) and were all undertaken using weighting factors for each stratum to minimize differential probability of selection and non-response. Images were created using RStudio version ‘2024.4.2.764’. Group comparison was performed using the Chi-square test for categorical variables and the t-test for continuous variables, with a P < .05 being considered statistically significant. We estimated the prevalence of MI awareness in individuals with MI and the prevalence of MI in individuals with MC for all respondents. For each of these prevalences, two logistic regression models with robust variance were performed. Model 1 was adjusted for age, sex, education, and household income per capita. Model 2 was adjusted for all variables. Collinearity diagnostics were evaluated using the Variance Inflation Factor derived from an equivalent linear regression model, showing no meaningful collinearity (for the logistic regression of MC in individuals with MI varied from 1.09 to 1.89, and for the logistic regression of MI in individuals with MC from 1.12 to 1.81).

Additionally, we conducted a sensitivity analysis using imputation for missing data. We used the logistic regression imputation method to impute missing binary variable values (memory complaints and self-rated health). Missing values of ordinal variables (depressive symptoms, number of chronic illnesses, education, b-ADL, i-ADL, and a-ADL) were imputed using an ordered logistic regression imputation method and missing values of continuous variables (objective memory performance) using a linear regression imputation method. We repeated the previous analysis after multiple imputation, using age, sex, and household income per capita as predictors of the variables with missing data.

Results

From the 9412 participants of the original sample, 7831 had information for all variables, and from this group, 739 (9.3%) had MI and 3402 (41.9%) had MC. Fifty-two per cent of those with MI had MC and 11.5% of individuals with MC had MI (n = 371) (Figure 1).

Among the subsample of individuals with MI, the majority were women (57.7%), with a mean age of 62.1 years (SD = 9.6) and an average of 5.7 years of education (SD = 4.6). Individuals aware of MI had lower education, lower household income per capita, a higher number of chronic illnesses, higher disability in b-ADL, i-ADL, and a-ADL, lower self-rated health and a higher number of depressive symptoms (Table 1).

Table 1.

Characteristics of Individuals With Memory Impairment (MI) in Relation to Awareness of Memory Complaints (MC) (n = 739)

	Total n = 739	Without MC (“unaware”) n = 368 (48.0%)	With MC (“aware”) n = 371 (52.0%)	P Value chi²/t test
Sex (female). n(wgt%)	449 (57.7)	218 (54.9)	231 (57.7)	Chi² = 0.71 P = 0.400<
Age (years). Mean (SD)	62.1 (9.6)	62.7 (10.1)	61.6 (9.1)	t-test = 1.55 P = 0.121
Education (years). Mean (SD)	5.7 (4.6)	6.1 (4.8)	5.2 (4.3)	t-test = 2.64 P = 0.008
Monthly household income per capita (Brazilian reais). Mean (SD)	935.8 (1108.5)	1041.3 (1361.3)	831.3 (768.8)	t-test = 2.58 P = 0.009
N. of chronic illnesses. Mean (SD)	2.3 (1.6)	1.9 (1.4)	2.6 (1.8)	t-test = −6.34 <0.001
B-ADL. Mean (SD)	0.3 (0.9)	0.2 (0.7)	0.4 (1.1)	t-test = −3.38 <0.001
I-ADL. Mean (SD)	1.3 (1.9)	0.9 (1.6)	1.6 (2.2)	t-test = −2.33 <0.001
A-ADL. Mean (SD)	10.0 (2.3)	9.8 (2.5)	10.2 (2.1)	t-test = −2.82 P = 0.019
Self-rated health (bad/very bad). n (wgt %)	112 (13.1)	28 (6.9)	84 (18.8)	Chi² = 32.47 <0.001
Depressive symptoms (CESD-8 total score). Mean (SD)	3.2 (2.4)	2.4 (2.2)	4.1 (2.4)	t-test = −10.32 <0.001

SD = standard deviation; wgt = weight; MI = memory impairment; SD = standard deviation; B-ADL = basic activities of daily living; I-ADL = instrumental activities of daily living; A-ADL = advanced activities of daily living; CESD = centre for epidemiologic studies depression scale.

The fully adjusted logistic regression models revealed that in individuals with MI, a higher number of chronic illnesses (OR = 1.16; 95% CI 1.03 to 1.32), higher disability in i-ADL (OR = 1.17; 95% CI 1.02 to 1.33) and a higher number of depressive symptoms (OR = 1.27; 95% CI 1.15 to 1.41) were associated with awareness of memory impairment (Table 2).

Table 2.

Association of Sociodemographic and Health Variables With Awareness of MI (n = 739)

	Adjusted OR^a (95% CI)P-value	Adjusted OR^b (95% CI)P-value
Sex (female)	1.25 (0.84-1.87) P = 0.270	0.93 (0.59-1.45) P = 0.750
Age (years)	0.99 (0.97-1.00) P = 0.164	0.98 (0.96-1.00) P = 0.060
Education (years)	0.96 (0.91-1.01) P = 0.131	0.98 (0.93-1.03) P = 0.446
Monthly household income per capita (Brazilian reais)	0.99 (0.99-1.00) P = 0.075	0.99 (0.99-1.00) P = 0.260
N. of chronic illnesses	1.33 (1.19-1.49) <0.001	1.16 (1.03-1.32) P = 0.019
B-ADL	1.24 (1.00 -1.52) P = 0.045	0.87 (0.69-1.08) P = 0.206
I-ADL	1.26 (1.13-1.42) <0.001	1.17 (1.02-1.33) P = 0.024
A-ADL	1.08 (1.01-1.18) P = 0.031	1.05 (0.96-1.15) P = 0.282
Self-rated health (bad/very bad)	2.88 (1.74 - 4.78) P < 0.001	1.37 (0.73-2.56) P = 0.328
Depressive symptoms (CESD-8 total score)	1.34 (1.22-1.48) <0.001	1.27 (1.15-1.41) <0.001

^aAdjusted for sex/gender, age, education, and household income.

^bMutually adjusted.

MI = Memory impairment; SD = standard deviation; B-ADL = basic activities of daily living; I-ADL = instrumental activities of daily living; A-ADL = advanced activities of daily living; CESD = centre for epidemiologic studies depression scale.

Among the subsample of individuals with MC, the majority were women (60.2%), with a mean age of 62.6 years (SD = 9.3) and an average of 4.7 years of education (SD = 3.9). Within this subsample, individuals with MI had higher education, lower household per capita income, greater disability in Basic Activities of Daily Living (b-ADL), Instrumental Activities of Daily Living (i-ADL), and Advanced Activities of Daily Living (a-ADL), worse self-perceived health, and a higher number of depressive symptoms (Table 3).

Table 3.

Characteristics of Individuals With Memory Complaints (MC) in Relation to Memory Impairment (MI) (n = 3402)

	Total n = 3402	Without MI n = 3031(88.5%)	With MI n = 371(11.5%)	P value chi²/t test
Total	3402	3031 (88.5)	371 (11.5)
Sex (female), n(wgt%)	2085 (60.2)	1854 (60.1)	231 (60.3)	Chi² = 0.17P = 0.682
Age (years), mean (SD)	62.6 (9.3)	62.7 (9.3)	61.6 (9.1)	t-test = 2.22P = 0.026
Education (years), mean (SD)	4.7 (3.9)	4.6 (3.8)	5.2 (4.3)	t-test = −3.10P = 0.002
Monthly household income per capita (Brazilian reais), mean (SD)	939.4 (1106.4)	952.7 (1140.3)	831.3 (768.8)	t-test = −3.10P = 0.046
N. of chronic illnesses, mean (SD)	2.7 (1.8)	2.7 (1.8)	2.6 (1.8)	t-test = 0.13P = 0.891
b-ADL, mean (SD)	0.2 (0.9)	0.2 (0.8)	0.4 (1.1)	t-test = −3.77<0.001
i-ADL, mean (SD)	1.1 (1.8)	1.1 (1.7)	1.6 (2.2)	t-test = −5.43<0.001
a-ADL, mean (SD)	9.6 (2.3)	9.5 (2.4)	10.2 (2.1)	t-test = −5.07<0.001
Self-rated health (bad/very bad), n(wgt %)	622 (17.8)	538 (17.6)	84 (18.8)	Chi² = 5.29P = 0.021
Depressive symptoms (CESD-8 total score), mean (SD)	3.6 (2.6)	3.5 (2.6)	4.1 (2.4)	t-test = −4.16<0.001

SD = standard deviation; wgt = weight; MI = memory impairment; SD = standard deviation; b-ADL = basic activities of daily living; i-ADL = instrumental activities of daily living; a-ADL = advanced activities of daily living; CESD = centre for epidemiologic studies depression scale.

The fully adjusted logistic regression model revealed that among individuals with MC, higher education (OR = 1.11; 95% CI 1.07 to 1.16), greater disability in i-ADL (OR = 1.13; 95% CI 1.03 to 1.24) and a-ADL (OR = 1.22; 95% CI 1.14 to 1.29), and a higher number of depressive symptoms (OR = 1.07; 95% CI 1.01 to 1.13) were associated with greater MI. In contrast, having a higher per capita household income (OR = 0.99; 95% CI 0.99 to 0.99) and a higher number of chronic health illnesses (OR = 0.91; 95% CI 0.86 to 0.97) were associated with less MI (Table 4).

Table 4.

Association of Sociodemographic and Health Variables With MI in Individuals With MC (n = 3402)

	Adjusted OR^a (95% CI)P-value	Adjusted OR^b (95% CI)P-value
Sex (female)	1.08 (0.84-1.39)P = 0.543	0.92 (0.70-1.21)P = 0.581
Age (years)	0.99 (0.97-1.00)P = 0.103	0.98 (0.97-1.00)P = 0.063
Education (years)	1.02 (0.99-1.06)P = 0.206	1.11 (1.07-1.16)<0.001
Monthly household income per capita (Brazilian reais)	1.00 (0.99-1.00)P = 0.503	0.99 (0.99-0.99)P = 0.044
N. of chronic illnesses	0.96 (0.90-1.01)P = 0.140	0.91 (0.86-0.97)P = 0.003
b-ADL	1.19 (1.08-1.32)<0.001	0.99 (0.86-1.16)P = 0.970
i-ADL	1.17 (1.11-1.24)<0.001	1.13 (1.03-1.24)P = 0.009
a-ADL	1.24 (1.18-1.31)<0.001	1.22 (1.14-1.29)<0.001
Self-rated health (bad/very bad)	1.05 (0.81-1.36)P = 0.723	0.81 (0.57-1.15)P = 0.242
Depressive symptoms (CESD-8 total score)	1.09 (1.03-1.14)<0.001	1.07 (1.01-1.13)P = 0.025

^aAdjusted for sex/gender, age, education, and household income.

^bMutually adjusted.

MI = memory impairment; SD = standard deviation; b-ADL = basic activities of daily living; i-ADL = instrumental activities of daily living; a-ADL = advanced activities of daily living; CESD = centre for epidemiologic studies depression scale.

After multiple imputation, 986 individuals from the full sample were identified as having MI, and 4160 individuals as having MC and were included in the sensitivity analysis. 492 individuals had MC and MI and were considered aware, representing 49.9% of individuals with MI and 11.8% of the individuals with MC. Group comparison found similar values except that household income and disability in b-ADL were not associated with awareness of MI (Supplemental Table 1). Similarly, logistic regression found similar associations to the model without the multiple imputation, except that lower education was associated with awareness in both models, while disability in b-ADL and a-ADL were not associated with awareness in the sociodemographic adjusted model, and disability in i-ADL was not associated with awareness in the mutually adjusted model while worst self-rated health was associated (Supplemental Table 2).

Group comparisons found that differences in age, education, household income and self-rated health were not associated with MI (Supplemental Table 3). The results of the logistic regression models were in the same direction as those without the multiple imputation, with the only difference being that mean household income per capita was not associated with MI in the fully adjusted model (Supplemental Table 4).

Discussion

This study is the first to use a nationally representative sample of Brazilians aged 50 years or older to assess objective memory impairment and its associations with MI awareness and memory complaints. We found an overall prevalence of MI of 9.3%, and in this group 52% were aware of their cognitive status. A higher number of chronic illnesses, higher disability in i-ADL, and a higher number of depressive symptoms were associated with a higher likelihood of awareness of MI. The overall prevalence of MC was 41.9%, and in this group 11.5% of individuals had MI. Higher education, higher disability in i-ADL and a-ADL, and a higher number of depressive symptoms were associated with the prevalence of MI in individuals with MC, while higher household income per capita and a higher number of chronic illnesses were associated with a decreased prevalence of MI.

MI awareness was not found to be associated with sex, age, education or income in the current study, although the sensitivity analysis demonstrated a positive association with education after multiple imputation. The 10/66 study found a similar positive association between education and MI awareness in Latin Americans.¹⁹ Education is a proxy for cognitive reserve and may enhance the ability to manage cognitive demands as one ages,²⁰ indicating a cognitive reserve paradox wherein higher educational attainment may enable individuals to better compensate for early cognitive deficits, thereby delaying the subjective perception of decline despite objective impairment. In contrast, individuals with lower education may have fewer compensatory strategies which contributes to a higher impact on their everyday functioning, therefore increasing their perception of memory difficulties^21.

In our study we used disabilities in different ADLs as a proxy for functional status. The association of awareness of MI with disability in ADL, is well described in literature,¹ but differences are found in respect of each activity. We found an association between awareness of MI and disability in instrumental and advanced but not basic activities of daily living. While other studies with a Brazilian sample found an association of awareness with i-ADL and b-ADL but not with a-ADL.^22,23 This probably relates to the older sample (above 65) used in the other studies, being more prone to present a greater decline in more basic ADLs.

Regarding neuropsychiatric symptoms, we found that MI awareness was associated with an increased number of depressive symptoms. Depression is considered a factor that could lead to higher awareness of MI²⁴ despite being itself a risk factor for dementia.²⁵ This also corroborates findings from the 10/66 Dementia research group study about MI awareness, which indicated an association with depressive symptoms in Latin America.¹⁹ Our results are also consistent with another analysis of the 10/66 data which revealed that mania was associated with worse MI awareness,²⁶ and with a systematic review that found an association of impaired awareness with lower depressive symptoms¹ suggesting that there is a relationship between mood and awareness.

Furthermore, we found an association between MI awareness and a higher number of chronic illnesses but not to self-rated health in the fully adjusted model. This demonstrated that presence of illnesses is more relevant to MI awareness than the perception of health. Another point to consider is that this association was positive even when some of the self-reported illnesses, such as diabetes and hypertension, are risk factors for dementia, and might have increased prevalence in a sample with MI.²⁷

With respect to MC, we found a prevalence of 41.9% which is in line with estimates in previous populational studies, ranging from 25-50%.²⁸ A previous study by our group suggested that the slightly higher figure than in other studies, particularly from HICs was probably due to the lower educational attainment in our sample.²⁹ A study using data from the English Longitudinal Study of Aging,³⁰ an HRS sister study, found a prevalence of 30% of MI in individuals with MC against only the 11.5% in our study, but they used a z-score under −1.0 SD in a cognitive task while we used a z-score of −1.5 SD in a memory task. Similarly to the ELSA, a study using an American population sample aged above 65 years found 31% of participants with cognitive impairment.³¹ Previous Brazilian studies have not a representative sample of the Brazilian population and found conflicting results, with only one indicating an association of MC with cognitive impairment,³² whereas the others did not.^13-15,33,34

Another reason for the slightly higher prevalence of MC in our study is that we considered directed complaints elicited by specific questions, (rather than spontaneous ones) which tend to have a higher prevalence, but a lower association with cognitive decline and dementia.^35,36 The only Brazilian populational study we found that compared the prevalence of spontaneous and directed complaints had a sample of 91 individuals above the age of 50, of which only 14% had spontaneous complaints, while 76% had directed complaints.³⁵

We found a higher prevalence of women with MC, which has also been described in most populational studies,³⁷ but no difference between sex and age in the prevalence of MI in individuals with MC, which we did not expect as higher age and female sex are more usually associated with dementia.^37,38 In our previous study,²⁹ we found that while women had a higher overall prevalence of MC, men tended to present complaints only with worse memory impairment and older age. The lack of association of sex and age with cognitive decline in individuals with MC was also found in an USA population-based study with an ethnically diverse sample aged 65 years and over.³¹ This finding could indicate that age and sex influence the prevalence of MC more than cognitive decline, in which men are more likely to complain only when memory failures are occurring with more intensity, while women developing MC earlier, continue with the progress of neurodegeneration leading to unawareness of cognitive decline.⁴

Regarding education, we found a positive association between higher education and MI in individuals with MC, which could also be explained by the cognitive reserve paradox.²¹ We also found a positive association between depressive symptoms and memory impairment in MC; however, we cannot determine the direction of this association, due to the cross-sectional design of this study. Three possible explanations may account for this phenomenon. First, individuals with dementia who also present depressive symptoms may have a greater tendency to perceive difficulties in general, including those related to cognition. Second, depression itself can mimic a dementing syndrome, in which case the underlying mechanism of cognitive deficits would differ from neurodegeneration.²⁵ Third, individuals who recognize their ongoing cognitive decline may experience psychological distress as a consequence, which in turn may give rise to depressive symptoms.⁸

Regarding other health factors, a higher number of chronic illnesses was associated with a smaller prevalence of MI in individuals with MC, while self-rated health was not associated. Both are usually associated with MC^39,40 and with MI as some of the self-reported illnesses are risk factors for dementia.²⁷ This finding could be explained by survivor bias or progression to impaired awareness.

Finally, disability in activities of daily living and lower household income were associated with MI in individuals with MC, which could be explained by these factors being associated with functional impairment.⁴¹ Household income can be affected by MI individuals stopping work and also by a family member giving up paid work to become an unpaid caregiver to a more dependent individual.⁴² Interestingly, all ADL were associated in sociodemographic adjusted regressions, but in the fully adjusted regression only advanced and instrumental ADL were associated while basic ADL was not. This finding may reflect disease progression: as cognitive impairment advances and disability in basic ADLs emerges, individuals may also lose awareness of their cognitive deficits.⁴

This study has some limitations that should be mentioned. First, our analysis was based on cross-sectional data and cannot infer temporality between the associated variables and MI. Second, we had a large number of excluded participants due to missing data, although the multiple imputation analysis revealed similar patterns of association between the selected variables and MI, awareness of MI, and MC when including data from excluded individuals the possibility of selection bias cannot be entirely ruled out, particularly if those excluded had more pronounced cognitive impairment. Third, our measure of memory complaints is limited as it was answered in a single dichotomic manner. Fourth, we considered awareness as the presence of memory complaints in individuals with memory impairment, and this does not cover the complex nature of this phenomenon. Fifth, a-ADLs were measured differently from the other disabilities, which might have inflated negative response. Finally, the group classified as having objective memory impairment is likely heterogeneous, encompassing individuals at different stages of cognitive decline, including milder preclinical impairment up to more advanced clinical conditions such as dementia. Since awareness of cognitive decline can decrease as impairment progresses, this heterogeneity may have influenced the observed prevalence and correlates of awareness when memory complaints were used as a proxy.

Despite these limitations, our study has some significant strengths, including the fact that it analyzed data from a large representative sample of the Brazilian population aged 50 and over. Furthermore, we included associated factors not only sociodemographic but also health-related information and daily living activities, thus allowing for the identification of significant associations while controlling for the others.

Our findings demonstrate that half of individuals with MI are aware of their cognitive status and that one in ten individuals with MC exhibits objective memory impairment. These results highlight the necessity for actively screening cognitive decline in the middle-aged and older adult population, while addressing other causes for memory complaints. Longitudinal data from LMICs, with more refined cognitive assessment and measurement of memory complaints are needed to better disentangle awareness across different stages of cognitive impairment in diverse populations.

Supplemental Material

Supplemental material - Memory Complaints and Impairment Among Middle-Aged and Older Brazilians: ELSI-Brazil Study

Supplemental material for Memory Complaints and Impairment Among Middle-Aged and Older Brazilians: ELSI-Brazil Study by Pedro J. M. R. Pinho, MD, Matheus Ghossain Barbosa, MD, PhD, Ari Alex Ramos, PhD, Lucas Martins-Teixeira, MD, Andrew C. C. Miguel, MD, Maria Fernanda Lima-Costa, MD, PhD, Daniel C. Mograbi, PhD, Cleusa Pinheiro Ferri, MD, PhD in Journal of Geriatric Psychiatry and Neurology

Footnotes

ORCID iDs

Pedro J. M. R. Pinho

Matheus Ghossain Barbosa

Ari Alex Ramos

Cleusa Pinheiro Ferri

Funding

The authors disclosed receipt of the following financial support for the research, authorship, and/or publication of this article: This study was supported by Sponsorship—ELSI‐Brazil was supported by the Brazilian Ministry of Health: Conselho Nacional de Desenvolvimento Científico e Tecnológico; Producitivity Research Fellow; DECIT/SCTIE, Ministério da Saúde (Grants: 404965/2012-1, TED 28/2017); COPID/DECIV/SAPS (Grants: 20836, 22566, 23700, 25560, 25552, 27510). Ferri CP, Mograbi DC and Lima‐Costa MF are recipient of a productivity research fellowship from CNPQ.

Declaration of Conflicting Interests

The authors declared no potential conflicts of interest with respect to the research, authorship, and/or publication of this article.

Supplemental Material

Supplemental material for this article is available online.

References

Azocar

Livingston

Huntley

. The association between impaired awareness and depression, anxiety, and apathy in mild to moderate alzheimer’s disease: a systematic review. Front Psychiatr. 2021;12:633081. doi:10.3389/fpsyt.2021.633081.

Warren

Reid

Whitfield

Moustafa

. Subjective memory complaints as a predictor of mild cognitive impairment and Alzheimer’s disease. Discov Psychol. 2022;2(1):13. doi:10.1007/s44202-022-00031-9.

Wilson

Barnes

Rajan

, et al. Antecedents and consequences of unawareness of memory impairment in dementia. Neuropsychology. 2018;32(8):931-940. doi:10.1037/neu0000437.

, et al. Temporal course and pathologic basis of unawareness of memory loss in dementia. Neurology. 2015;85(11):984-991. doi:10.1212/wnl.0000000000001935.

Razafimahatratra

Guieysse

Lejeune

, et al.

Can a failure in the error-monitoring system explain unawareness of memory deficits in Alzheimer’s disease?

Cortex. 2023;166:428-440. doi:10.1016/j.cortex.2023.05.014.

Verhülsdonk

Quack

Höft

Lange-Asschenfeldt

Supprian

. Anosognosia and depression in patients with Alzheimer’s dementia. Arch Gerontol Geriatr. 2013;57(3):282-287. doi:10.1016/j.archger.2013.03.012.

Smid

Studart-Neto

César-Freitas

, et al. Subjective cognitive decline, mild cognitive impairment, and dementia – syndromic approach: recommendations of the scientific department of cognitive neurology and aging of the Brazilian academy of neurology. Dement neuropsychol. 2022;16(3 Suppl 1):1-24. doi:10.1590/1980-5764-DN-2022-S101EN.doi:

Mogle

Hill

Bhargava

Bell

Bhang

. Memory complaints and depressive symptoms over time: a construct-level replication analysis. BMC Geriatr. 2020;20:57. doi:10.1186/s12877-020-1451-1.

Comijs

Deeg

DJH

Dik

Twisk

JWR

Jonker

. Memory complaints; the association with psycho-affective and health problems and the role of personality characteristics. A 6-year follow-up study. J Affect Disord. 2002;72(2):157-165. doi:10.1016/s0165-0327(01)00453-0.

10.

Estimation of the global prevalence of dementia in 2019 and forecasted prevalence in 2050: an analysis for the global burden of disease study 2019. Lancet Public Health. 2022;7(2):e105-e125. doi:10.1016/S2468-2667(21)00249-8

11.

Burmester

Leathem

Merrick

. Subjective cognitive complaints and objective cognitive function in aging: a systematic review and meta-analysis of recent cross-sectional findings. Neuropsychol Rev. 2016;26(4):376-393. doi:10.1007/s11065-016-9332-2.

12.

Bertola

Suemoto

Romero Alberti

, et al. Prevalence of dementia and cognitive impairment no dementia in a large and diverse nationally representative sample: the ELSI-brazil study. J Gerontol A Biol Sci Med Sci. 2023;22:glad025. doi:10.1093/gerona/glad025. Published online January.

13.

Aguiar

ACPde O

Ribeiro

Jacinto

. Subjective memory complaints in the elderly May be related to factors other than cognitive deficit. Dement Neuropsychol. 2010;4(1):54-57. doi:10.1590/S1980-57642010DN40100009.

14.

Brucki

SMD

Nitrini

. Subjective memory impairment in a rural population with low education in the amazon rainforest: an exploratory study. IPG. 2009;21(01):164-171. doi:10.1017/S1041610208008065.

15.

Minett

TSC

Da Silva

Ortiz

Bertolucci

PHF

. Subjective memory complaints in an elderly sample: a cross-sectional study. Int J Geriatr Psychiatr. 2008;23(1):49-54. doi:10.1002/gps.1836.

16.

Lima-Costa

de Andrade

de Souza

PRB

, et al. The Brazilian longitudinal study of aging (ELSI-Brazil): Objectives and design. Am J Epidemiol. 2018;187(7):1345-1353. doi:10.1093/aje/kwx387.

17.

Sonnega

Faul

Ofstedal

Langa

Phillips

Weir

. Cohort profile: the health and retirement study (HRS). Int J Epidemiol. 2014;43(2):576-585. doi:10.1093/ije/dyu067.

18.

Kozlov

Dong

Kelley

Ankuda

. The epidemiology of depressive symptoms in the last year of life. J Am Geriatr Soc. 2020;68(2):321-328. doi:10.1111/jgs.16197.

19.

Mograbi

Ferri

Sosa

, et al. Unawareness of memory impairment in dementia: a population-based study. Int Psychogeriatr. 2012;24(6):931-939. doi:10.1017/S1041610211002730.

20.

Stern

. Cognitive reserve in ageing and Alzheimer’s disease. Lancet Neurol. 2012;11(11):1006-1012. doi:10.1016/S1474-4422(12)70191-6.

21.

Stern

Barnes

Grady

Jones

Raz

. Brain reserve, cognitive reserve, compensation, and maintenance: operationalization, validity, and mechanisms of cognitive resilience. Neurobiol Aging. 2019;83:124-129. doi:10.1016/j.neurobiolaging.2019.03.022.

22.

Hartle

Mograbi

Fichman

, et al. Predictors of functional impairment and awareness in people with dementia, mild cognitive impairment and healthy older adults from a middle-income country. Front Psychiatr. 2022;13:941808. doi:10.3389/fpsyt.2022.941808.

23.

Mograbi

Morris

Fichman

, et al. The impact of dementia, depression and awareness on activities of daily living in a sample from a middle-income country. Int J Geriatr Psychiatr. 2018;33(6):807-813. doi:10.1002/gps.4765.

24.

Mograbi

Morris

. On the relation among mood, apathy, and anosognosia in Alzheimer’s disease. J Int Neuropsychol Soc. 2014;20(1):2-7. doi:10.1017/S1355617713001276.

25.

Wilson

Capuano

Boyle

, et al. Clinical-pathologic study of depressive symptoms and cognitive decline in old age. Neurology. 2014;83(8):702-709. doi:10.1212/WNL.0000000000000715.

26.

Mograbi

Ferri

Stewart

, et al. Neuropsychological and behavioral disturbance correlates of unawareness of memory impairment in dementia: a population-based study. J Geriatr Psychiatr Neurol. 2015;28(1):3-11. doi:10.1177/0891988714541868.

27.

Livingston

Huntley

Liu

, et al. Dementia prevention, intervention, and care: 2024 report of the lancet standing commission. Lancet. 2024;404(10452):572-628. doi:10.1016/S0140-6736(24)01296-0.

28.

Jonker

Geerlings

Schmand

. Are memory complaints predictive for dementia A review of clinical and population-based studies. Int J Geriatr Psychiatr. 2000;15:983-991. doi:10.1002/1099-1166(200011)15:11<983::AID-GPS238>3.0.CO;2-5.

29.

Pinho

PJMR

Bertola

Ramos

, et al. Subjective memory complaints: prevalence, associated factors and sex differences in the ELSI-brazil study. Int J Geriatr Psychiatr. 2023;38(11):e6026. doi:10.1002/gps.6026.

30.

Brailean

Steptoe

Batty

Zaninotto

Llewellyn

. Are subjective memory complaints indicative of objective cognitive decline or depressive symptoms? Findings from the English longitudinal study of ageing. J Psychiatr Res. 2019;110:143-151. doi:10.1016/j.jpsychires.2018.12.005.

31.

Dhana

DeCarli

Dhana

, et al. Association of subjective memory complaints with white matter hyperintensities and cognitive decline among older adults in Chicago, Illinois. JAMA Netw Open. 2022;5(4):e227512. doi:10.1001/jamanetworkopen.2022.7512.

32.

Mattos

Lino

Rizo

Alfano

Araújo

Raggio

. Memory complaints and test performance in healthy elderly persons. Arq Neuropsiquiatr. 2003;61(4):920-924. doi:10.1590/S0004-282X2003000600006.

33.

Silva

LDSV e

Silva

TBLda

Falcão

DVda S

, et al. Relations between memory complaints, depressive symptoms and cognitive performance among community dwelling elderly. Arch Clin Psychiatry (São Paulo). 2014;41:67-71. doi:10.1590/0101-60830000000013.

34.

Paulo

DLV

Yassuda

. Queixas de memória de idosos e sua relação com escolaridade, desempenho cognitivo e sintomas de depressão e ansiedade. Rev psiquiatr clín. 2010;37(1):23-26. doi:10.1590/S0101-60832010000100005.

35.

Pereira

de Vasconcelos

THF

de Oliveira

AAR

, et al. Memory complaints at primary care in a middle-income country: clinical and neuropsychological characterization. Dement Neuropsychol. 2021;15(1):88-97. doi:10.1590/1980-57642021dn15-010009.

36.

Burmester

Leathem

Merrick

. Assessing subjective memory complaints: a comparison of spontaneous reports and structured questionnaire methods. Int Psychogeriatr. 2015;27(1):61-77. doi:10.1017/S1041610214001161.

37.

Tomita

Sugawara

Kaneda

, et al. Sex-specific effects of subjective memory complaints with respect to cognitive impairment or depressive symptoms. Psychiatr Clin Neurosci. 2014;68(3):176-181. doi:10.1111/pcn.12102.

38.

Abdulrahman

Richard

van Gool

Moll van Charante

van Dalen

. Sex differences in the relation between subjective memory complaints, impairments in instrumental activities of daily living, and risk of dementia. J Alzheimers Dis. 2021;15:283-294. doi:10.3233/JAD-215191. Published online November.

39.

John

Evans

Hanfelt

Loring

Goldstein

. Subjective memory complaints in white and African American participants. J Geriatr Psychiatr Neurol. 2020;33(3):135-143. doi:10.1177/0891988719868305.

40.

Pedro

Mercedes

Ramón

Borja

. Subjective memory complaints in elderly: relationship with health status, multimorbidity, medications, and use of services in a population-based study. Int Psychogeriatr. 2016;28(11):1903-1916. doi:10.1017/S104161021600106X.

41.

Ikeda

Maruta

Shimokihara

Nakamura

Han

Tabira

. Difficulties in the use of everyday technology among older adults with subjective memory complaint and cognitive decline. Gerontology. 2022;68(6):655-663. doi:10.1159/000518390.

42.

Ferretti

Sarti

Nitrini

Ferreira

Brucki

SMD

. An assessment of direct and indirect costs of dementia in Brazil. PLoS One. 2018;13(3):e0193209. doi:10.1371/journal.pone.0193209.

Supplementary Material

Please find the following supplemental material available below.

For Open Access articles published under a Creative Commons License, all supplemental material carries the same license as the article it is associated with.

For non-Open Access articles published, all supplemental material carries a non-exclusive license, and permission requests for re-use of supplemental material or any part of supplemental material shall be sent directly to the copyright owner as specified in the copyright notice associated with the article.

0.00 MB

0.11 MB